,. , . . . _ V . V . .6 3.1.5.9 _, . . «Viki. V . . . . V V . V. . V Vr......,... V . . V V, . . . .. . V . .. . _ q . , V V v I . , ,~ V . .q . V ¢ . V n . V . . .. . . V . . V . V . .,. . . . . . . , , V .V V. V . . . . V n . . VV . V. V . . , . V .V a V . y . V . . , ~ . .. .. . . . .., ., u . . v V . .V . V. . V . . V V . V V . ,. . V. . . . . . V . V . V. , . V ,. . VV V V . . V . . . . V . _ V .. .. u . V o . V . . . V . V , V V . _ . V . . I V v , V , .V V A . . V V , . . . ,, \. V ,‘1 QV ., . .. ,V , . V V V V.., . .r 9003’ LIBRARY Michigan State University This is to certify that the thesis entitled IRRIGATION METHODS FOR ABIES FRASER! (PURSH) POIR CHRISTMAS TREE PRODUCTION presented by NICHOLAS J. GOOCH has been accepted towards fulfillment of the requirements for the MS. degree in Forestry WW I Major Professor’s Signature Lama/08 Date MSU is an affinnative-action, equal-opportunity employer PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 5/08 K‘IProglecsPres/ClRC/DateDue Indd IRRIGATION METHODS FOR ABIES FRASER! (PURSH) POIR CHRISTMAS TREE PRODUCTION By Nicholas J. Gooch A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Forestry 2008 ABSTRACT IRRIGATION METHODS FOR ABIES FRASER! (PURSH) POIR CHRISTMAS . TREE PRODUCTION By Nicholas J. Gooch Irrigation may improve the growth and survival of Abies frasen’ Christmas trees produced in Michigan. However, specific guidelines for irrigation do not exist. The goal of this study was to assess soil matric potential (SMP WW) and plant stress (stern water potential SWP, crop water stress index CWSI) as tools for irrigation scheduling in A. frasen'. The specific objectives were to design, construct, and implement an automated irrigation system; determine the effect of changes in environmental conditions on SMP; and determine the effect of SMP on growth and water stress of A. frasen' at various stages of a Christmas tree rotation. Over the course of this study SMP were linearly (R2=0.94-O.98) or exponentially (R2=O.75-0.98) correlated with the adjusted evapotranspiration (AET) depending on soil type and moisture depletion. Seedling survival was positively affected by low SMP. Changes in growth were positively related with decreasing SMP, and differences were observed among treatments (P<0.05) in relative height and basal area on smaller trees (<0.9 rn tall). Stem water potential (SWP) and crop water stress index (CWSI) were also positive related with decreasing SMP although results were less consistent across irrigation treatments, compared to growth. Additional research is needed to examine the application of CWSI and SWP as tools for irrigation scheduling in A. fraseri. ACKNOWLEDGEMENTS I would like to thank my major professor, Dr. Pascal Nzokou for providing me with the knowledge, guidance, and support that I received over the past two years. This project has been a very good learning experience and has helped me expand my knowledge and thought process immensely. I am also thankful for my committee members Dr. Bert Cregg and Dr. Laurent Matuana who spent many hours reading my thesis and also for providing guidance and help during my study. I would also like to thank my mother, father, Peggy, and Tony for supporting my decision to continue my education and the support they provided for me during this whole time. I would also like to thank my sister Tiffany, grandparents, and all my other family members who always encouraged and believed in me. Most important I would like to thank my loving and caring girlfriend Andrea. Without her mental and emotional support I don't know how I would have gotten through everything. I am thankful for the help of my fellow graduate student Paligwende Nikiema and several undergraduate. I would like to thank Mr. Rex Korson and Mr. Mike Gwinn for being so helpful and accommodating with all the research taking place on their farms. Finally I would like to acknowledge the MCTA and project GREEEN for the funding and financial support of this research project. TABLE OF CONTENTS LIST OF TABLES ............................................................................. vi LIST OF FIGURES ........................................................................... viii LIST OF ABBREVIATIONS ................................................................. xi INTRODUCTION .............................................................................. 1 Literature cited ........................................................................ 6 CHAPTER ONE LITERATURE REVIEW ............................................................. 8 Biological and physical characterization of Fraser fir ....................... 9 Taxonomical Background .................................................. 9 Range & Distribution ........................................................ 9 Morphology & Physiology ................................................. 1O Economical Importance ................................................... 13 Irrigation systems in Fraser fir Christmas tree production ................. 15 Center Pivot Irrigation ...................................................... - 15 Traveling Water Gun ........................................................ 16 Drip Line Irrigation ........................................................... 17 Irrigation practices in agriculture ................................................. 19 Plant-based Assessments ................................................ 19 Canopy Temperature ............................................. 19 Stem Water Potential ............................................. 22 Physical Characteristics............................................ 24 Visual Indices ....................................................... 25 Soil-based assessments .................................................. 25 Hand-feel Method .................................................. 26 Tensiometers ....................................................... 28 Time Domain Reflectometry .................................... 32 Gravimetric Water Content ...................................... 33 Literature Cited ....................................................................... 35 CHAPTER TWO Building an automated system for Christmas tree irrigation ............... 41 Introduction ................................................................... 42 Research Locations ........................................................ 43 Setup and Design ........................................................... 44 Summary ...................................................................... 62 Literature Cited ............................................................... 68 Tables .......................................................................... 71 Figures ......................................................................... 74 CHAPTER THREE Relation between environmental factors and soil matric potential in Fraser fir (Abies frasen) production ............................................ 76 Introduction .................................................................. 77 Materials and Methods ................................................... 80 Results ........................................................................ 85 Discussion .................................................................... 90 Conclusion ................................................................... 94 Literature Cited ............................................................. 96 Tables ......................................................................... 98 Figures ........................................................................ 100 CHAPTER FOUR The effects of matric potential irrigation on the growth and water stress in Fraser fir (Abies frasen) production ................................. 106 Introduction .................................................................. 107 Materials and Methods ................................................... 109 Results ........................................................................ 114 Discussion ................................................................... 122 Conclusion ................................................................... 125 Literature Cited ............................................................. 127 Tables ......................................................................... 1 30 Figures ........................................................................ 1 34 CONCLUSION ................................................................................ 146 LIST OF TABLES TABLE PAGE 1.1 1.2 2.1 2.2 2.3 3.1 3.2 4.1 Soil moisture, appearance, and description chart for determining the percent of available water (difference between field capacity and wilting point) for various soil types, adapted from (VanderGulik, 1997)..............27 Interpretation of tensiometer readings and the need for irrigation adapted from Jamieson et al., 2002 ............................................................ 30 Lateral movement for ponded water for various soil types adapted from Boswell, 1984 ............................................................................. 71 Example of hypothetical on/off tolerances to maintain various tension levels in zones controlled by a tensiometer based automated irrigation system ...................................................................................... 72 Example of total costs of components and materials associated with building an automated irrigation system and a manually controlled irrigation system based on experiences installing automate irrigation systems in Horton and Sidney, Ml .................................................. 73 2006 and 2007 weather data summary for Horton, MI research location. Data represents average wind speed, temperature, and relative humidity and total precipitation and solar radiance ......................................... 98 2006 and 2007 weather data summary for Sidney, Ml research location. Data represents average wind speed, temperature, and relative humidity and total precipitation and solar radiance ......................................... 99 Mean relative height growth and standard error (cm/cm) for all size classes and treatments in 2006 and 2007 for Horton. Similar letters indicate no significance between treatments means (LSD 0.05) .......... 130 vi 4.2 4.3 4.4 Mean change in relative basal area and standard error (mm2/mm2) for all size classes and treatments in 2006, 2007, and total change in basal area for the 2 years at Horton. Similar letters indicate no significance between treatments means (LSD 0.05) ...................................................... 131 Significance of treatments indicated by ANOVA F-test p-values of various growth measurements for different size classes for 2006 and 2007 at Horton research location ............................................................. 132 Mean separation (LSD=0.05) for stem water potential (bar) where significance (P<0.05) was observed for small (top), medium (middle), and large (bottom) size class trees for each of the matric potential irrigation treatments. Means with similar letters are not significantly different (P<0.05) .................................................................................. 133 vii LIST OF FIGURES FIGURE PAGE 1.1 National distribution of Fraser fir (USDA, 2007) ................................. 10 1.2 Mature A. frasen' seed cone adapted from (Britton and Brown, 1913).....13 1.3 A central pivot irrigation system in Christmas tree production ............... 16 1.4 Hose drawn water gun irrigation system .......................................... 17 1.5 Design of a drip line irrigation system adapted from (FAO, 1988) .......... 18 1.6 IRT continuously measuring canopy temperatures of A. frasen' ............. 21 1.7 Diagram of pressure chamber (bomb) (Scholander et al., 1965) ............ 23 1.8 Diagram of a tensiometer, courtesy (Jamieson et al., 2002) ................. 29 1.9 Diagram of a TDR probe (CSI, 2007) .............................................. 32 2.1 Example of an irrigated field divided into zones, and the valves and pipes allowing irrigation events to be zone-specific .................................... 74 2.2 Change in matric potential in response to changes in water content affected by different soil types from (Hall et al., 1977) ......................... 75 3.1 Changes in matric potential measured at 30 cm (open circles) and 60 cm (closed circles) depths as influenced by precipitation (vertical bars), recorded at the Horton, Ml research location in 2006 (a) and 2007 (b)..100 viii 3.2 3.3 3.4 3.5 3.6 4.1 4.2 4.3 Changes in matric potential measured at 30 cm (open circles) and 60 cm (closed circles) depths as influenced by precipitation, indicated by vertical bars, recorded at the Sidney, Ml research location in 2006 (a) and 2007 (b) .......................................................................................... 101 Changes in AET affected by precipitation (A); matric potential measured at 30 cm (open circles) and 60 cm (closed circles) and cumulative AET (B); and regression analysis between AET and soil matric potential for Julian days 203 to 239 (C) for Horton, MI research location 2006 ................. 102 Changes in AET affected by precipitation (A); matric potential measured at 30 cm (open circles) and 60 cm (closed circles) and cumulative AET (B); and regression analysis between AET and soil matric potential for Julian days 155 to 177 (higher data points) and 184 to 200 (lower data points) (C) for Horton, MI research location 2007 ....................................... 103 Changes in AET affected by precipitation (A); matric potential measured at - 30 cm (open circles) and 60 cm (closed circles) and cumulative AET (B); and regression analysis between AET and soil matric potential for Julian days 216 to 234 (C) for Sidney, Ml research location 2006 ................. 104 Changes in AET affected by precipitation (A); matric potential measured at 30 cm (open circles) and 60 cm (closed circles) and cumulative AET (B); and regression analysis between AET and soil matric potential for Julian days 155 to 184 (C) for Sidney, MI research location 2007 ................. 105 Daily precipitation and soil matric potential at 30 cm depth for Horton in 2006 (A) and 2007 (B) ................................................................ 134 Daily precipitation and soil matric potential at 30 cm depth for Sidney in 2006 (A) and 2007 (B) ................................................................ 135 Mean relative height growth (cm/cm) in 2006 and 2007 for Sidney. Similar letters indicate no significance between treatments means (LSD 0.05). *' **' ***Significance P < 0.01, 0.001, or 0.0001 ....................... 136 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 Mean relative change in basal area (mmzlmmz) for 2006, 2007, and 2 year total change for Sidney. Similar letters indicate no significance between treatments means (LSD 0.05). *' **' ***Significance P < 0.01, 0.001 , or 0.0001 ........................................................................ 137 Seedling mortality for irrigation treatments for Sidney in 2007. Similar letters indicate no significance between treatments (LSD 0.05) ........... 138 Seedling mean relative height growth (cm/cm) for Sidney in 2007. Similar letters indicate no significance between treatments means (LSD 0.05) ....................................................................................... 139 Crop water stress index (CWSI) for control (A) and 15 kPa treatments (B) for Horton in 2006 ..................................................................... 140 Crop water stress index (CWSI) for control (A), 25 kPa (B), and 15 kPa treatments (C) for Horton in 2007 .................................................. 141 Pressure bomb mean stem water potential (l-PStem) measurements (bar) for small height class trees (<0.9 m), subjected to various irrigation treatments, for Horton in 2007 ...................................................... 142 Pressure bomb mean stern water potential (“’39,“) measurements (bar) for medium height class trees (0.9-1.8 m), subjected to various irrigation treatments, for Horton in 2007 ...................................................... 143 Pressure bomb mean stem water potential (IPStem) measurements (bar) for large (>1.8 m) height class trees, subjected to various irrigation treatments, for Horton in 2007 ...................................................... 144 Summary of mean pressure bomb stem water potential (“’smm) measurement (bar) for all treatments for tree height classes, small (<0.9 m), medium (0.9 — 1.8 m), and large (>1.8 m) .................................. 145 LIST OF ABBREVIATIONS ASAE ........................ American Society of Agricultural and Biological Engineers AET ................................................................. Adjusted evapotranspiration CWSI ..................................................................... Crop water stress index ET ............................................................................... Evapotranspiration ETSz ...................................... Standardized reference crop evapotranspiration ID .................................................................................... Inner diameter IRT ........................................................................... Infrared thermometer NCTA .................................................... National Christmas tree association NESDIS ............ National Environmental Satellite, Data, and Information Service OD .................................................................................. Outer diameter PET ................................................................. Potential evapotranspiration RSU .......................................................................... Remote sensing unit SDD ............................................................................. Stress degree day SMP ........................................................................... Soil matric potential SWP .......................................................................... Stem water potential TDR .................................................................. Time domain reflectometry USDA .............................................. United States Department of Agriculture WUE .......................................................................... Water use efficiency VPD ........................................................................ Vapor pressure deficit xi INTRODUCTION The Christmas tree production industry plays important economic and social roles in the United States. It is estimated that 90% of the 32 million Christmas trees sold in the US are produced in the regions of the Great Lakes, the Pacific Northwest, the Northeast, and North Carolina (Nzokou et al., 2006). Michigan accounts for approximately 15% of the national supply and has long been recognized as a major producer (Koelling et al., 1992), producing 4 million trees annually valued at more than 100 million dollars. There are approximately 800 Christmas tree growers in operation in Michigan providing, 5,000 full time jobs and 35,000 seasonal positions (Nzokou et al., 2006). Michigan offers favorable weather conditions, a wide range of topography, and a variety of soil types, allowing Christmas tree growers to produce a variety of species (Jones et al., 1997). During the last few years, the popularity of species has shifted in the market from pines, primarily Scotch (Pinus sylvestn's), to true firs, notably Fraser fir (Abies frasen') (Nzokou et al., 2006). Fraser fir has good needle retention, color, fragrance, and a selling price two to three times higher than Scotch pine prompting many growers to plant large acreages of Fraser fir. However, Fraser fir is not native to Michigan Fraser fir is native to the southern Appalachian Mountains of southwestern Virginia, western North Carolina, and eastern Tennessee (Beck, 2006). In these regions rainfall ranges from 1900 to 2500 mm annually and temperatures average 15.5° C in the peak of the growing season. A considerable amount of fog cover contributes to seasonally cool weather and precipitation totals (Mark, 1958; Smathers, 1982). In contrast, rainfall in Michigan averages less than 800 mm annually and average temperatures during the peak of the growing season often exceed 27° C (NESDIS, 2006). High summer temperatures combined with a rainfall total 1/3 that of the native range present problems in terms of survival and quality of Fraser fir. Consequently, production of Fraser fir as a Christmas tree in Michigan presents some management challenges. To solve these management issues, supplemental water must be applied to meet the physiological needs of Fraser fir and produce an acceptable crop of Christmas trees in Michigan. Fraser fir is often irrigated through the entire rotation, typically lasting 8 to 10 years, with a cost of approximately $150 0.4 ha'1 (1 acre) yr'1 (Nzokou et al., 2006). Current irrigation practices are based on a grower’s personal observations or empirical knowledge, with a rule of thumb guideline of 2.5 cm (1 in) of water applied weekly in the absence of rainfall. Instruments to assess soil moisture conditions are often not used and research data for irrigation on Fraser fir are unavailable. Improper use of irrigation can have many negative impacts on the production of Fraser fir and on the environment. Failing to provide enough water to meet the needs of Fraser fir could result in a decrease in productivity and quality, increased rotation time, and mortality. The production of Fraser fir as a Christmas tree often requires high inputs in the form of chemical fertilizers to meet quality demands and shorten rotation time (Hinesley et al., 2000). Excess irrigation has the potential to reduce the effectiveness of these inputs, and contaminate surface and ground water supplies (Hussein and Bhattarai, 2007). Also, frequent high volume applications of irrigation waters can compact the soil (Chang and Hills, 1993), reducing drainage and creating an environment favorable for pathogen development, particularly Phytophthora root rot, a serious problem for Fraser fir growing in poorly drained soils (Benson and Grand, 2000). Additionally, operating expenses of irrigation systems in the form of labor, fuel, and system maintenance add to production costs. Furthermore, water use restrictions are becoming a focal point of state and local government. The Michigan Department of Agriculture (MDA) has adopted the Generally Accepted Agricultural and Management Practices (GAAMP) for irrigation and water usage. The irrigation GAAMP’s are designed to improve stewardship for the water, soil, crop, and the agricultural sector of Michigan’s economy (MDA, 2006). Although the GAAMP are voluntary, they provide guidelines and help introduce agriculture to the possibility of water usage legislation. Current water use legislation contained in the Natural Resources Environmental Protection Act requires registration, reporting, and conservation measures for water use systems having the capacity to withdraw greater than 100,000 gallons daily (Michigan, 2006). It is likely other agencies including state and local governments will continue to amend current water usage legislation to develop stronger regulations and enforcement of water quality standards. If Michigan Christmas tree growers are to stay competitive, popular Christmas trees species like Fraser fir must be made available for sale. As water resource demands increase, the need for efficient water management practices will also increase. Improper use of irrigation water could result in serious negative impacts on the environment and the production system. In order to prepare for these upcoming challenges, scientific research is necessary to develop an appropriate irrigation framework for Fraser fir production. Irrigating based on changes in soil moisture conditions is relatively simple and easily to apply in practice (Jones, 2004). Although this assessment method would be applicable to large production systems, irrigation based on soil conditions does not necessarily relate to plant needs. Irrigating based on plant stress is likely to be more efficient in terms of water usage, providing the plant with the amount of water it actually needs (Jones, 1990). This assessment method, although difficult to implement in large production systems, can offer insight into the relation between plant and soil based irrigation methods for Fraser fir Christmas tree production. Specific Objectives The goal of this study was to assess soil moisture and plant stress based irrigation methods as tools to develop guidelines for irrigation scheduling in Fraser fir Christmas tree production. The specific objectives of this study were to: 1. Design, construct, and implement an automated irrigation system for Fraser fir plantations; 2. Determine the effect of changes in environmental conditions on soil matric potential; 3. Determine the impact of different irrigation levels on the growth of Fraser fir at various stages of the production rotation and on plant moisture stress measured by changes in canopy temperature and stem water potential. These results can then be used to develop irrigation scheduling guidelines for Michigan growers, resulting in a more efficient usage of water and a reduction in production costs, financially and environmentally. Literature cited Beck, DE. 2006. Abies fraseri (Pursh) Poir [Online] http://wwwngfsfed.us/spfo/pubs/silvics mamflNolume 1/abies/fraseri.htm (verified October 11, 2006). Benson, D.M., and LP. Grand. 2000. Incidence of Phytophthora root rot of Fraser fir in North Carolina and sensitivity of Phytophthora cinnamomi to metalaxyl. Plant Dis. 84:661-664. Chang, W.J., and DJ. Hills. 1993. Sprinkler droplet effects on infiltration. II: Laboratory study. Journal of Irrigation and Drainage Engineering 119:157-169. Hinesley, L.E., L.K. Snelling, and CR. Campbell. 2000. Nitrogen increases fresh weight and retail value of Fraser fir Christmas trees. Hortic. Sci. 35:860-862. Hussain, l., and M. Bhattarai. Comprehensive assessment of socio-economic impacts of agricultural water uses: Concepts, approaches and analytical tools [Online] http://www.iwmi.cgiarorgQropoor/files/ADB Proiect/Research Papers/Compreh ensive%2fobagemdf. (verified April 14, 2008) Jones, D.M., L.A. Leefers, and MR. Koelling. 1997. Cost and Returns in Michigan Christmas Tree Production, 1997. Research Repot. Michigan State University, East Lansing. Jones, H.G. 1990. Plant water relations and implications for irrigation scheduling. ' Acta Horticulturae 278:67-76. Jones, H.G. 2004. Irrigation scheduling: advantages and pitfalls of plant-based methods. Jounal of Experimental Botany 55:2427-2436. Koelling. M.R., J.B. Hart, and L. Leefer. 1992. Status and Potential of Michigan Agriculture-Phase ll. Michigan State University, East Lansing. Mark, AF. 1958. The ecology of the southern Appalachian grass balds. Ecological Monographs 28(4): 294-336. MDA. 2006. Generally Accepted Agricultural and Management Practices for Irrigation Water Use, Lansing. Michigan, 8.0. 2006. Natural Resources and Environmental Protection Act, In D. o. E. Quality, (ed.), Vol. 451. NESDIS. 2006. NOAA Satellite and Information service [Online]. Available by National Climatic Data Center US. Department of Commerce mpzllwww1 .ncdc.noaa.mlpt_1b/orders/787F813D-007B-77B1-AF45- ABF25930;115.P[£ (verified October 3, 2006). Nzokou, P., L.A. Leefers, and DE. Keathley. 2006. Cost and Returns in Michigan Christmas Tree Production 2006. Research Report. Michigan State University, East Lansing. Smathers, GA. 1982. Fog interception on four southern Appalachian mountain sites. Journal of the Elisha Mitchell Scientific Society 98:119-129. CHAPTER 1 LITERATURE REVIEW Biological and physical characterization of Fraser fir Taxonomical Background Fraser fir, Abies frasen' (Pursh) Poir, is a popular conifer used for landscape and Christmas tree production. The name Abies fraseri (Pursh) Poir comes from a history of several botanists’ reports and discoveries of this particular tree. In 1787 French botanist Andre Michaux and Scottish botanist John Fraser set out on a journey through the Carolinas. Fraser eventually separated from Michaux and later discovered an unknown species of fir and was later credited for his discovery (Coffey, 2007). Frederick Traugott Pursh was a botanist who emigrated to the United States from Germany. He reported on Fraser fir and many other plants collected on the Lewis and Clark expedition, the first American overland expedition to the Pacific coast, and contributed to botany with his book “A systematic Arrangement and Description of The Plants of North America” which was published in 1813 (Reveal, 1998a). Years later a French botanist Jean Louis Marie Poiret, described Fraser fir and other plant species in a series of publications known as Encyclopedia methodique botanique co-authored with French botanist Jean Baptiste Antoine Pierre Monnet de Lamarck (Reveal, 1998b). This series of botanists reporting’s of Abies frasen' gave rise to the name Abies fraseri (Pursh) Poir. The genus name Abies comes from the Latin meaning silver fir or fir trees and the taxonomic serial number is 181829 (ITIS, 1996; USDA, 2007) Range & Distribution Fraser fir is native to a small region with unique growing conditions that allow this species to thrive. It is indigenous to the high elevations of the Appalachian chain In southwestern Virginia, western North Carolina, and eastern Tennessee (Fig. 1.1) (Beck, 1990). These regions are characterized by mild summer and winter temperatures and frequent precipitation. Elevations of this range average 1524 m (5,000 ft) above sea level with annual precipitation rates between 1900 and 2540 mm (75 to 100 in) and summer temperatures averaging 15° C (59° F) (Beck, 1990). Also, a considerable number of days (65%) have fog cover, adding a cooling effect and contributing to precipitation totals due to fog drip (Mark, 1958). Figure 1.1. National distribution of Fraser fir (USDA, 2007) Morphology & Physiology Fraser fir is a relatively small perennial tree approximately 15 to 18 m (50 to 60 ft) tall and less than 30 cm (12 in) diameter at breast height (DBH) at full 10 maturity (USDA, 2007). Fraser fir is relatively short lived with a life span of approximately 150 years (Oosting and Billings, 1951). The trees have a uniform conical shape with dark green foliage and strong branches pointing slightly upward. The needles, approximately 1.3 to 2.5 cm (0.5 to 1 in) in length, are flat with a medial groove on top and the undersides have two silvery white bands indicating the presence of rows of stomata (McKinley, 2007). The bark is smooth and thin with a brown-gray appearance, and becomes slightly scaly in older trees. There are also resin filled blisters on the trunk of the tree, distinguishing it from the close relative Abies balsamea (McKinley, 2007). Fraser fir is very tolerant to shade, capable of growing under dense canopies for many years with growth rates as low as 2.5 to 5.1 cm (1 to 2 in) per year (Beck, 1990). Stands of Fraser fir tend to develop very dense canopies, particularly in plantation settings where there is little competition. Growth can later slow and canopies begin to naturally thin as trees approach pole stage (Crandall, 1958). In high intensity production systems like plantations, height growth can often exceed 30 cm (1 ft) per year. The typical rotation for Fraser fir Christmas tree production is 8 to 10 years, resulting in a harvestable tree between 2.1 and 2.7 m (7 to 9 ft) in height (Nzokou et al., 2006). Fraser fir naturally grows in regions with shallow and rocky mineral soils (Oosting and Billings, 1951). These soils tend to be very acidic with a pH range of 3.8 to 4.2 (Beck, 1990). Rooting of Fraser fir tends to be shallow but may vary depending on the depth of soil present. In deeper soils, Fraser fir can develop a deeper root system allowing it to withstand drier conditions in comparison to its 11 frequent companion species Picea rubra (Crandall, 1958). The soil depth, type, compaction, and bulk density can have a major impact on growth of the root system (Foil and Ralston, 1967; Halverson and Zisa, 1982). Fraser fir reaches reproductive maturity at approximately 15 years in its natural environment. The species is monoecious with male and female strobuli appearing on the same tree. Flowering typically takes place between May and June with the female flowers appearing in the upper portion of the crown and male flowers slightly below the female (Beck, 1990). As with most fir species, Fraser fir has a 2 year reproductive cycle. In the first year undifferentiated bud set occurs, emerging as vegetative or cone buds the following year. Reproductive bud differentiation coincides with vegetative growth both of which place high demands on available resources (Arnold et al., 1992). The amount and frequency of cone production can be influenced by sunlight, temperature, water availability (Owens et al., 2001 ), nutrition (Arnold et al., 1992; Owens et al., 2001), and tree size (Seki, 1994). Seed cones are approximately 6.4 cm (2.5 in) in length with bracts longer than cone scales (McKinley, 2007) (Fig. 1.2). Immature cones exhibit a green to somewhat reddish appearance, finally turning brown when growth is complete. The cones grow in an upright position and are resinous. 12 'i'i-f, 7x. ..\\ , \ :L l \ \‘ V , 4 N '/’ ‘\ I, /// 21533 ~ ./ \ Figure 1.2. Mature A. fraseri seed cone adapted from (Britton and Brown, 1913) Seed dispersal typically takes place during the fall when cones are fully developed. Dispersal is done primarily by wind and when complete, erect cone stalks persist leaving an unsightly appearance for Christmas tree production (McKinley, 2007). Economic Importance Christmas tree production plays an important role in the agricultural economy. In the United States alone 30 to 35 million trees are sold annually throughout all 50 states (Dungey, 2007). Approximately 90% of this production comes from the Great Lakes states, the Pacific northwest, the northeast, and North Carolina (Koelling et al., 1992). Michigan is the leading Christmas tree producer among the Great Lakes states. The annual harvest of cut trees in Michigan is approximately 4 million, with a total value of more than 100 million dollars (Nzokou et al., 2006). With a variety of landscapes, topography, soil types, and micro climates, Michigan is favorable for growing many different tree 13 species. This ability to grow many species keeps Michigan competitive and recognized as one of the top Christmas tree producers nationally. Previously, the leading Christmas tree species for planting and sale was Scotch pine (Pinus sylvestn's). Over the last decade consumer interest has begun to shift towards true firs (A. balsamea, A. concolor, and A. frasen) with Fraser fir becoming the most popular. According to a 2006 survey of Michigan Christmas tree growers, Scotch pine, Douglas fir, and Fraser fir ranked as the top threes species in terms of area per grower in prodcution (Nzokou et al., 2006). Scotch pine and Douglas fir rank highest in terms of overall acreages because many growers have large plantations of these previously popular species nearing harvestable size. As these species are harvested Fraser fir will likely be planted in their place due to the popularity of this tree. Fraser fir has a high selling price of $27.39 wholesale and $47.00 choose & cut compared to Scotch pine at $14.13 wholesale and $15.00 choose & cut (Nzokou et al., 2006). This higher selling price, however, comes with some additional costs to the grower. Since Fraser fir Is a non-native species, the physiology of the tree may not be fully adapted to the conditions to which it is subjected. Michigan, on average, receives approximately 800 mm (31.5 in) of precipitation annually with average summer temperatures of 27° C (80° F) during the growing season (NESDIS, 2006). Also the elevation of Michigan is approximately 274 m (900 ft) above sea level with a very small percentage of days experiencing fog cover, considerably less in comparison to the native range of Fraser fir. Given low precipitation rates, average summer temperatures 12° C 14 higher than in the native range, and extended periods of summer drought, producing Fraser fir in Michigan does present some management challenges. Irrigation systems in Fraser fir Christmas tree production Due to climatic differences between Michigan and Southern Appalachians, Fraser fir plantations are commonly irrigated in Michigan. Common irrigation methods are center pivot, hose drawn travelers, and micro-irrigation or drip line tubing mnning through each planted row at the base of the trees. Each irrigation system has advantages and disadvantages that need to be taken into consideration when choosing an irrigation system. Center Pivot Irrigation Center pivot irrigation systems are connected to a central point where water is fed through the system and dispensed in a circular pattern. These systems offer good uniformity of application and are less labor intensive compared to other system (Figure. 1.3). 15 Figure 1.3. A central pivot irrigation system in Christmas tree production. To ensure all parts of the field are adequately irrigated boom lengths and heights necessary to take into consideration. Another area of concern and importance is speed of travel and the amount of water applied. Since these systems travel around a central point in a circular pattern, the farthest point from the pivot will cover a greater area than points closer to the pivot. Furthermore, these systems only irrigate in a circular pattern, limiting the size of the planted field or requiring the addition of multiple systems to ensure total coverage. Traveling Water Gun Hose drawn traveling systems offer more flexibility and mobility in their range of use. This systems use a large water canon connected to a hose linking it to a pump and water source. As the water is pumped through the system the water pressure slowly drives the system in a straight line across a field (Figure. 1.4). Figure 1.4. Hose drawn water gun irrigation system. This type of irrigation system offers flexibility in location of use, speed of travel, and the amount of water applied. Problems include: low uniformity, high operating pressures, and large droplet size. Typically these systems operate at very high pressures to deliver a large volume of water over a large area. The large water droplets and application rates can damage planted crops and lead to a reduction in soil quality from crusting (Morin et al., 1981), affecting infiltration rates (Chang and Hills, 1993), and resulting in runoff and a less efficient use of water. These unidirectional systems are also very labor intensive because the labor of one or more persons is required to restart the same irrigation m or transport the system to another location. Drip Line Irrigation 17 Drip irrigation systems offer high efficiencies in water delivery to plants (OuldAhmed et al., 2007a; Shalhevet, 1991). Drip systems consist of small diameter plastic tubing running the length of each planted row, with irrigation emitters spaced along the tubing. The tubes connect to a mainline containing a valve or system of valves allowing irrigation to be controlled in various sections of the fields. These systems operate at low pressures and various application rates are achieved by using emitters with different flow rates (Fig. 1.5). .’ r \4; - ‘1 ‘ I c. '.‘J \ ‘ ‘ I Figure 1.5. Design of a drip line irrigation system adapted from (FAO, 1988). Drip line systems are not affected by windy conditions since the emitters are on the soil surface. Application rates are low resulting in greater infiltration and less runoff. These systems can be equipped with manual or electronic valves to allow system control and automation. Although efficient in water use, drip irrigated systems are prone to salt accumulation potentially affecting crop yields (Katerji et al., 2003; Shani and Dudley, 2001). Poor quality water sources containing chemical and/or biological pollutants can affect the discharge of the system 18 emitters, decreasing uniformity of application rates (OuldAhmed et al., 2007b). For this reason it is often necessary to include a filter or system of filters at the water source; these filters require periodic cleaning or replacement. System design and maintenance can be another concern if large equipment or machines are used, increasing the likelihood of damage to the drip line tubing. Compared to other irrigation systems, when designed properly, drip irrigation is a means of increasing water use efficiency (Thorbum et al., 2003). Irrigation practices in agriculture Plant-based Assessments Plant-based irrigation is based on measured or visual indices of plant water stress. This can be done by a number of direct and indirect measurements (canopy temperature, stem water potential, plant physical characteristics, visual indices) which can be used to develop irrigation scheduling and automation. Canopy Temperature The temperature of a plant’s canopy can provide a good indication of the water stress a plant is under, based on the principle that in non-water limited environments, transpiration through the stomata has a cooling effect on the plant. As water becomes limiting the stomata begin to close, reducing the rate of transpiration. If the leaf surface continues to absorb the same amount of solar energy, the surface temperature will increase and can often exceed ambient temperature. This difference in temperature can be substantial. Ehrler and 19 Bavel (1967) found that the differences between the leaf surface and air temperature (AT) ranged from -1.5° C on wet soils to 45° C on water-depleted soils. Similarly, AT decreased by 1° C for two days following an irrigation event in cotton (Ehrler, 1973). These AT values can be influenced by solar energy, but more importantly the vapor pressure deficit (VPD) can also have a direct effect (Ehrler et al., 1978). This was further tested in several microenvironments with alfalfa and it was found that AT and vapor pressure showed a linear relationship, normalizing AT for environmental variability (Jackson et al., 1981 ). Infrared thermometry (IRT) offers a good understanding of the change in temperature of a plants leaf surface and canopy. This technology works on the principle that all objects absorb and radiate energy. The temperature can be measured in the infrared range (750 nm to 1000 nm) by knowing the energy emitted and the emissivity of an object. IRT is advantageous over other techniques in the sense that it can be mounted above crops to make measurements on a single crop or over an entire field (Figure. 1.6). 20 Figure 1.6. IRT continuously measuring canopy temperatures of A. frasen'. Jackson et al. (1977) developed a calculation of the summation of the AT measured daily over a period of time, termed stress degree day (SDD). This SDD value was used in conjunction with predicted evapotranspiration rates to determine the time and amount of irrigation that needed to be applied (Jackson et al., 1977). ldso et al. (1981) used the relationship between AT and VPD to develop an empirical calculation of a crop water stress index (CWSI) by determining non-water stressed and non-transpiring baseline. This calculation is a simple equation with three variables: measured canopy temperature (Tc) at a calculated VPD, a point on the non-transpiring baseline (TMAx) calculated at the same VPD as Tc, a point on the non-water stressed baseline (TMIN) calculated at the same VPD as Tc. However, this method of reporting the CWSI has been in question since it does not account for changes in temperature related to solar radiance and changes in wind speed. Jackson et al. (1981) developed a theoretical CWSI equation that takes these two factors into consideration, 21 requiring a slightly more difficult calculation than the empirical equation. Many researchers have used the empirical CWSI method to assess water stress and schedule irrigation events on a variety of crops including com (lrmak et al., 2000; Yazar et al., 1999), sunflower (Erdem et al., 2006), watermelon (Orta et al., 2003), and grass and forage crops (Al-Faraj et al., 2001; Payero et al., 2005). Stem Water Potential For more than 100 years, the cohesion tension theory has been used to explain how water moves to the top of a tall tree (Dixon and Joly, 1894). This theory states that, water moving through transpiring stomata creates a negative pressure (tension) on the xylem water column. Due to the strong cohesive forces of water, this column remains continuous throughout the tree, thus pulling water from the soil through the roots. This theory was validated by Scholander et al. (1965) with the design and experimentation of a pressure chamber. This Scholander pressure chamber, more commonly known as pressure bomb, uses branch cuttings from a tree to measure the xylem tensile state. The cutting is placed in the chamber with the cut end protruding from a small orifice, with the remainder of the cutting sealed into an air tight chamber. Pressure is gradually applied with nitrogen gas until sap is noticed at the cut end (Fig. 1.7). 22 Figure 1.7. Diagram of pressure chamber (bomb) (Scholander et al., 1965). The pressure causing sap flow is known as the balancing pressure, pressure necessary to reach equilibrium. It equals the amount of vascular tension the branch was experiencing prior to removal from the tree. This observed measure can be used to calculate stem water potential by the following equation: WW = P + W3 Where WW is the water potential, P is the pressure applied by the chamber, and W3 represents the osmotic effects of the solutes in the xylem sap of the plant (Boyer, 1967). Care must be taken in preparing cuttings for the pressure chamber to get accurate results. Pressure bomb measurements could overestimate pressure by 0.5 MPa if the leaf it left uncovered during transport (Turner and Long, 1980). A concern in rapidly transpiring leaves is when tension is released, following removal, is the xylem equilibrates with the mesophyll cells, affecting measurements (Passioura, 1991). Accumulations of solutes in the mesophyll cells can contribute to the osmotic potential of the water gradient. This difference in water potential between the xylem and mesophyll cells would 23 be more apparent on transpiring plants than non-transpiring plants. This phenomenon was documented by comparing pressure chamber and pressure probe data obtained from transpiring and non-transpiring sugar cane and maize plants (Melcher et al., 1998). In non-transpiring plants, suitable results were obtained by both methods. Wei et al. (2001) discussed proper techniques to obtain xylem pressure measurements with the pressure probe, which were found to agree with measurements made by the pressure chamber, validating its use. Further validating its use, Cochard et al. (2001) found that changes in water content in stems were correlated with changes in water potential measured by the pressure chamber. Although useful for making quick in situ measurements, incorporating the pressure chamber into an automated irrigation system is impractical (Jones, 2004). Physical Characteristics Beyond the physiological measures of plant water stress, changes in growth also have some merit for assessing water needs. As water becomes limited plants attempt to conserve water by holding water more strongly in some parts than others. If water becomes limited that the plant can no longer store it, cell turgor can be lost causing cells to shrink resulting in a reduction of the size and thickness of fruit, leaves, or stems. In the trunks and stems of trees, the cambium layer is particularly sensitive to moisture stress. Changes in diameter or caliper could serve as a predictor of water stress in trees. These changes have been useful for assessing water stress in almond (Fereres and Goldhamer, 24 2003; Nortes et al., 2005) and olive trees (Moriana and Fereres, 2002). Although this method would be difficult to incorporate into an automation system due to daily changes in diameter, Fereres and Goldhamer (2003) showed that maximum daily shrinkage was a suitable approach for scheduling irrigation events. A potential drawback of this method is accuracy of measurement and the expensive equipment required to achieve it. Visual Indices Visually assessing the level of moisture stress in a plant can be difficult and a poor indication of the initiation of water stress. This method is not precise and reduction in growth or yield typically happens before wilt is noticed visually (Jones, 2004). If visual indicators do not become evident until soil moisture conditions have reached permanent wilting point, damage may be substantial and the chances of plant recovery are poor. Soil-based Assessments The majority of the water plants need for growth and survival comes from the soil. Some of the factors that need to be taken into consideration are the rooting depth of the crop and the soil type. Water holding capacity of soils can range from 21% or more in peat and muck soils to less than 6% in coarse sands. The frequency and quantity of irrigation will have to be adjusted according to soil type. There are several methods of measuring soil moisture including the hand feel method, tensiometers, time domain refiectometry, and gravimetric water 25 content. Although these methods do not directly relate to the amount of stress a plant is experiencing, they are valuable for understanding how irrigation events influence the soil moisture state. Hand-feel Method Assessing the moisture conditions of a particular soil can be done quickly by hand. This method involves taking a sample of soil by hand or probe and using a set of guidelines to determine the soil moisture content. This procedure requires practice and experience to ensure accuracy. Soil texture must first be determined, followed by observations of how the soil behaves when balled and then formed into a ribbon in the hand (Bolen, 1984). Based on the amount of water that can be squeezed from the soil, or the soil particle’s ability to adhere to each other, inferences can be made on the moisture status using charts that describe the properties of the soil type (table 1.1). 26 Table 1.1. Soil moisture, appearance, and description chart for determining the percent of available water (difference between field capacity and wilting point) for various soil types, adapted from (VanderGulik, 1997). field capacity (field capactiy) when is bounced. no wet not form a with not form through squeezing, free water wet left on Makes ribbon. weak ball, easily. not slick. pressure but holds dry, will form ball. loose, flows fingers. squeezing, water outline left on Ribbons 2.50m. ball, very pliable, readily if clay. somewhat can slicks pressure. slighty easily squeezing, free water wet on hand. about 5 between slick fibbons thumb forefinger. pliable, ball under cracked, crumbs on Although this method is subjective, proficiency can be obtained by repeated comparisons of soils with known gravimetric water content to develop baselines for the hand feel method (Schneekloth et al., 2007). When using this technique 27 to assess soil moisture, it is important to know the soil texture present. In a field with multiple different soil types or horizons, multiple samples might have to be taken to ensure any level of accuracy. Hand feel soil moisture observations should not be made immediately following an irrigation or precipitation event as this could result in an overestimation of soil moisture content. Measurements should be delayed until the water has infiltrated the soil and wet the profile which will vary based on texture. Tensiometers A more objective measure of soil moisture content is direct measurement using instruments. Tensiometers are used to measure matric potential (WW), or the force that plant roots must overcome to extract water from the soil (Jamieson et al., 2002). Water is a polar molecule exhibiting strong attractive forces to other water molecules (cohesion) and attractive forces to particles (adhesion). As soil moisture decreases, the adhesive forces between soil and water molecules become stronger, making it more difficult to extract the water. Tensiometers measure this change in WW through a series of components. Tensiometers consist of a water—filled column with a porous tip at one end and a sealed reservoir at the other (Fig 1.8). 28 Reservoir Ceramic f— Tip Figure 1.8. Diagram of a tensiometer, adapted from (Jamieson et al., 2002). The porous tip, typically ceramic, acts as an interface between the water in the soil and the reservoir allowing the free passage of water and the exclusion of gases (Young and Sisson, 2002). The reservoir has a removable cap for servicing and filling the water column as necessary, and a gauge or pressure sensitive apparatus is attached slightly below this point. Bourdon gauges are commonly used. They consist of a flexible tube connected to a diaphragm that distorts based on the tension it is placed under, and this amount of distortion produces a reading on the gauge face (Young and Sisson, 2002). Gauge readings typically range from 0 to 93 centi-bars (cb), or greater, where 1 ob = 1 kPa. Gauges are accurate to :1 kPa, restricting this type of measurement to low-cost operations where detailed precision and accuracy are of lesser importance (Young and Sisson, 2002). For more detailed measurements, tensiometers can be equipped with an electronic pressure transducer that can be wired to data logging equipment for monitoring and controlling irrigation systems. 29 How these readings are interpreted and used can vary in individual situations. Typically, a tensiometer reading of 10 kPa indicates a soil at field capacity. Lower readings indicate a soil reaching saturation, with a reading of 0 kPa indicating a flooded system. Depending on the crop type, age, quality desired, and soil type, irrigation is recommended at tensions of 20 to 60 kPa (table 1.2) (Jamieson et al., 2002). Table 1.2. Interpretation of tensiometer readings and the need for irrigation adapted from Jamieson et al., 2002. Readings Interpretation O to 5 kPa Soil is nearly saturated. This will occur following a rainfall event or in waterlogged areas. Anaerobic conditions can develop. 5 to 20 kPa Field capacity range, the Optimal range for plants. Irrigating beyond this point can result in wasted water and leaching. 20 to 60 kPa Depending on soil type, plant age, rooting depth, and other stresses this is the point to turn on the irrigation system. Large trees with deep root system likely will tolerate higher tension levels compared to seedling with an immature root system. The amount of irrigation applied at this point should match the water holding capacity of the soil to avoid leaching. Higher than 60 kPa Readings greater than 70 kPa, for most plants, will result in water stress. Soil tension at this level can cause damage and mortality if it remains elevated for to long. Tension readings greater than 60 kPa begin approaching permanent wilting point, or the point at which plants remain wilted unless the soil is rewetted (Kozlowski and Pallardy, 1997b). Dry soils with tensions in excess of 80 kPa 3O usually cause the water contact between the tip and soil to break, resulting in air entering the instrument and a sharp decrease in reading (Jamieson et al., 2002), indicating the upper limit of a tensiometer as a measurement tool. Tensiometers can be an important tool for high input agriculture, in which fertilizers and pesticides are applied. Oki et al. (1996) found that using an automated irrigation system controlled by three different soil tension thresholds resulted in more efficient water use, reductions in pollution run-off, and increased growth compared to a grower-controlled system. The reduction in water use decreases the overall cost of operating the irrigation system. Furthermore, automating the irrigation system based on tension resulted in more efficient water than in a manual irrigation system (Oki et al., 1996). This application would be particularly useful for the production of Fraser fir since large acreages are often irrigated, taking several days or more to complete. The reduction in labor required for managing these large scale systems and assess soil moisture status could prove substantial in a tensiometer-monitored and-controlled system. Munoz-Carpena et al. (2005) found that maintaining soil tension levels of 15 kPa resulted in a 73% reduction in water use and no adverse effect on quality compared to a manually irrigated system. This further supports what Smajstrla and Locascio (1996) found when maintaining irrigation levels at 10 kPa: total marketable yield decreased linearly from tension levels of 10 kPa to 20 kPa (Smajstrla and Locascio, 1996). These findings confirm the benefit of high frequency low volume irrigation to keep soil at field capacity at all times in order to avoid water stress due to environmentally caused fluctuations in soil moisture 31 content. Also, keeping soils moistened reduces the potential development of a hydrophobic (water repelling) soil surface, resulting in a reduction in runoff if large precipitation events were to occur. Time Domain Reflectometry Another method of determining the soil water content is through time domain reflectometry (TDR). TDR was introduced in 1980 as a non-destructive method to determine the soil water content (T opp et al., 1980). TDR works by sending pulses from one post on the probe to another, with the time it takes the pulse to travel the distance between the probes as an indicator of water content. Typically, a TDR probe consists of two or more metal posts inserted into the ground and connected to the instrument body, which passes the current through the probes. The instrument is usually wired to a controller or data logging device (Fig. 1.9). Figure 1.9. Diagram of a TDR probe (CSI, 2007). Soil water content will affect the velocity of the pulse emitted from the TDR, with lower velocities representing a wetter soil (Evett, 2003). These measurements tend to be very accurate and consistent, although in certain situations TDR probes do not perform well. Accuracy is reduced in soils with high organic matter 32 due to its low permittivity (Evett, 2003). In this situation, custom calibration coefficients may be necessary to ensure a high degree of accuracy. Field calibration is often necessary to ensure accuracy since different manufacturers use different calibration techniques which can lead to measurements over or under-predicting the actual water content when field tested (Plauborg et al., 2005). In drip-irrigated applications with frequent irrigation events, or where a fertigation system is used, the accumulation of soluble salts can lead to increased EC, which can affect the accuracy of TDR probe measurements and requires calibration adjustments (Plauborg et al., 2005). TDR probes require a source of electricity and connection to a measuring device like a data logger or computer for interpretation of results. In addition, TDR systems tend to be very expensive ($4000 to $7000), due to the need of specialized equipment for system and data control, limiting their application to research (VanderGulik, 1997). Given the cost and the nature of this measurement method, permanent application of multiple TDR probes throughout a field is rarely feasible. Gravimetric Water Content A very basic and accurate method to assess soil moisture status is by calculating the gravimetric water content. Gravimetric water content is defined as the mass of the water per unit of mass dry soil. A hollow corer or pipe is inserted into the soil to remove a somewhat undisturbed column of soil. The initial mass is recorded (Mwet) and then the soil is dried at 60° C for 24 hours. The mass of 33 the dry soil (Mdry) is recorded and gravimetric water content is calculated by the equafion: = Mwater = Mwet “dry 99 Msoil Mdry (Bilskie, 2001) Limitations of this method include the inability to measure moisture status continuously using logging equipment or computer control, variability in measurements among soil types, and the labor intensive nature of this process; it is limited to situations where instant assessments of soil moisture status are not necessary. 34 Literature Cited AI-Faraj, A., GE. Meyer, and CL. Horst. 2001. A crop water stress index for tall fescue (Festuca arundinacea Schreb.) irrigation decision-making -- a fuzzy logic method. Computers and Electronics in Agriculture 32:69-84. Arnold, R.J., J.B. Jett, and H.L. Allen. 1992. Identification of nutritional influences on cone production in Fraser fir. Soil Science Society of America Journal 56:586- 591. Beck, D.E. 1990. Silvics of North America: Abies fraseri (Pursh) Poir. [Online] h_t__tp://www.na.fs.fed.th/spfo/pgbs/silvics mangalNolgme 1/abies/fraseri.htm (verified August 2, 2007). Bilskie, J. 2001. Soil water status: content and potential. Campbell Scientific, Inc. App. Note: ZS-I. Bolen, KR. 1984. Estimating soil moisture by appearance and feel. University of Nebraska Cooperative Extension ReportzGB4-690-A. Boyer, J.S. 1967. Leaf water potentials measured with a pressure chamber. Plant Physiology 42:133-137. Britton, ML, and A. Brown. 1913. USDA-NRCS PLANTS Database - Illustrated flora of the northem states and Canada. Vol.1: 63. Chang, W.J., and DJ. Hills. 1993. Sprinkler droplet effects on infiltration. II: Laboratory study. Journal of Irrigation and Drainage Engineering 119:157-169. Coffey, R.K. 2007. Magnolias and Firs: The John Fraser Connection [Online] mtg://wvm~.appvoices.org/index.th?/site/voice stories/magnolias gid firs the i ohn fraser connection/issue/543 (verified 9/25/07). Crandall, UL 1958. Ground vegetation patterns of the spruce-fir area of the Great Smoky Mountains National Park. Ecological Monographs 28(4):337-360. CSI. 2007. Campbell Scientific Inc. (CSI) Image of: CS605-L 3-rod TDR Probe [Online] hfip://www.camgbellsci.com/images/cs6053ij (verified 10/1/07). Dixon, H.H., and J. Joly. 1894. On the ascent of sap. Annals of Botany 8:468- 470. Dungey, R. 2007. National Christmas Tree Association: Quick tree facts [Online] mpzllwww.christmastree.org/facts.cfm (verified August 3, 2007). 35 Ehrler, W.L. 1973. Cotton leaf temperatures as related to soil water depletion and meteorological factors. Agronomy Journal 65:404-409. Ehrler, W.L., S.B. ldso, R.D. Jackson, and R.J. Reginato. 1978. Diurnal changes in plant water potential and canopy temperature of wheat as affected by drought. Agronomy Journal 70:999-1004. Erdem, T., Y. Erdem, A.H. Orta, and H. Okursoy. 2006. Use of a crop water stress index for scheduling the irrigation of sunflower (Helianthus annuus L.). Turk J Agric For 30:11-20. Evett, SR. 2003. Soil Water Measurement by Time Domain Relfectometry, p. 894-898, In B. A. Stewart and T. A. Howell, eds. Encyclopedia of Water Science. Marcel Dekker, New York. FAO. 1988. Food and Agriculture Organization of the United Nations. Irrigation water management, Training manuals - 5: Chapter 6. Drip irrigation [Online] http://www.fao.org/docrep/88684gls8684e1ggfl (verified September 12, 2007). Fereres, E., and DA. Goldhamer. 2003. Suitability of stem diameter variations and water potential as indicators for irrigation scheduling of almost trees. Journal of Horticultural Science and Biotechnology 78:139-144. Foil, RR, and CW. Ralston. 1967. The establishment and growth of loblolly pine seedlings on compacted soils. Soil Science Society of America Proceedings 31:565-568. Halverson, H.G., and RP. Zisa. 1982. Measuring the response of conifer seedlings to soil compaction stress. USDA Forest Service, Northeastern Forest Experiment Station Research PapeerE-509. lrmak, 8., DZ. Haman, and R. Bastug. 2000. Determination of crop water stress index for irrigation timing and yield estimation of corn. Agronomy Journal 92:1221-1227. ITIS. 1996. Integrated Taxonomic Information System Report Abies fraseri (Pursh) Poir. [Online] http://www.itis.gov/servlet/SingleRpt/SinqleRbt?search topic=TSN&search value =181829 (verified August 2, 2007). Jackson, RD, R.J. Reginato, and SB. ldso. 1977. Wheat canopy temperature: A practical tool for evaluating water requirements. Water Resources Research 13:651-656. 36 Jackson, RD, S.B. ldso, R.J. Reginato, and P.J. Pinter. 1981. Canopy temperature as a crop water stress indicator. Water Resources Research 17:1133-1138. Jamieson, T., R. Gordon, L. Cochrane, A. Madani, and G. Patterson. 2002. Tensiometers and their use in irrigation scheduling. Nova Scotia Agricultural College. Water conservation factsheet No. 577.100-2 Jones, H.G. 2004. Irrigation scheduling: advantages and pitfalls of plant-based methods. Journal of Experimental Botany 55:2427-2436. Katerji, N., J.W. van Hoorn, A. Hamdy, and M. Mastrorilli. 2003. Salinity effect on crop development and yield analysis of salt tolerance according to several classification methods. Agricultural Water Management 62:37-66. Koelling. M.R., J.B. Hart, and L. Leefer. 1992. Status and potential of Michigan agricultural phase II. Christmas tree production. MAES special report number 61 :Michigan State University, East Lansing, Michigan. Kozlowski, T.T., and 8.6. Pallardy. 1997b. "Physiology of Woody Plants." Academic Press, San Diego. Mark, AF 1958. The ecology of the southern Appalachian grass balds. Ecological Monographs 28(4)294-336. McKinley, CR. 2007. National Christmas Tree Association Tree Index: Fraser fir [Online] h_ttp://www.christmastree.orgjtrees/frasercfm (verified August 2, 2007). Melcher, P.J., F.C. Meinzer, D.E. Yount, G. Goldstein, and U. Zimmermann. 1998. Comparative measurements of xylem pressure in transpiring and non- transpiring leaves by means of the pressure chamber and the xylem pressure probe. Journal of Experimental Botany 49:1757-1760. Moriana, A., and E. Fereres. 2002. Plant indicators for scheduling irrigation on young olive trees. Irrigation Science 21:83-90 Morin, J., Y. Benyamini, and A. Michaeli. 1981. The effect of raindrop impact on the dynamics of soil surface crusting and water movement in the profile. Journal of Hydrology 52:321-335. NESDIS. 2006. NOAA Satellite and lnforrnation service [Online]. Available by National Climatic Data Center US. Department of Commerce [Online] h_ttg://www1 .ncgc.noaa.gov/pt_1b/orders/787F813D-0tfl-77B1-AF45- ABF2593021 15.PQE (verified 10/03/06). 37 Nortes, PA, A. Perez-Pastor, G. Egea, W. Conejero, and R. Domingo. 2005. Comparison of changes in stem diameter and water potential values for detecting water stress in young almond trees. Agricultural Water Management 77:296-307. Nzokou, P., L.A. Leefers, and DE. Keathley. 2006. Costs and returns in Michigan Christmas tree production. MAES special report. Michigan State University, East Lansing, Michigan. Oki, L.R., J.H. Lieth, and S. Tjosvold. 1996. Tensiometer-based irrigation of cut- flower roses: Report to the California cut-flower commission. University of Callfomia, Davis. Oosting, H.J., and WD. Billings. 1951. A comparison of virgin spruce fir forest in the northern and southern Appalachian system. Ecology 32(1):84-103. Orta, A.h., Y. Erdem, and T. erdem. 2003. Crop water stress index for watermelon. Scientia Horticulturae 98:121-130. OuldAhmed, B.A., T. Yamamoto, H. Fujiyama, and K. Miyamoto. 2007b. Assessment of emitter discharge in microirrigation system as affected by polluted water. Irrigation Drainage Systems 21297-107. OuldAhmed, B.A., T. Yamamoto, V. Rasiah, M. lnoue, and H. Anyoji. 2007a. The impact of saline water irrigation management options in a dune sand on available soil water and its salinity. Agricultural Water Management 88:63-72. Owens, J.N., L.M. Chandler, J.S. Bennett, and T.J. Crowder. 2001. Cone enhancement in Abies amabilis using 6A4”, fertilizer, girdling and tenting. Forest Ecology and Management 154:227-236. Passioura, J.B. 1991. An impasse in plant water relations? Botanica Acta 104:405-411. Payero, J.O., C.M.U. Neale, and J.L. Wright. 2005. Non-water-stressed baseline for calculating crop water stress index (CWSI) for alfalfa and tall fescue grass. Transactions of the ASAE 48:653-661. Plauborg, F., B.V. lversen, and PE. Laerke. 2005. In situ comparison of three dielectric soil moisture sensors in drip irrigation sandy soils. Soil Science Society of America Journal 4:1037-1047. Reveal, J.L. 1998a. Discovering Lewis & Clark: Frederick Traugott Pursh (1774- 1820) [Online] h_ttp://www.lewis-clark.org_/content/content- a_rticle.asp?Articlell_D=50_2_ (verified 9/25/07). 38 Reveal, J.L. 1998b. Discovering Lewis & Clark: Menzies, Lambert and Poiret [Online] http://www.lewis-clark.org/content/content-artjcle.asp?ArticlelD=1 509 (verified 9/27/07). Schneekloth, J., T. Bauder, l. Broner, and R. Waskom. 2007. Colorado State University Extension: Measurement of Soil Moisture [Online] mpzllwwwextcolostateedu/drought/soilmoist.html (verified 9/21/07). Scholander, P.F., H.T. Hammel, E.D. Bradstreet, and EA. Hemmingsen. 1965. Sap pressure in vascular plants. Science 148:339-346. Seki, T. 1994. Dependency of cone production on tree dimensions in Abies man'esii. Canadian Journal of Botany 72:1713-1719. Shalhevet, J. 1991. Proceedings of the Binational China-Israel Workshop, Beijing, China, p. 17-53, In J. Shalhevet, Cangmimg, L., Yuexian, X., ed. Using water of marginal quality for crop production: major issues, water use efficiency in agriculture Peril Publishing. Rehovot, Israel. Shani, U., and L.M. Dudley. 2001. Field studies of crop responses to water and salt stress. Soil Science Society of America Journal 65:1522-1528. Smajstrla, AG, and SJ. Locascio. 1996. Tensiometer-controlled, drip-irrigation scheduling of tomato. Applied Engineering in Agriculture 12(3):315-319. Thorburn, P.J., F.J. Cook, and KL. Bristow. 2003. Soil-dependent wetting from trickle emitters: implications for system design and management. Irrigation Science 22:121-127. Topp, G.C., J.L. Davis, and AP. Annan. 1980. Electromagnetic determination of soil water content: Measurements in coaxial transmission lines. Water Resources Research 16:574-582. Turner, NC, and M.J. Long. 1980. Errors arising from rapid water loss in the measurements of leaf water potential by the pressure chamber technique. Aust. J. Plant Physiol 7:527-537. USDA. 2007. USDA-NRCS PLANTS Database Abies frasen' (Pursh) Poir. [Online] http://plants.usda.gov/iavalbrofile?svmbol=A_BFR (verified August 2, 2007) VanderGulik, T. 1997. Water conservation fact sheet: Irrigation scheduling techniques. British Columbia Ministry of Agriculture and Food Report:577.100-1. Yazar, A., T.A. Howell, D.A. Dusek, and KS. Copeland. 1999. Evaluation of crop water stress index for LEPA irrigated corn. Irrigation Science 18:171-180. 39 Young, M.H., and J.B. Sisson. 2002. Tensiometry, p. 575-609, In J. Dane and C. Topp. eds. Methods of soil analysis, Part 4, SSSA Book Series: 5. American Society of Agronomy, Madison, WI. pp. 575-609. 40 CHAPTER 2 BUILDING AN AUTOMATED DRIP SYSTEM FOR CHRISTMAS TREE IRRIGATION 41 Introduction Irrigation for Christmas tree production in Michigan is a relatively new idea, due to the increasing interest in Fraser fir (Abies frasen) production. Christmas tree production systems are unique in comparison to other agricultural crops. Christmas trees are perennial crops grown from seed in a nursery for 3 to 5 years then moved into a plantation where they are grown for an additional 6 to 9 years until they reach a mature harvestable size, approximately 2.1 m (Chastagner and Benson, 2000). Trees are planted in rows with a spacing approximately 1.8 m between trees, and 1.8 m between rows. During the entire production cycle, plantations are managed for weeds, pests, and fertility along with annual pruning and shearing to produce high quality trees. These management factors need to be considered when deciding to irrigate. Additionally, the frequency and intensity of these cultural and maintenance practices is often labor intensive, requiring the use of heavy equipment and machinery all of which can damage an irrigation system if proper care is not taken. Fraser fir, in particular, tends to require more maintenance and inputs, compared to other species, to produce a quality crop (Leuty, 2005). For this reason, it is necessary to investigate all aspects of design, set-up, integration, and problems associated with building an automated drip irrigation system for Fraser fir Christmas tree production. lnforrnation gained from this system will better help growers understand and make informed decisions on how to implement an irrigation system of this nature into their production system and some of the concerns associated with doing so. A system of this nature should 42 allow efficient use of irrigation waters resulting in a more sustainable system meeting the needs of current and future water use legislation. This paper is an overview looking into all aspects of building an automated irrigation system for Christmas tree production. The processes and suggestions are based on irrigation literature and lessons learned while building automated irrigation systems at Christmas tree farms at two research sites in Michigan. Information provided is a starting point for building a system and product names, manufacturers, and brands do not represent an endorsement but are mentioned for educational purposes. Research Locations This study was conducted in cooperation with two large Christmas tree plantations in Michigan starting in 2006. The first location, Horton Michigan, had a micro, or drip line irrigation system, that was controlled through manual on/off valves. The existing irrigation system was excavated and modified to divide the field into smaller zones allowing independent control of irrigation for each zone. Tensiometers (model R-RSU lrrometer Co. Riverside, CA) and a data logging station were also installed. Sidney, Michigan was the second location chosen for this study. The type of irrigation used at this farm was an overhead water cannon system. Although this system allows for the irrigation of large areas, control and uniformity is quite limited. For this reason, the existing Irrigation system in 2007 was discontinued 43 and a drip line irrigation system, similar to Horton, was designed and installed allowing better monitoring and control of irrigation events and quantities. Setup and Design Planning and zoning One of the most important aspects of designing a successful irrigation system is planning. The initial planning stage involves taking into consideration all aspects of production and how these factors might affect or interact with the irrigation system. The first step in planning is to designate irrigation zones. Zoning essentially involves the division of large areas into smaller, easy to manage plots. By dividing a field into zones, micro managing the amount and frequency of water being applied to each zone is possible. This is accomplished through a pipe connected to a pump or water source, leading to a manifold, housing a system of valves, which then diverts the water flow to a series of mainlines in the field (Figure 2.1). A system of sub-mainlines connects to the mainline allowing the water flow to be directed laterally into the desired zone where the sub-mainline is then connected to a series of drip lines irrigating the field. The number of zones necessary to efficiently irrigate all parts of the plantation will vary based on specific field conditions. Since irrigation is constant throughout a zone, limiting the number of zones may results in poor water use efficiency due to differences in soil types and plant water requirements. Rationale for delineation of irrigation zones may include: crop water needs, soil type, landscape topography, drainage, and weed pressure. Incorporating too 44 many zones into an area, in an attempt to micro-manage and achieve the greatest water use efficiency, could result in increased costs and maintenance associated with additional piping, labor, and repairs. It is important to plan and determining the conditions and factors present that might have an effect on the layout and design of a drip irrigation system. An optimum zone size is one that uniformly and efficiently delivers water over a given area based on the irrigation needs for the given crop. Additionally, the size of each zone will be determined by pumping capacity, elevation changes, pipe size and friction loss, application rate, and emitter output. Pump size and output Regardless of the size and type of irrigation system present, energy is required to transport water from a source to the field. This can be done using a variety of different pumps, typically powered by electricity or a fuel source. Due to the rural remote locations of many plantations, routing the necessary cabling from utilities, to provide power for an electric pump, could be quite extensive and cost prohibitive. For this reason, pumps powered by diesel engines are often preferred. Diesel powered pumps offer greater flexibility and mobility in comparison to electric. In our situation, water was delivered to the irrigation system by the water pump for the growers house. Irrigation was possible using a small pump by irrigating a smaller area for a longer period of time and rotating through the irrigation zones. The pump available at the Sidney location was a larger more powerful pump capable of delivering large quantities of water to a 45 greater area. It would be advisable to select the most powerful pump, financially feasible, to reduce the time needed to irrigate and reduce the workload a smaller pump would be subjected to. To determine the water power required to irrigate a given area, a number of factors need to be taken into consideration and can be calculated by the following equation: Q*Hp 102*E WP= (1) Where: WP = water power (kW) Q = system flow rate (L/s) Hp: total head (m) E = pump efficiency in decimal form (%) 102 = Numeric constant unique to units of measure (Clark et al., 2007; Tyson, 2002) System flow rate is likely the easiest variable to control by changing the size of the area to be irrigated. The benefit of drip irrigation systems is the ability to operate at low pressures, reducing energy use compared to other pressurized irrigation systems (Ayars et al., 2007). Low output pumps can be used by increasing the number of zones and irrigating on a rotation schedule. Total head deals with pressure associated with change in elevation due to gravitational force. Minimizing the effects of this component might be difficult due to restrictions in water source locations. The energy required to offset these factors is typically expressed in terms of pressure (kPa) or an equivalent height of a water column (m) often referred to as head pressure or head (Clark et al., 46 2007). An energy balance equation can be used to explain the effects of change in elevation and friction on head pressure: HA+ HP = “3* HF(A-B) (2) (Clark et al., 2007) Total system head is affected by starting point HA and an ending point H3. The amount of head pressure, gained or lost, between HA and H3 is affect by difference in elevation and acceleration due to gravity. The starting point HA for most systems typically is the location of the pump. Head lost due to friction HHA- 3) is based on the distance traveled from HA to H3 and varies based on pipe material and size. Depending on the change in elevation from the pump to different regions within the field, zone size or pump capacity may have to be adjusted accordingly. If the amount of head pressure added by the pump (Hp) is insufficient to keep the energy equation in balance, irrigation uniformity or system efficiency may suffer. The power output efficiency of pumps varies, with electric pumps often capable of operating at 100% of their rated power. The power of diesel pumps, or those utilizing an internal combustion engine, typically needs to be estimated approximately 15% greater than their rated power output, to factor wear and other operating inefficiency restricting them from being used at 100% capacity continuously (Tyson, 2002). To understand what size of zones are compatible with the system capacity and the amount of water desired, a mass balance approach can be used: (sts)(Tc)=(2-778)(A)(IGR) (3) 47 Where: st3 = system volumetric flow rate (L/s) TC = operating time per irrigation cycle (h) A = area irrigated (ha) IGR = depth per irrigation cycle (mm) 2.778 = multiplicative constant specific to the units used (Clark et al., 2007). Increasing zone size in attempt to irrigate a greater area could offset the equation resulting in poor uniformity or lack of coverage in certain areas. In our situations, the area of each zone was approximately 675 m2 in Horton and approximately 210 m2 in Sidney. If a larger area is desired to be irrigated, a larger pump is one solution, but may not be financially feasible, and an increased flow rate could place additional stress on pipes and fittings, causing damage to the system and additional maintenance concerns. Although the system pump may be capable of producing a certain flow rate, flow rates does not stay constant throughout the field. Variables such as the number of outlets, change in elevation, and friction inside pipes can have a major impact on the velocity of the water. A poor understanding of these factors might lead to increased flow rates, resulting in damage to pipes or fittings close to the pump and decreased application uniformity far from the pump. Calculating pipe size 48 The design and layout of the pipes is dictated by site conditions. It is important to keep system design simple and efficient. In both of our research locations 5 cm (2 in) pipe was used for mainlines, 2.54 cm (1 in) pipe for sub- mainlines, and 1.6 cm (5/8 in) drip-line tube was used. Additional piping and valves can increase costs and future maintenance. Typically, as the distance from the pump increases the diameter of the pipe should decrease to maintain velocity and ensure a continuous water supply to all irrigated areas. The velocity of the water inside the pipes is important when choosing a pipe size. To minimize damage to pipes and other components, by surge pressure or water hammer, a maximum flow velocity of 1.5 m/s is recommended (ASAE, 2000). Additionally, minimum velocities of 0.3 m/s or greater should be maintained to allow suspended particles to flush and avoid buildup and clogging (Clark et al., 2007). Determining pipe diameter based on rules for safe velocities and a known volumetric pump flow rate can be done with the following equation: a: 1000(0) (V) (4) Where: Q = systems volumetric flow rate (US) v = average velocity (m/s) a = cross section area of flow (mmz) (Clark et al., 2007) Typically, the output of any pump is constant and cannot be adjusted. Water velocity is determined by pump output and pipe size. Using a large pipe will result in safer velocities, due to an increased surface area, but might increase 49 costs substantially. Ideally, using the smallest pipe size capable of maintaining a safe velocity would result in the most cost effective system. Friction within pipes, fittings, and connections also reduces velocity. This loss associated with friction, friction head loss HF, should be taken into consideration for all parts of the irrigation system to ensure uniformity in application and can be calculated by: HF: Hf+ Hm (5) Where Hg: is the total head loss (m) in a given distance of pipe and Hm is the sum of minor friction losses from water flowing through connection, couplings, valves, etc (Clark et al., 2007). Pipe manufactures often provide friction coefficients for various types of pipe. If this information is not available, then it is necessary to calculate it. It is important to realize that different types of pipe, having the same outer diameter (OD), do not necessarily have the same inner diameter (ID). Variations of wall thickness due to quality, material, and construction method, can have a considerable effect on the ID. The Hazen-Williams equation, essentially allows one to calculate the friction loss, depending on the roughness of the pipe. The equation is as follows: 1.852 Hf: L(1.212x101°)(%) (04-87) (6) Where: L = pipeline length (m) Q = pipeline flow rate (L/s) D = inner diameter (mm) C = Hazen-Williams friction coefficient (unitless) 5O (Clark et al., 2007) The Hazen-Williams friction coefficient varies from 100 (old steel pipe with high friction) to 150 for various types of smooth plastic pipe having much less friction effect on flowing water (Clark et al., 2007). Other types of metal pipe, based on the age and roughness of the interior of the pipe, have intermediate values. Although this formula uses a qualitative evaluation of pipe friction, it is the most commonly used formula to calculate friction head loss (Clark et al., 2007). The majority of newer irrigation systems use plastic pipe with a Hazen-Williams coefficient of 150 for pipeline friction head loss calculations. Emitter spacing and wetting pattems Another very important factor to take into consideration before installation of the irrigation system is emitter spacing. When determining emitter spacing, one of the most important factors to consider is the soil type. The amount of lateral water movement in the soil depends on a variety of factors including soil type, textures, stratification, and application rates (Burt and Styles, 1994). The typical amount of lateral water movement for different soil types are shown in Table 2.1. These values, presented by Boswell (1984), represent the amount of lateral movement of surface ponded water on various soil types. Interpreting these numbers should be used with caution because many factors affect water movement, and the amounts of lateral movement presented may not represent field conditions. Generally, wetting patterns in a sandy soil tend to be more vertical than horizontal (Bresler et al., 1971; Levin et al., 1979), with silt loams 51 relatively uniform in all directions (Hachum et al., 1976), and this pattern becomes more horizontal as soil texture reaches a clay soil (Bar-Yosef and Sheikholslami, 1976). Soil texture is typically the criterion used to determine system design and emitter spacing (Hung, 1995; Reddy, 1998). However, many physical characteristics of the soil including structure may have a greater impact on wetting (Haverkamp et al., 1999), and relying solely on texture to predict wetting patterns is unreliable (Thorburn et al., 2003). Generally, regardless of soil type, an increase in application rate results in a narrow wetted area due to gravity influences (Roth, 1974). Surface wetting also has a major impact on the wetting pattern within the soil. This depends on the soil type, moisture content, and discharge rate of the emitter. The higher the discharge rates and the lower the infiltration rate of the soil, the greater the surface area that will be wetted (Howell et al., 1980). Understanding how all aspects of the soil interact with water movement is the key to choosing the right spacing for emitters. Testing multiple areas, where different soil properties are expected, would give the best determination of how the irrigation water responds when applied to the soil. Due to the long production time of Christmas trees and relatively wide spacing (1.8 rn), emitter spacing needs to accommodate a growing tree and a developing root system. In our situation, at both locations, the space between emitters was 61 cm which is the general recommendation for Christmas tree production systems (personal communication with irrigation salesman). Some 52 growers opt for wider 91 cm spacing although care needs to be taken to ensure emitter placement is close to each tree, due to this wider spacing. Development of the root system can depend on the location and quantity of water delivered and also the soil type and its physical properties (Barley and Greacen, 1967). A single emitter placed close to the tree might deliver a high volume of water and encourage leaching since the water will only reach a small area of the root system. This single emitter could limit root growth to the portion of soil wetted (Ayars et al., 2007). Including more than one emitter per tree will allow wetting of a larger area and encourage root growth. Any increased costs associated with using multiple emitters may be offset by a higher quality tree due to a more extensive root system. However, multiple emitters have drawbacks. Since spacing between emitters is constant, spacing allowing multiple emitters per tree would result in increased water use, irrigating large portions of the soil between trees. Drip line piping including multiple emitters per tree and spacing between emitter sets, to avoid irrigating bare soil, would require a custom application and might be cost prohibitive. System installation After determining pipe size, quantities, emitters, system design, and pumping capacity, the next step is system installation. When installing piping, routine maintenance and cultural practices need to be considered including the use of machinery and heavy equipment, all of which are capable of damaging piping, resulting in additional maintenance for the irrigation system. Typically, 53 mainline and sub-mainline pipes are buried, and the point of connection for the drip line tubing is exposed above ground. The drip line pipe is run along the base of the trees in each row and the space between rows is empty allowing space for equipment and machinery. There are various grades and qualities of drip line tube and emitters. The longevity of these components can be affected by frequency of use, water quality, water pressure and velocity, and environmental factors. In some situations, drip tape, a disposable less expensive alternative to drip pipe, might be applicable. Following Christmas tree harvest, if stumps are to be removed or the field is tilled in preparation for spring planting, the drip line irrigation present has to be dealt with accordingly. In our Horton research location the drip irrigated plot is a choose and cut system. As the trees are harvested, additional seedlings are planted along the drip line. In our Sidney location, the single aged field is grown for whole sale. At the time of harvest the field will be clear cut and the drip lines will have to be rolled up and moved or disconnected allowing the field to be prepared for another planting. A more disposable drip line system that can be replaced following each harvest may be more cost-effective. As with any irrigation system, drip line irrigation is not maintenance free. The rates of deterioration of the drip line piping will vary based on materials, quality of pipe, and environmental conditions. Application quantity and frequency Determining the amount of water that needs to be applied is one of the most important parts in deciding when to irrigate. Irrigating too little can lead to 54 water stress, resulting in less than optimum growth or mortality in severe situations. Excess irrigation causes leaching, soil degradation, and increases costs associated with system maintenance and operation. The empirical rule for irrigating Fraser fir, although not supported by scientific evidence, is 2.54 cm per week in the absence of rainfall, as reported by growers. Determining this application amount in terms of the irrigation system capacity can be done with equation 3. From a practical standpoint, specifying a time to irrigate offers little help unless the flow rate of the pump can be modified, subject to limitations of safe velocities. It would be more appropriate to modify equation 3, assuming a constant flow rate and a known irrigation amount, to understand the amount of time required to achieve such a goal. Volume = Depth * Area Flow rate Constant * Flow (7) Time = Solving equation 7 will provide an understanding of the time required to irrigate a given amount of water based on the limitations of the system and the area of coverage desired. This gives insight into the amount of water the system is capable of delivering and the time required to do so. If the time required to irrigate is unacceptable for the quantity of irrigation desired, decreasing the amount of irrigation per zone could help reduce this time. If decreasing application rates is not an option then it might be advisable to prioritize various zones, concentrating on the areas more prone to drying, or providing water to the trees that are most sensitive to water stress first. Furthermore, the output of the system should be matched to the output of the emitters. Emitters typically are designed to provide constant application rates at a specified operating pressure. 55 In our situations, for example, the emitters used had an application rate of 1.59 L hr". In heavier soils, or those with low infiltration rates, emitters with application rates of 0.98 L hr‘1 are more commonly used and irrigation is done for a longer period of time to reduce run off. Applying a lower volume over a longer period of time would also be a good strategy to utilize with a hydrophobic soil. Soil based irrigation An alternative to basing irrigation on a set application amount is to irrigate based on changes in soil moisture. Soil water holding capacity is based on particle size, density, and soil type. Furthermore, the availability of the water in different soil types can be divided into soil water content and soil water potential (Warrick and Or, 2007). Soil water can be expressed in terms of gravimetric or volumetric content, where gravimetric water content is the ratio of the mass of water held in the soil against gravity and the mass of the soil particles (glg or kg/kg). Similarly, volumetric water content is the ratio of the volume of the soil particles to the volume of the water in the soil (v/v) (Warrick and Or, 2007). Although labor intensive to obtain, gravimetric and volumetric water content can offer a good understanding of the water status and holding capacity of different soil types. Gravimetric water content can be measured by obtaining a soil core and oven drying it. The initial mass is taken and compared to the mass when oven dried, to determine water holding capacity. This information could then be used to make irrigation decisions. This might be difficult since the amount of water present in the soil can be affected by precipitation, temperature, 56 evaporation, evapotranspiration, and other environmental factors. To understand the water status of the soil at any given moment, samples would have to be taken, posing a challenge due to the lag time for obtaining results. Time domain reflectometry and neutron probes offer a non-destructive method of assessing gravimetric water content, but often require specialized training and are expensive to purchases, making them less than ideal for a Christmas tree production system. Tensiometers for irrigation A more practical method of making irrigation decisions would be to use an instrument or tool that monitors soil water conditions. Soil water potential deals with the energy status of the water present. Using tensiometers, this potential can be measured and used to assess irrigation needs. When using tensiometers, it is important to determine the quantity, size, and placement of the instruments. Tensiometers typically are available in a range of length allowing placement at various depths within the soil. Placement should be done where roots are actively growing to observe changes in water tension in the root zone. In our research locations, 30 and 60 cm tensiometer (lrrometer Company Riverside, CA) were placed in each zone. The 30 cm tensiometer was used to make irrigation decisions and the 60 cm tensiometer was used to asses the possibility of leaching. Location of the tensiometers is very important to ensure accurate reporting and representation of moisture tension for the various irrigation zones. High and low spots should be avoided as well as isolated areas 57 containing soil types, not uniform throughout the whole zone. Placement should be along the drip line, spaced away from the emitters, representative of where the plants are taking up water (Tam, 2006). This distance the tensiometers are from the emitter(s) should be similar to the distance the trees are from the emitter(s) to ensure tensiometers are in the wetted area (lrrometer, 2007). Determining when to irrigate, using tensiometers, will be based on many factors including size and physical condition of the trees, soil type, and forecasted weather. Tensiometer measurements of soil moisture tension are based on matric potential (kPa) ranging from field capactiy (5-20 kPa) to a soil reaching permanent wilting point; (>70 kPa) (Jamieson et al., 2002) the point at which water contained in the soil is no longer plant available. Monitoring change in tensiometers should be done on a daily basis to develop an understanding on how soil moisture tension changes with time and environmental conditions. Depending on the heterogeneity of soil among different zones, certain areas might respond more quickly to drying and wetting cycles than others. Water holding capacity is much greater in soils with smaller particles (clays and silts) than in soils with larger particles size (sands) (Figure 2.2). Based on water holding capacity, the length of time to reach an irrigation tension threshold will vary. The presence of sands and other soils with low water holding capacity would likely result in light frequent irrigation, compared to heavy infrequent irrigation in clays. Maintaining a soil near field capacity (5-20 kPa) would require frequent visits and good record keeping it soil types and tree size vary between zones. When the soil approaches permanent wilting point, 58 irrigation should be applied until there is adequate moisture to return to field capacity. The time necessary to irrigate a zone and bring tension levels to field capacity can vary based on application rates, infiltration, and moisture holding capacity requiring close observation of changes in tension rates to avoid runoff and leaching. This constant monitoring is very labor intensive and cost prohibitive prompting a need for an automated system. Automation using tensiometers Starting an irrigation event when a tensiometer reaches a desired tension might be a simple observation, but knowing how much to apply or when to stop irrigating can be difficult since the amount of soil water depletion is unknown. Checking tensiometer readings periodically during an irrigation event can be time consuming if multiple tensiometers are used throughout an irrigated area. To solve this problem, we recommend incorporating a method of automating the irrigation system. Many companies that sell tensiometers have options of equipping the tensiometer, with a pressure transducer. In our research locations, tensiometers were equipped with remote sensing units (RSU) for connection to a data logger (CR1000 Campbell Scientific Logan, Utah) allowing continuous monitoring of changes in soil moisture tension, and through a system of control components (AM16/32 multiplexer and SDM-CD16AC AC/DC controller, Campbell Scientific Co.) system automation was possible. To control irrigation events, a range of soil tension levels need to be specified. Specifying interval could result in many small frequent irrigation cycles, possibly cycling pumps and 59 valves too frequently prompting concern about system maintenance and longevity. Specifying a wide irrigation range might lead to excessive soil drying resulting in leaching or run off when irrigation does occur. An example of target irrigation tension thresholds and start/stop ranges is given in Table 2.2. The decision to Irrigate was based on a 12 kPa tension range for each of the different zones, with the exception of the non-irrigated zone. If the tension exceeded this range the system would activate a valve, initiating an irrigation event in the corresponding zone. Irrigation would stop when the tension reached the low range, of the 1:2 kPa threshold, based on the target tension. Irrigation decision could be made based on the 30 or 60 cm tensiometer and the target tension may vary depending on the soil type, tree size, etc. In our situation, the system was set up using a series of commands in a simple user-defined program. Programming could be changed and adapted to various situations to accommodate different tree sizes and water needs. An additional advantage to an automated system is accurate data logging and collection of the irrigation systems operation. Monitoring and wireless communication The principle of successful automation is based on the idea that all components, critical to the system, are working properly and in unison. In the event that the data logging station responsible for making irrigation decisions encounters an error, irrigation will stop. This could be costly in terms of water stress and reduction of quality of the crops being irrigated. Thus, it is important 60 to monitor data and ensure that the system is functioning properly, avoiding additional expenses. Tensiometers in particular fail in extremely dry soil (Jamieson et al., 2002). Frequent trips to check that the tensiometers and irrigation system are properly functioning negate the labor savings of automation. For this reason, it might be advisable to utilize wireless communication. In the automatic irrigation systems set up at our two research locations, the Campbell Scientific CR1000 data logger offers different options for collecting data. Using either Campbell Scientifics PC400 or LoggerNet data logger support software, direct connection is possible using the RS-232 port, linking a portable computer and the data logger with a serial cable. Direct cable connection does pose some limitations as a computer is required at the location of the data logger (Cheek and Wilkes, 1994). Doing so on a regular basis would require several trips to the field and in the event of system malfunction, the data will only be accurate to the last time the system was accessed. If the system is not checked routinely, data loss could span many days or weeks. Using wireless communication, it is possible to set up an automated irrigation system that can be controlled by any computer with the appropriate software and an active intemet connection. This setup allows greater flexibility with system monitoring and data collection by using a computer in a home or office. Customization of the software and data collection allows incorporation of charts and graphs for a visual representation of data and alarm notifications in the event a system malfunction takes place. 61 Summary Before making a decision to integrate an automated irrigation system into a new or existing stand of Christmas trees, it is critical to consider the amount of time required to do so. The best time to start planning for the installation of the irrigation system would be following harvest and holiday sales. This would allow sufficient time before the start of the growing season. In Michigan and other states with cold winters, low temperatures may place restrictions on system installation in terms of underground piping. Although it is still possible to dig and install pipe in sub-freezing temperatures, doing so increases the likelihood of damage to pipe and fittings. Also glued connections and fittings are more prone to failure if proper care is not taken when constructing these unions in sub- freezing temperatures. Following the holiday season when the coldest weather is likely to occur, measuring fields and determining pipe size and the number of zones needed would be the best place to start. This would include measurements for main and sub-mainlines, drip tubing, determining emitter spacing, and all the fittings necessary for proper connection. Also it would be an ideal time to determine tensiometer quantity and placement within the field and the amount of wire required to connect the tensiometers to the data logger and control station. It is advisable, and often easier for future repairs and maintenance to route wiring and piping along the same paths. When the number of tensiometers and valves has been determined, the next step would be to order the data logger and controllers. In our research locations, each station was equipped with a 62 CR1000 data logger, one AM16/32 multiplexer capable of connecting up to 32 tensiometers, and a SDM-CD16AC controller capable of controlling up to 16 valves. Along with the control components, software for accessing, programming, and acquiring data from the data logger is required with PC400 as the entry level requiring direct connection of Campbell Scientific data loggers and LoggerNet software for wireless communication. Prior to field deployment it is advisable to set up the data logging equipment and become familiar with system operations and integration to ensure a smooth transition to the field. The complexity of system wiring and programming varies based on the system size and the detail required in the irrigation system. Regardless of the complexity of the irrigation system, adequate time should be allotted to become familiar with system wiring, programming, and how to make modifications in the event that the irrigation plan does not meet the growers needs. Due to the nature of tensiometers, de-airing and minor maintenance might be necessary for a few days following field installation. Once properly installed and de-aired, based on manufacturer recommendations, maintenance should be minimal for the remaining of the growing season. As with any irrigation system, automated or not, periodic maintenance checks should be made to avoid any problems. An automated drip line system should require much less labor to maintain than a manually controlled irrigation system. Problems may include broken pipes and wires, emitter clogging, and 63 tensiometer failure. Additional time may be required during seasonal startup and shutdown times. Budgeting and costs Compared to a traditional overhead or manually controlled irrigation system, an automated system is more expensive initially due to purchasing equipment. However, these costs, averaged over the life of the irrigation system, are likely to be considerably cheaper than the manually controlled alternative. Table 2.3 shows a summary of costs of the various components and labor required to implement a system of this nature. Based on our experience, cost and quantity of components, the cost for building an automated irrigation system on 1 ha is $8,142. The cost of a drip irrigation system, exclusive of automation, can range from $1,500 to $3,500/ha with maintenance costs ranging from $50 to $200/ha/yr (Ayars et al., 2007). In our situations, the irrigation system cost was $2,600/ha based on the assumption that 80% of the cost was for drip tube, 15% for main and sub-mainlines, and 5% for connections and valves. Depending on the quantity, quality, and complexity of the desired components, costs could vary greatly. Tensiometers used in a manually controlled system are considerably cheaper, $65 30 cm and $85 60 cm, since the RSU component is unnecessary. The cost for adding the automation part of the irrigation system ($3,942) is based on the assumption that one 30 cm and 60 cm tensionmeter are installed for each zone, and increasing the number of zones will add to costs. Set up costs associated with the automated system ($1000), based on our experience, is the 64 time required to setup and properly implement the system in situ. Due to the complexity of connecting and programming an automated irrigation system, this estimate could vary considerably possibly requiring technical support and a lengthy learning curve. Since the tensiometers require connection to the data logger and peripherals, increasing the tensiometers beyond the means of the data logger might require the purchase of additional peripherals, further increasing costs. The wireless service necessary to access, modify, and view the workings of the irrigation system is based on a standard limited access data account and can vary among wireless carrier and usage. Labor associated with each type of system is assumed to be $200 in costs associated with installing and removing tensiometers from the field each year. Since the automated system functions without the aid of human labor, the operating costs associated with the automated system of $400 are based on a 4 month irrigation season (May — August) and a tech visiting the system for one 2 hour period each week at a rate of $10 hr to ensure all components are functioning properly. Similarly, the operating costs for a manually controlled system includes $200 in labor costs for installing and removing tensiometers and no additional set up costs since the automation components are not used. The maintenance costs for a manually controlled system are based on a 4 month irrigation season (May - August) and a tech visiting the system daily for a 2 hour period, making observations and determining when to irrigate at a rate of $10 hr. The labor intensive nature of the manually controlled system results in much greater operation costs when compared to the automated system, $2400 compared to $400, respectively. 65 Although the manually controlled irrigation system, initially is cheaper than an automated system ($8,142 vs. $5,350) the additional $2000 in operating costs for the manually controlled system result in the automated system being cheaper to overally in the second year of operating. Assuming components stay in good working order, this cost savings will become more substantial with time. Based on a 9 year rotation, an automated system results in a savings of $15,208 compared to a manually controlled system contributed to overall savings in labor. The cost summary listed in Table 2.3 is presented as a starting point to the investment required in building an automated system. Before deciding to implement such a system, vendors and suppliers should be contacted to get a more accurate cost based on the size and nature of the system required. Problems and considerations Although an automated irrigation system should be much less problematic than a manually controlled system, there are still a few areas of concern that should be considered. Tensiometers need to be installed and removed at the beginning and end of each growing season if freezing winter temperatures are expected. Tensiometers also have a tendency to fail in very dry soils. If an irrigation plan calls for excessive drying between irrigation events, other soil moisture measurement instruments might be considered. Also, soil tension reported is only true for the location the tensiometer is present, emphasizing the importance of instnrment placement. Tensiometer placement, relative to emitters, should be similar in relation to emitter spacing for the trees. Aside from 66 tensiometer functionality, care should be taken in securing exposed wire connections. The data logger and controllers experienced few problems in both of our research locations. Periodically, the data collected produced errors, but few data points were missing or erroneous. This was more likely to occur when the frequency of data collection was increased. Using wireless data acquisition, connecting quickly and maintaining a long connection was often problematic, due to intermittent cellular coverage in the area where the data logger and modern were placed. For this reason, it is advisable to check which wireless carriers offer the strongest coverage in the area to be irrigated. 67 Literature Cited ASAE. 2000. Design, installation and performance of underground thermoplastic irrigation pipelines. ASAE $376.2 JAN98. 1m ASAE Standards, St. Joseph, Michigan. Ayars, J.E., D.A. Bucks, F.R. Lamm, and PS Nakayama. 2007. Introduction, In F. R. Lamm, et al., eds. Microirrigation for Crop Production. Elsevier, Oxford. Bar-Yosef, B., and MR. Sheikholslami. 1976. Distribution of water and ions in soils irrigation and fertilized from a trickle source. Soil. Sci. Soc. Am. J. 40:575- 582. Barley, K.P., and EL. Greacen. 1967. Mechanical resistance as a soil factor in influencing the growth of roots and underground shoots. Advances in Agronomy 1921-43. Boswell, M.J. 1984. Micro-Irrigation Design Manual. Hardie Irrigation, El Cajon, CA. Bresler, l., J. Heller, N. Diner, l. Ben-Asher, A. Brandt, and D. Goldberg. 1971. Infiltration from a trickle sources. ll. Experimental data and theoretical predictions. Soil. Sci. Soc. Am. J. 35:683-689. Burt, CM, and SW. Styles. 1994. Drip and Microirrigation for Trees, Vines, and Row Crops. California Polytechnic State University, San Luis Obispo, CA. Chastagner, GA, and D.M. Benson. 2000. The Christmas Tree: Traditions, Production, and Diseases. Online. Plant Health Progress doi:10.1094/PHP-2000- 1013-01-RV. Cheek, S., and R. Wilkes. 1994. Monitoring processes using wireless data acquisition. Water Eng. Mgt. 144122-23. Clark, G.A., D.Z. Haman, J.F. Prochaska, and M. Yitayew. 2007. General System Design Principles, In F. R. Lamm, et al., eds. Microirrigation for Crop Production. Elsevier, Oxford. Hachum, A.Y., J.F. Alfaro, and LS. Willardson. 1976. Water Movement in soil from a trickle source. ASCE J. lrrig and Drainage Div. 102(IR2):179-192 Hall, D.G.M., M.J. Reeve, A.J. Thomasson, and V.F. Wright. 1977. Water Retension, Porosity and Density of Field Soils. Soil Survey Tech. Monogr. No. 9. Harpenden, UK. 68 Haverkamp, R., F. Bouradui, C. Zammit, and R. Angulo-Jaramillo. 1999. Movement of moisture in the unsaturated zone., In J. W. Delleur, ed. Groundwater engineering handbook. CRC, Boca Raton, FL., pp. 5.1-5.50. Howell, T.A., D.S. Stevenson, F.K. Aljibury, H.M. Gitlin, I.P. Wu, AW. Warrick, and FAQ Raats. 1980. Design and operation of trickle (drip) systems, In M. E. Jensen, ed. Design and operation of farm irrigation systems. ASAE, St. Joseph, MI. Hung, J.Y.T. 1995. Determination of emitter spacing and irrigation run time including plant root depth., In F. R. Lamm, ed. Microirrigtion for a changing world: conserving resources/preserving the environment. Proceedings of the fifth international microirrigation congress, Orlando, FL, 2-6 April, 1995. ASAE, St. Joseph, Ml., p. 292-296. lrrometer. 2007. lrrometer reference book - #24 [Online]. Available by lrrometer Company http://www.irrometer.com/pdf/24lrrometer_w-pictures.pdf (verified November, 2007). Jamieson, T., R. Gordon, L. Cochrane, A. Madani, and G. Patterson. 2002. Tensiometers and their use in irrigation scheduling. Nova Scotia Agricultural College. Water conservation factsheet No. 577.100-2 Leuty, T. 2005. Fraser fir for Christmas trees and landscape transplants [Online]. Available by Ontario Ministry of Agriculture, Food and Rural Affairs http://www.omafra.gov.on.ca/english/crops/facts/info_fir.htm (verified January 29, 2005) Levin, I., P.C. VanRooyen, and F.C. VanRooyen. 1979. The effect of discharge rate and intermittent water applications by point source on the soil moisture distribution pattern. Soil. Sci. Soc. Am. J. 4328-16. MDA. 2006. Michigan Department of Agriculture: Water Use Reporting. Act 451 of 1994 Part 327. Reddy, J.M. 1998. Selection of emitter spacing in trickle irrigation. In: Proceedings of the Fourther international microirrigation congress, Albury— Woodonga, October 1988, paper 6B-1. The Congress, Parkville, Victoria Roth, R.L. 1974. Soil moisture distribution and wetting pattern from a point source. In: Proceedings of the second international drip irrigation congress, San Diego, CA, p. 246-251. Shukla, S., C.Y. Yu, J.D. Hardin, and F.H. Jaber. 2006. Wireless data acquisition and control systems for agricultural water management projects. HortTechnology 16:595-604. 69 Tam, S. 2006. Irrigation scheduling with tensiometers. Water conservation factsheet No. 577.100-2. Vancouver, British Col. :British Columbia Ministry of Agriculture and Food. Thorburn, P.J., F.J. Cook, and KL. Bristow. 2003. Soil-dependent wetting from trickle emitters: implication for system design and management. Irrigation Science 22:121-127. Tyson, AW. 2002. Factors to consider in selecting a farm irrigation system. University of Georgia Cooperative Extension Service. Bulletin 882. Warrick, A.W., and D. Or. 2007. Soil Water Concepts, In F. R. Lamm, et al., eds. Microirrigation for Crop Production. Elsevier, Oxford. 70 Table 2.1. Lateral movement for ponded water for various soil types adapted from (Boswell, 1984). Additional Lateral Movement Soil Type ft m Coarse Sand 0.5 - 1.5 0.15 - 0.46 Fine Sand 1.0 - 3.0 0.30 - 0.91 Loam 3.0 - 4.5 0.91 - 1.37 Heavy Clay 4.0 - 6.0 1.22 - 1.82 71 Table 2.2: Example of hypothetical on/off tolerances to maintain various tension levels in zones controlled by a tensiometer based automated irrigation system. Zone Tensiometer depth Stop irrigation 1 2 3 30 cm 60 cm 30 cm 60 cm 30 cm 60 cm 30 cm 60 cm 30 cm 60 cm s 13 kPa s 23 kPa S 33 kPa S 43 kPa Target tension Non-irrigated 15 kPa 25 kPa 35 kPa 45 kPa Start irrigation 2 17 kPa 2 27 kPa 2 37 kPa 2 47 kPa 72 Table 2.3: Example of total costs of components and materials associated with building an automated irrigation system and a manually controlled irrigation system based on experiences installing automate irrigation systems in Horton and Sidney, Ml. Automated System Manual System Item Description Costs (5) Description Costs (5) Components CR1000 Data logger 1,350 Tensiometer (30 cm) 65 SDM-CD16AC Controller 695 Tensiometer (61 cm) 85 AM16/32 Multiplexer 560 Tensiometer (30 cm) 165 Tensiometer (61 cm) 185 CURS100 Resistor 52 Wireless Modem 340 LoggerNet Software 545 Wireless service1 50 Piping Trickle tubez 2000 Trickle tubez 2000 Pipes: main, sub-main2 375 Pipes: main, sub-main2 375 Connections / valves2 125 Connections / valves2 125 installation2 100 Installation2 100 Labor Set up 1,000 Operation 400 Operation 2,400 Maintenance3 200 Maintenance3 200 Total cost 8,142 5,350 1cost/mo 2cost/ha 3cost/yr 73 Zone 1 Zone 2 Zone 3 ................... ---I Zone 2 Zone 3 Zone 1 ——————————————————— u I —I Zone 3 Zone 1 Zone 2 - Vt L ...... —————— -I 1- J Zone 1 Zone 2 Zone 3 Figure 2.1. Example of an irrigated field divided into zones, and the valves and pipes allowing irrigation events to be zone-specific. 74 Matric suction (kPa) ‘~~~~‘ ‘s 40.... ‘s“ T‘s‘ ‘ \ \‘\‘ ‘\“ \\‘ x ‘\ \ N‘ \ \ \ ‘\ ‘x ‘ ‘ §\ \ \ \\ \‘ \ \‘ \ \ x \ \ \\ TT \\ \‘\ ‘\ 30“ \\ \ \‘\\\ ‘\ \\ Clay \\\‘ ‘\\\ \x \ \‘\“ \\\\ \\ ' E \\ \\\\ \\ \\\\ o \ \ ‘5 \ . O \ ‘5“ x ‘\ Silty clay loam o \ \ \ ‘~ \ or I m I: 201 \ \ \ \‘ ay ca 0) \ \\ \ \\ E \ \ ‘ \ 3 \\ \ \\ ~ Sandy clay loam 'H ‘x \\ \\\ C ‘s \\ \\\\ . 9:3 ( ‘\ \\ \\‘\‘ Sandy srlt loam 8 \ \\ \V‘ Silt loam .0.J ,0, \\ \\ Sandy loam H ‘\ (U \\ “~ s‘ ‘- 3 "-~“ Loamy sand \“‘ “‘----- Sand 6‘ r r r r l r 5 10 50 100 500 7000 Figure 2.2. Change in matric potential in response to changes in water content affected by different soil types from (Hall et al., 1977). 75 CHAPTER 3 RELATION BETWEEN ENVIRONMENTAL FACTORS AND SOIL MATRIC POTENTIAL IN FRASER FIR (ABIES FRASERI) PRODUCTION 76 Introduction Fraser fir (Abies fraseri) is an important species for landscape and Christmas tree production (NCTA, 2002). Fraser fir, native to remote locations of the Appalachian Mountains, is physiologically adapted to mild temperatures and plentiful precipitation (Beck, 1990). Unlike many other conifers produced in Michigan, water stress is of major concern with Fraser fir. This creates conflict between market demands for the species and the ability of Michigan growers to produce it. Among growers, it is commonly accepted that irrigation is essential to the quality and survival of Michigan-grown Fraser fir and irrigation decisions are qualitative, based on experience, common among agricultural and horticulture crops (Fereres, 1996; Fereres et al., 2003). Although a variety of quantitative methods are available to assess soil moisture status including gravimetric water content, water balance, and soil matric potential, these methods have not been applied to scheduling irrigation of Fraser fir. Soil matric potential (SMP) is a very useful method to assess changes in soil moisture and has been reported to provide a better characterization of crop water availability compared to gravimetric or volumetric assessments (Wang et al., 2007). Tensiometer have been used in a variety of different crops and regions as a tool for providing SMP information for scheduling irrigation (Hegde, 1987; Hegde and Srinivas, 1989; Phene and Beale, 1976). Benefits of using tensiometers for irrigation scheduling include increased yields (Hoppula and Salo, 2007; Smajstrla and Locascio, 1996) and water use efficiency (Shae et al., 1999; Wang et al., 2007). Compared to other tools commonly used for irrigation 77 scheduling, tensiometers offer flexibility in ease of use and affordability. Considering the benefits of using tensiometers, adapting these tools, as a basis for irrigation scheduling in Fraser fir, could serve as a starting point for irrigating in a quantitative manner. However, predicting changes in SMP can be difficult since similar site conditions rarely exist. SMP can be quite variable among spatial variability and soil water holding capacity (Hall et al., 1977). Furthermore, changes in environmental conditions also affect SMP. To make informed irrigation decisions effectively, SMP predicitive models, accounting for these changes, are needed. Potential evapotranspiration (PET) calculations attempt to account for plant and soil water loss, which is affected by environmental conditions including temperature, solar radiance, wind, vapor pressure deficit (VPD), and precipitation. Changes in heat flux, mainly due to temperature and solar radiation, interact strongly soil water (Buchan, 2001). Although the effect of these factors on the flow of liquid water is negligible (de Vries, 1975), water vapor flow is more strongly affected (Buchan, 2001). Although the interaction of individual factors on plant and soil evaporative demand might show a relation, collectively, these interactions could offer a better understanding of changes in SMP associated with water loss, based on potential evapotranspiration (PET). Considerable attention has focused on predicting PET with approximately 50 models developed (Lu et al., 2005). They vary in data requirements and tend to be site specific (Grismer et al., 2002). Among these models the Penman- Montheith model is the most widely used. This combination model is based on an 78 energy balance mass transfer equation (Penman, 1948) and a derivation of thermodynamics (Monteith, 1965), can be adapted to a site if the aerodynamic and surface resistance are known. The complexity of this model is less than ideal for Fraser fir, due to calculations associated with aerodynamic and surface resistance, on a complex canopy over a long rotation. A simplified model, the American Society of Civil Engineers (ASCE) combination Penman-Montheith model, uses a general approach for short and tall crops to simplify calculations. To test the suitability of this model, we applied the ASCE combination model for tall crops to Fraser fir production. Modeling changes in SMP based on PET estimates has a potential benefit for forecasting irrigation needs. Since the SMP is affected by many environmental factors, and PET attempts to account for these factors, developing a model using daily PET values to predict changes in SMP could serve as a starting point for forecasting irrigation needs in Fraser fir. The purpose of this study was to examine the relationship between SMP and environmental conditions at two locations in Michigan. The objectives on this study were to first use daily weather data to develop a PET model based on the ASCE combination Penman-Montheith equation for tall crops, and second, determine the suitability of PET as a predictor for changes in SMP. The information gained from this study can be used as a first step for quantitative irrigation scheduling guidelines for Fraser fir Christmas tree production in Michigan. 79 Materials and Methods The study included 2 years of data collection at two different Christmas tree farms with existing Fraser fir plantations beginning in 2006. The first location, Horton Michigan, is approximately 88 km south Michigan State University (MSU) East Lansing, MI. The site chosen for research was a choose- and-cut drip irrigated Fraser fir plantation approximately 0.4 ha in size. Due to the nature of choose-and-cut, the plantation contained a variety of tree sizes, ranging from new planted seedling to mature trees ready to harvest (approximately 2.5 m). Tensiometers were used to monitor changes in soil matric potential at a 30 and 60 cm depth. The tensiometers were lrrometer model R-RSU from the lrrometer Company Riverside, CA and were equipped with a remote sensing unit (RSU) which connected to CR1000 data logger from the Campbell Scientific Company, Logan, UT. Connected to the data logger were weather instruments for precipitation, humidity, temperature, solar radiance, wind speed and direction. The second research location was in Sideny Michigan, which is approximately 100 km northwest of the MSU campus. This site, 0.4 ha in size, consisted of similar sized Fraser fir approximately 4 years old. The plot was previously irrigated using an overhead water cannon, which for the purpose of this research was no longer used. Similar to the other location, a zone separated drip irrigation system was devised for the purpose of future research. The same set up was used for the tensiometers, data logger, and weather monitoring instruments as was used for the Horton plot. Two locations were used to help 80 develop a better understanding of how environmental conditions affect SMP under different soil types and production systems. A custom data logger program was written to collect temperature, relative humidity, solar radiance, soil tension, wind speed and direction, precipitation and tensiometer measurement data. Reference ET was calculated using the ASCE standardized reference evapotranspiration equation for tall crops: Cr) 0.408A(Rn 'G)+Y T+273 U2 (33 'ea) (1 ) A+V(1+Cd * 02) ETSZ = (ASCE, 2005) Where: ETSZ = standardized reference crop evapotranspiration (mm/d) Rn = calculated net radiation at crop surface (MJ/m2/d) G = soil heat flux density at soil surface(MJ/m2/d) T = mean air temperature (°C) u2 = mean wind speed (m/s) eS = saturation vapor pressure (kPa) ea = mean actual vapor pressure (kPa) A = slope of saturation vapor pressure-temperature curve (kPa/ °C) y = psychrometric constant (kPa/ °C) Cn = numerator constant that changes with reference type and calculation time step (K mm s3/Mg/d) 81 Cd = denominator constant that changes with reference type and calculation time step (s/m) (ASCE, 2005) The ETsz equation is a standardized equation with values for surface aerodynamics predetermined base on short or tall crops. Values for Cn and Cd were 1600 and 0.38, respectively, representing the constants for tall crops with a daily calculation interval. The minimum variables for computation of this equation are air temperature, solar radiation, and wind speed. Meeting the minimal requirements for ET calculations, some additional calculations were necessary before a value could be computed. Calculating for y, a psychrometric constant, requires information regarding atmospheric pressure and elevation. Elevations for both locations were obtained from the National Oceanic and Atmospheric Administration (NOAA) to calculate mean atmospheric pressure at both locations using a simplified Universal Gas Law equation: 293-0.00652 293 5.26 ] <2) P=101.3( (Burman et al., 1987) Where P is the mean atmospheric pressure (kPa) at the site determined by station elevation z (m) above sea level (ASCE, 2005). The mean atmospheric pressure P can then be multiplied by a constant to obtain the value of y by the following equation: v=0.000665(P) (3) 82 The slope of the vapor pressure-temperature curve A (kPa/°C) can be calculated from the daily mean air temperature T (°C) and the following equation: 2503exp[17'27 T] T+237.3 (T+27:-3.3)2 A= (4) To calculate saturation vapor pressure e$ (kPa) and actual vapor pressure ea (kPa) a function e°(T) (kPa) must first be calculated using T (°C): (5) e°(T)=0.6108exp[ "'27 T) T+237.3 The value obtained from eq. 5 can then be used to calculate ea (kPa) when relative humidity (%) is known: _ RH o ea-fie (T) (5) Saturation vapor pressure es (kPa) is determined by using daily minimum (Tmin) and maximum (Tmax) temperatures (°C) and eq. 5 to develop the functions e°(Tmin) and e°(Tmax) which are used in the equation: = eo(“l-max )TGOITmin) (7) es 2 Soil heat flux density was calculated using mean monthly air temperatures T (°C) using eq. 8: Gmonth,i = 0-07(Tmonth,i+1-Tmonth,i-1) (8) Where: Tmonth’i = mean air temperature of month | (°C) 83 Tmonth.,_1 = mean air temperature of previous month (°C) Tmomhm = mean air temperature of next month (°C) (ASCE, 2005) Combining the values calculated from equation 5 into equations 6 and 7, and then substituting the calculated values for equations 2-6, 7-8 into equation 1 will give a reasonable estimate for the daily ET predicted by the standardized ETsz value for tall crops. Weather data were summarized into daily values for the computation of ETsz values. Since ET losses are measured in mm/day and precipitation over the data collection period is measured in mm/day, a parameter was defined that essentially creates a water budget accounting for input and losses of water associated with ET and precipitation. This parameter, adjusted ET (AET) is defined by the following equation: i i AET = 2(ET) — 2(Precipitation) n n Where AET is the summation of the water losses associated with ET (mm) for the observation i for n observations and subtracted is the summation of the precipitation received (mm) for the i observation for n observations in the data collection period. The changes in AET and soil moisture tension were compared and relationships were determined using regression analysis. The strength of the relationship was determined with correlation coefficients. Data analysis was independent for both locations over the two year study. 84 Results The weather data used for the computation of ETsz were summarized into weekly averages (wind speed, temperature, relative humidity) and weekly totals (solar radiance, precipitation). Table 3.1 shows weather data for the two year collection period for Horton, Ml. Weather stations were installed in 2006 resulting in a shorter data collection period compared to 2007 (Table 3.2). Changes in Matric potential Changes in SMP based on soil moisture tension measurements taken at various depths give a good understanding how the frequency and quantity of precipitation affect soil moisture conditions. These changes were observed at Horton in 2006 and 2007. The soil moisture tension conditions for 2006 maintained at or near field capacity condition until day 219 when a drought induced rise was observed (Fig. 3.1a). Prior to this point rainfall intervals were frequent enough and volume was sufficient to keep soil in a field capacity condition. The increase in tension from day 219 continued for approximately 17 days when 15.2 mm of precipitation was received, causing a decrease in tension. Additional precipitation events were necessary, in the following days, before the tension, measured at 60 cm, was returned to a field capacity condition. Hydrophobic soil conditions may have been a contributor to this lag time, or the precipitation received at day 236 may have only been enough to affect moisture tension in the top 30 cm of the profile. Due to excessive drought conditions during this period, the small precipitation events observed on days 228 to 231 had little effect on returning the soil to a field capacity state. 85 Changes in soil moisture tension in 2007 at Horton were more variable than 2006 (Fig. 3.1b). The large precipitation event received after the start of the data collection period maintained the soil near a field capacity condition until day 161 where tensions continued to rise, peaking at 67 kPa, until precipitation events on days 198 and 200. The tension measured at 60 cm remained high, peaking at 65 kPa, until day 217 when 25 mm of precipitation was received. The rapid change in tension measured at 30 cm was likely due to the lack of available moisture at deeper depths resulting in quick water uptake in the top profile of the soil. It appears, based on observations made in Horton 2007, as the time between precipitation event increases, the amount of rainfall necessary to return soil to a field capacity conditions, also increases, more importantly at deeper depths in the soil. Changes in soil moisture tension at the 30 and 60 cm depths were observed for Sidney in 2006 and 2007. In 2006 the tension measured at the 60 cm depth stayed near field capacity during the entire data collection period (Fig. 3.2a). The change in tension measured at 30 cm was more variable and peaked at 23 kPa on day 234. This variability and quick flux in tension was likely due to the low water holding capacity of the soil. With this rapid flux in soil moisture tension due to the soils lack of water holding capacity, it would be probable to assume that the large precipitation event on day 192 did not have any more effect on soil moisture tension than the precipitation events on days 195 and 198. Changes in soil moisture tension observed in 2007 were more noticeable, compared to 2006, due to lower precipitation (Fig. 3.2b). A period of drought, 86 starting after the precipitation events to day 154, extended until day 184 when 14.7 mm of precipitation was received. Prior to this tension levels reached 71 kPa and although the 60 cm tension was lower, the increasing trend was still apparent. After the decrease in tension on day 184, levels quickly increased to 66 kPa on day 195. Again, this was likely contributed to the low water holding capacity of the soil and quicker moisture depletion in the top of the soil profile compared to the deeper depths. The trend of frequent increasing and decreasing tension, lasting until the end of the data collection period, suggests precipitation during this period was inadequate to counter the high evaporative demands, common during the droughty summer months of the growing season. This trend can be further emphasized with the tension at the 60 cm depth gradually Increasing over the data collection period, and the precipitation events during this span have little impact of decreasing the 60 cm tension to field capacity conditions. As the soil dries, moisture depletion first occurs at the soil surface, gradually progressing to deeper depths until plant available water may not be obtainable even at the deepest depths plant roots may penetrate. Understanding how evaporative demand affects soil moisture tension at the various depths would be a key point in forecasting the need of irrigation events based on different soil moisture tension levels. SMP and AE T Horton Research Location 87 Figure 338 shows how SMP, at 30 and 60 cm depths, and AET changed in 2006. AET and changes in SMP follow a similar increasing trend until day 236 when a precipitation event was received, 15.2 mm (Fig. 3.3A), decreasing the soil moisture tension at the 30 cm depth. Additional precipitation events, following day 239 were needed before a decrease in SMP, measured at the 60 cm depth, began to decrease. Prior to the decrease in SMP observed at day 239, SMP and AET increase agreed reasonably well with a positive exponential relationship (Figure 3.30). AET and SMP at the 30 and 60 cm depth share a positive exponential relationship R2 = 0.85 and R2 = 0.76, respectively. The relationship between 60 cm tension and AET is not as strong as the former, possibly due to higher moisture content at the deeper depths. The data at the beginning of the correlation line follow a linear relationship before exhibiting an exponential increase likely due to changes in soil water status. The trends of increasing AET and SMP appear to be in reasonable agreement during periods absent to rainfall and would likely develop stronger relationships as this time between rainfall events increases. In 2007 these relationships between increasing tension and AET were very apparent in part due to the high evaporative demands and low precipitation amounts received. Early in the data collection period, SMP values remained high and AET decreased slightly until day 237 where it began to increase steadily (Figure 3.4B). The increase in tension and AET appeared to be in reasonable agreement for the entire data collection period with exception to sporadic changes in tension at 60 cm from day 202 to 216. Following day 216 tension 88 levels stayed low for the remainder of the data collection period due to frequent precipitation events (Fig. 3.4A and 3.4b), despite the fact that AET continued to increase. Although it is likely that a strong exponential trend would be noticed between increasing AET and SMP from days 155 until precipitation events on days 198 and 200 a system error resulted in a data gap from days 178 to 183. Although a continuous relationship could not be drawn over this time period, the relationship between increasing AET and SMP showed a positive exponential relationship before and after the data gap. From days 155 through 177 and 184 through 200 increases in AET and SMP, at the 30 and 60 cm depths, showed a exponential relationship R2=0.76 and R2=0.65, respectively (Figure 3.40). The stronger relationship between the AET and the 30 cm tension, in comparison to the 60 cm SMP, was likely due to lower water content at the shallower depth. In the event that rainfall was not received for a further extended period of time, one can speculate that the 60 cm AET correlation would become stronger with a more exponential resemblance. Sidney Research Location The changes in soil moisture tension for Sidney in 2006 appear to agree with AET linearly although the SMP values never exceed 24 kPa (Figure 3.5B). Also the precipitation received was frequent enough to be more in balance with the losses from AET, indicated by a decreasing overall trend in AET during the data collection period (Fig. 3.5A). The relationship between the change in AET and SMP at 30 and 60 cm (Julian days 216 to 235) has a strong linear correlation 89 R2 = 0.986 and R2 = 0.936 respectively (Figure 3.5C). Due to the low water holding capacity of the soil at this location, it is assumed that if SMP would continue to rise the relationship with AET would resemble a more exponential relationship similar to what was seen the Horton location. In 2007 numerous occasions were observed where SMP values were much greater than those observed in 2006. Lack of precipitation and high evaporative demand caused 30 cm SMP to reach high levels on days 183, 195, 207, 218, and 231 (Figure 3.6B). Precipitation was less than adequate to replace soil moisture lost by ET indicated by the increasing AET trend (Figure 3.6A). This trend of increasing AET and SMP, for each of the drying cycles after a precipitation event, follows a strong correlation. SMP at the 30 cm depth had a strong correlation (R2 = 0.982) with AET and SMP at 60 cm also had a strong correlation relationship (R2 = 0.95) although somewhat more linear (Figure 3.60). This difference in relationship (linear vs. exponential) again was likely due to soils low water holding capacity. The rapid changes in SMP indicate the soil water quickly depleting and precipitation being readily adsorbed. The frequent wetting and dry of the top soil profile likely did not allow the soil at the 60 cm depth to become moisture depleted enough to express this exponential trend. Also the rooting depth of the trees present might have been primarily in the soil layers closer to the soil surface resulting in water greater water uptake from these layers in comparison to deeper depths. Discussion 90 Based on the results in this two year study, AET losses had a strong correlation with changes in soil matric potential measured at both a 30 cm and 60 cm depth. Difference in weather conditions between 2006 and 2007 were factors in the difference between SMP changes at both locations. Different soil types exhibit different matric potentials based on water content (Hall et al., 1977). Large particle soils, like sands, release more water at low matric potentials, in comparison to clays which release water gradually across a range of increasing matric potentials (Townend et al., 2001). In our two research locations the soil types were very different. The soil in Horton, MI is a Montcalm McBride sandy loam (NRCS, 2006) composed of approximately 50% sand and 50% silt and clay (NRCS, 2008). In contrast, the soil in Sidney is a Boyer Oshtemo loamy sand (NRCS, 2006) composed with as much as 85% sand with the remaining fraction silt and clay (NRCS, 2008). This difference in the soil textures at these two locations is a possible explanation for the difference in changes in SMP (Figures 3.1 and 3.2). This further emphasizes the importance of determining the relationship, between AET losses and changes in matric potential, on an individual basis instead of developing generalized recommendations. Additionally, the production systems at these two locations are very different with the Horton research location having a heterogeneous range of tree ages and sizes. The tree present in Sidney had very similar sizes and were all planted at the same time period, approximately 4 year prior to the start of this research. At this time the field was prepared for planting, cultivating the existing soil before the seedlings were planted in the field. The relationship 91 between changes in soil moisture content and matric potential is different in a disturbed soil compared to an undisturbed soil (Unger, 1975). Although the effects of the soil disturbance may have lesser effect on soil moisture characteristics in the later years, it further emphasizes the importance of quantifying soil moisture release on a site specific basis. The degree of increase in SMP for both research locations was less in 2006 compared to 2007. Due to frequent precipitation in 2006, SMP rates for Horton remained low with exception to one observation, approximately day 235 (Fig. 3.1a). Furthermore, SMP levels in Sidney 2006 failed to reach levels which soil moisture became depleted (Fig. 3.2a). The lack of observations of high soil moisture tension levels at both locations, for 2006, was likely the reason that a stronger correlation was not observed in Horton (Fig. 3.30) and the correlation observed for the Sidney data was linear in nature (Fig. 3.50). This linear relationship was likely a ‘lag phase’ or start of the exponential relationship, between increased ET and increase soil moisture tension, likely becoming more evident if precipitation was less, allowing SMP levels to increase far beyond field capacity. The data collected from 2007 allowed testing of relationship, between AET and changes in SMP, during extended drought periods. In one occasion, tension levels reached approximately 70 kPa in both locations (Fig. 3.1b Horton, Fig. 3.2b Sidney), and there were additional observations were tension levels reached slightly lower levels. These periods of increased tension allowed for a better understanding of how AET correlated with increase soil moisture tension 92 (Figures 3.40 and 3.60) and provide a more accurate realization of how soil moisture depletion and increased tension follow an exponential pattern. The strong correlation and additional data points gives a more accurate model to use for predicting these changes. These models would likely be more accurate at predicting changes in soil moisture tension, based on AET, and could serve as a tool to forecast irrigation decisions for growers throughout the growing season when irrigation is practiced. Although the models do not take into consideration the rapid changes in SMP affected by precipitation events, the main concern of water stress, in Fraser fir, would be the time period between these irrigation events. Overall, the standardized ASCE reference equation for PET was in agreements with changes in AET and soil moisture tension, for both locations during this two year study. A possible concern for using this equation in predictions is the limitations of its use to predetermined crops sizes. The coefficients used in the equation for this research were those for a tall crop, 0.5 m (ASCE, 2005). This poses some limitations and concern since the height of Fraser fir throughout the production cycle ranges from <05 m at planting to >25 m at harvest. Accurately predicting water losses associated with ET require knowledge of surface architecture and aerodynamic surface resistance, which could prove time consuming and difficult, due to long production time and variability between trees, thus the limitation of using a more tradition ET calculation equations (Monteith, 1965; Penman, 1948). Using the coefficients for crops 0.5 m in heights may lead to an over estimation of ET in trees less than 93 this height, and under estimation of ET in trees greater than this height. This may not necessarily be a limitation since it would likely result in liberal water usage when the trees are small, and prone to water stress due to immature roots system, and a conservative use of irrigation waters as the trees increase in size >0.5 m, when an established root system is likely developed. Conclusions The results from this research show that cumulative AET is a good predictor of changes in SMP at 30 and 60 cm measurement depths. The relationship becomes stronger over prolonged periods of time without rainfall likely due to moisture release characteristics of the soil. Although AET is not a good continuous predictor of SMP values following large precipitation events, the predictive models showed merit between these rainfall periods. This information is of particular use to forecast what soil moisture tension levels are likely to be reached (site specific) based on daily ET rates. Following a large precipitation event, irrigation is of lesser concern than during periods of prolonged drought. Also, it is important to understand the soil type present and how this will play a role in water release characteristic, as predictive models, in this situation, were site specific. From a production standpoint, if a grower has an ideal matric potential level which they initiate irrigation, this information could be particularly useful aiding in predicting irrigation scheduling. Although the effects of irrigation on this relationship between changes in AET and SMP were not modeled, it is expect that similar results can be obtained. Additional research is needed to 94 verify this relationship and furthermore aid in irrigation scheduling for Fraser fir produced in Michigan. 95 Literature Cited Allen, R.G., L.S. Pereira, D. Raes, and M. Smith. 1998. Crop evapotranspiration - Guidelines for computing crop water requirements. Food and Agriculture Organization of the United Nations, Rome. ASCE. 2005. The ASCE standardized reference evapotranspiration equation. ASCE-EWEI Task Committee Report, January, 2005 Beck, DE. 1990. Silvics of North America: Abies fraseri (Pursh) Poir. [Online] http://www.na.fs.fed.us/spfo/pubs/silvics_manualNolume_1/abies/fraseri.htm (verified August 2, 2007). Buchan, GD. 2001. Soil Temperature Regime, p. 539-594, In K. A. S. 0. E. Mullins, ed. Matric potential. Marcel Dekker, New York. Burman, R.D., M.E. Jensen, and RC. Allen. 1987. Thermodynamic factors in evapotranspiraton, p. 28-30, In L. G. James and M. J. English, eds. Proc. lrrig. and Drain. Spec. Conf.,, Vol. July. ASCE, Portland, Oregon. de Vries, DA. 1975. Heat transfer in soils, In D. A. d. Vries and N. H. Afgan, eds. Heat and Mass Transfer in the Biosphere, New York: Scripta, pp. 5-28. Fereres, E. 1996. Irrigation scheduling and its impact on the 21st century, In C. R. Camp, et al., eds. Evapotranspiration and irrigation scheduling. Proceedings of the International Conference on American Society of Agricultural Engineers. 547-553, San Antonio, TX, USA. Fereres, E., D.A. Goldhamer, and LR. Parsons. 2003. Irrigation water management of horticultural crops. Hortscience 38: 1 036-1 042. Grismer, M.E., M. Orang, R. Snyder, and R. Matyac. 2002. Pan evaporation to reference evapotranspiration conversion methods. Journal of Irrigation and Drainage Engineering 128:180-184. Hall, D.G.M., M.J. Reeve, A.J. Thomasson, and V.F. Wright. 1977. Water Retention, Porosity and Density of Field Soils. Soil Survey Tech. Monogr. No. 9. Harpenden, U.K. Hegde, D.M. 1987. Effect of soil matric potential, method of irrigation and nitrogen fertilization on yield, quality, and nutrient uptake and water use of radish. Irrigation Science 8:13-22. Hegde, D.M., and K. Srinivas. 1989. Effects of soil matric potential and nitrogen on growth, yield, nutrient uptake and water use of banana. Agricultural water management 16: 1 09-1 17. 96 Hoppula, KL, and T.J. Salo. 2007. Tensiometer-based irrigation scheduling in perennial strawberry cultivation. Irrigation Science 25:401-409. Lu, J., G. Sun, S.G. McNuIty, and D.M. Amatya. 2005. A comparison of six potential evapotranspiration methods for regional use in the southeastern united states. Journal of the American Water Resources Association 41:621-633. Monteith, J.L. 1965. Evaporation and environment, In G. E. Fogg, ed. Symposium of the Society for Experimental Biology, The State of Movement of Water in Living Organisms, Vol. 19. Academic Press, Inc., NY. NCTA. 2002. National Christmas Tree Association (NCTA) Real Tree Agricultural Census [Online] http://www.christmastree.org/statistics_industry.cfm#findings (verified January 15, 2008). NRCS. 2006. Web Soil Survey [Online] httpzllwebsoilsurvey.nrcs.usda.gov/app/WebSoilSurvey.aspx (verified October 3, 2006). NRCS. 2008. Natural Resources Conservation Service, Soil Texture Calculator [Online] http://soils.usda.gov/technical/aids/investigations/texturel (verified March, 2008). Penman, H.L. 1948. Natural evaporation from open water, bare soil, and grass. Proc. Roy. Soc. London A193z120-146. Phene, C.J., and CW. Beale. 1976. High-frequency irrigation for water nutrient management in humid regions. Soil Science Society of America Journal 40:430- 443. Shae, J.B., D.D. Steele, and BL. Gregor. 1999. Irrigation Scheduling Methods for Potatoes in the Northern Great Plains. Transactions of the ASAE 42:351-360. Smajstrla, AG, and SJ. Locascio. 1996. Tensiometer-controlled, drip-irrigation scheduling of tomato. Applied Engineering in Agriculture 12(3):315-319. Townend, J., M..J Reeve, and A. Carter. 2001. Water Release Characteristic, In K A. Smith and C. E. Mullins, eds. Soil and Environmental Analysis, Physical Methods Second Edition. Marcel Dekker, Inc., Basel, NY. Unger, P. W. 1975. Water retention by core and sieved soil samples. Soil Science Society of America Proceedings 39:1197-1200. Wang, F.-X., Y. Kang, S.-P. Liu, and X.-Y. Hou. 2007. Effects of soil matric potential on potato growth under drip irrigation in the North China Plain. Agricultural water management 88:34-42. 97 Table 3.1. 2006 and 2007 weather data summary for Horton, Ml research Data represents average wind speed, temperature, and relative humidity and total precipitation and solar radiance. locafion. Horton 2006 Weather Summa Month Days Week SolarTRadiance Wind Speed Temp Relative Humidity Precipitation (kW/m2) (km/hr) (°C) (%) (mm) June 28-4 1 24.5 4.5 21.1 75.3 40.4 July 5-11 2 246* 2.7 20.3 69.4 13.5 July 12-18 3 * 3.0 24.4 71.8 31.8 July 19 -25 4 28.7* 3.6 22.3 72.1 0.0 July 26 -1 5 38.5 4.2 25.8 82.3 33.3 Aug 2-8 6 41.0 3.8 23.7 75.4 10.7 Aug 9-15 7 43.9 3.7 20.5 69.5 0.0 Aug 12-28 8 33.2 3.0 20.9 75.7 6.6 Aug 23 -29 9 20.5 2.4 20.9 84.5 51.3 Aug 30 -5 10 26.8 2.9 17.5 80.4 1.5 Sept 6-12 11 21.8 3.2 17.6 82.7 25.9 Sept 13 -19 12 19.8 3.9 17.0 85.8 10.7 Sept 20 -26 13 20.6 6.1 13.4 80.1 9.7 Sept 27 - 30" 14 11.8 4.9 11.2 8_0.0 12.4 Average 3.7 19.8 77.6 Sum 302.3 247.7 Horton 2007 Weather summary _ June 1-7 1 34.2 4.8 19.2 76.8 63.0 June 8-14 2 47.7 3.1 21.5 61.0 0.0 June 15-21 3 44.2 4.0 23.1 61.9 8.1 June 22-28 4 299* 3.0 21.9 68.0 0.1 June 29-5 5 22.4" 3.8 20.5 58.5 1.0 July 6-12 6 50.1 6.4 24.2 57.4 0.0 July 13-19 7 30.6 3.5 20.2 71.3 17.5 July 20-26 8 43.0 2.9 19.7 66.4 8.4 July 27-2 9 43.6 2.1 24.4 69.1 9.7 Aug 3-9 10 27.8 3.1 24.3 77.7 65.5 Aug 10-16 11 31.4 3.0 22.6 75.6 0.8 Aug 17-23 12 18.8 4.5 19.0 84.6 93.0 Aug 24-30 13 30.6 3.0 21.5 78.3 15.0 Aug 31-6 14 34.1 2.4 21.1 71.9 0.0 Sept 7-13 15 19.9 4.4 17.9 81.2 30.2 Sept 14-20 16 28.2 4.0 15.9 68.3 0.0 Sept 21-27 17 24.7 3.9 19.9 71.4 9.7 Sept 28-30” 18 11.9 3.8 15.4 70.7 0.3 Average 3.6 20.5 fi .6 Sum 394.7 251.0 * Data not available due to system malfunction **Shortened week (<7 days of data included) 98 Table 3.2. 2006 and 2007 weather data summary for Sidney, Ml research location. Data represents average wind speed, temperature, and relative humidity and total precipitation and solar radiance. Sidney 2006 Weather Summary Month JDays Week Solar Radiance Wind Speed Temp Relative Humidity PTecipitation (kW/m’) (km/hr) (°C) (%) (mm) July 7-13 1 41.8 3.8 21.8 76.3 76'5 July 14-27 2 44.3 4.1 23.8 77.8 50.5 July 21 -27 3 272* 5.1 22.0 77.7 26.4 July 28-3 4 * 5.8 26.2 75.6 40.4 Aug 4-10 5 37.5 3.3 22.0 73.5 0.3 Aug 11-17 6 45.7 3.1 19.0 73.0 0.0 Aug 18-24 7 32.1 2.8 19.8 79.5 17.3 Aug 25-31 8 29.2 4.7 19.8 83.4 3.6 Sept 1-7 9 33.5 2.7 17.6 79.2 0.3 Sept 8-14 10 14.3 4.4 15.4 88.8 60.7 Sept 15-21 11 26.4 3.7 14.4 84.0 2.0 Sept 22-28 12 20.3 5.0 12.5 84.9 36.3 Sept 29-30 13" 1.65" 2.5 8.3 91.5 3.7 Average 3.9 18.7r 80.4 Sum 325.1 317.9 Sidney 2007 Weather Summary June 1-7 1 34.1 4.0 18.4 797» 45".5' June 8-14 2 50.3 3.1 20.4 66.7 0.8 June 15-21 3 45.7 4.5 22.4 66.4 4.6 June 22-28 4 45.3 3.4 21.1 71.1 0.0 June 29-5 5 42.8 3.3 19.1 65.9 15.0 July 6-12 6 44.7 6.7 22.7 67.6 5.8 July 13-19 7 34.6 4.0 19.1 77.7 12.4 July 20-26 8 41.1 3.0 19.5 69.5 9.9 July 272 9 43.5 3.0 23.8 71.5 13.0 Aug 3-9 10 30.9 2.5 22.5 80.0 14.5 Aug 10-16 11 34.2 3.0 21.9 76.4 9.7 Aug 17-23 12 22.3 4.1 17.8 85.3 70.6 Aug 24-25 13** 4.6 4.1 21.2 89.7 12.2 3.7 20.8 74.4 Summary 474.2 213.9 * Data not available due to system malfunction **Shortened week (<7 days of data included) 99 Horton 2006 80 70 ft"? " 60 a (a) A is 60 2 - 50 E m v 5 ~40 5 9 40 — ,3 8. - 30 I: 0 Q 'g 20 — ‘ 20 g 5 1" M -_ 10 0 _ Y I I g l.‘ trail 0 179 189 199 209 219 229 239 249 259 269 Julian Day Horton 2007 80 70 O) O I N O I Precipitation (mm) Matric potential (kPa) h C 0 A 152161 170179188197 206 215 224 233 242 251260 269 Julian Day Figure 3.1. Changes in matric potential measured at 30 cm (open circles) and 60 cm (closed circles) depths as influenced by precipitation (vertical bars), recorded at the Horton, Ml research location in 2006 (a) and 2007 (b). 100 Sidney 2006 80 70 ’5 (a) - 60 A : 5° 7 ~ 50 E a V “E r 40 .5 5 40 d — 30 ‘5 .‘E o- .9 '5’ 20 - ’ 20 E) ’2“ 10 °- 0 — 1 - r 0 188 198 208 218 228 238 248 258 268 Julian Day Sidney 2007 80 70 9:3 7°“ (b) 6‘? 9 ”30 e :: 6° ‘ 69 E d'- L 50 g £3 50 ‘ 0° ' d9 69 d? — 40 F 0:) 40 3 .° 0 p 05065? '- Cd’? 6’ .° . 1% 8 30 1 . ‘. .0.- .0 ... $.v .0 V' l- 30 :3- § 20 T n? . o o" 0 I — 20 E ‘2“ 10 - Lin-av- I ' I u 10 (1. 0g I..- fir Ell r WA I.IIIV I'll I 0 152 162 172 182 192 202 212 222 232 Julian Day Figure 3.2. Changes in matric potential measured at 30 cm (open circles) and 60 cm (closed circles) depths as influenced by precipitation, indicated by vertical bars, recorded at the Sidney, Ml research location in 2006 (a) and 2007 (b). 101 25 200 E . g 20 4- 150 E C 15 _ 19;; 41 100 f, E, 5 _ 50 Q 0 - 0 202 212 222 232 242 252 262 272 Julian day 80 200 A (3) («Lo 60 - — 150 E 5 E .5 40 — — 100 F5 (I) LIJ 5 <1: ,_ 20 - — 50 0 1 """ ' 1 1 1 1 1 1 0 202 212 222 232 242 252 262 272 Julian day 00 O J (C) C) O 1 y = 7 654460.0114)‘ R2 = 0.8511 ' ‘3' Tension (kPa) h 0 20 - o y = 5.3368e0.0124x R’- = 0.7549 0 1 1 1 1 0 50 100 150 200 AET (mm) Figure 3.3. Changes in AET affected by precipitation (A); matric potential measured at 30 cm (open circles) and 60 cm (closed circles) and cumulative AET (B); and regression analysis between AET and soil matric potential for Julian days 203 to 239 (C) for Horton, MI research location 2006. 102 A 50 E E 40 - (A) 8 30 - E E 10 1 ‘L I 0 n. l I 155167 179 191 203 215 227 239 251 263 Julian day 80 200 B ”a? 60 — ( ) A .5 40‘ if V’ LU g < 1. 20 9 0 1 1 g 155 167 179 191 203 215 227 239 251 263 Julian Day (C) 80 7 y = 11.1636”186x o 7,: 60 _ R2 = 0.7627 ° 0. O as .5 40 F o ° ‘3 o 0 o o ‘D 20 - ° 0 ' y = 9.7464e0'01m '— ' ’ ' " ' .R2=06464 l G T l l l T l -20 0 20 40 60 80 100 AET (mm) Figure 3.4. Changes in AET affected by precipitation (A); matric potential measured at 30 cm (open circles) and 60 cm (closed circles) and cumulative AET (B); and regression analysis between AET and soil matric potential for Julian days 155 to 177 (higher data points) and 184 to 200 (lower data points) (C) for Horton, Ml research location 2007. 103 A 30 E _ E, 25 g 20 — E 15 — 2:: 10 - 0 e 5~ 0. 0 216 223 230 237 244 251 258 265 272 Julian day 80 200 711‘ 1 (B) as E .5 40 — — 100 I: 2 Lu ,9 20 — - — 50 < O l I T l l l I T T O 216 223 230 237 244 251 258 265 272 Julian Day 80 1 A C g 60, 1 ) as y = 0.154x + 7.557 .5 40 - R2 = 0.9855 (D E: 20 y=0.0518x+ 12.526 R2=O.9358 0 50 100 150 200 AET (mm) Figure 3.5. Changes in AET affected by precipitation (A); matric potential measured at 30 cm (open circles) and 60 cm (closed circles) and cumulative AET (B); and regression analysis between AET and soil matric potential for Julian days 216 to 234 (C) for Sidney, Ml research location 2006. 104 35 500 E 30 1 —— 400 g :3 ‘ (A) —— 300 ’E‘ o _1 '5 l 1— V g. 15fl 200 E .g 10 , - 100 < 3% 51 1 10 0 ~ i ' 1. — -100 152 162 172 182 192 202 212 222 232 Julian day 80 500 B 9°? 1 - 400 ’5 60 — ( ) a? : ". A g 1; - 7;; - 300 E v Jo '1? ' .0 o . E '1. 111°: 1. .. - 9 ~ . - LIJ oz) 20 1 ..° .. - 0° .8, ' — 100 < I— ” n . 9. ' . O 9%,. :1» 0 l T I 1 r r I r I ”100 152 162 172 182 192 202 212 222 232 l Julian Day 80 - / A (C) y = 10.2230'012x 260° 5; 60 1 R2 = 0.9817 66° 9 es 9 c 40 - ‘ .% coo/03°» 3 oo,’ 1— 20 - - . 659x+ 13.537 383? R2=O.9496 -50 0 50 100 150 200 AET (mm) Figure 3.6. Changes in AET affected by precipitation (A); matric potential measured at 30 cm (open circles) and 60 cm (closed circles) and cumulative AET (B); and regression analysis between AET and soil matric'potential for Julian days 155 to 184 (C) for Sidney, Ml research location 2007. 105 CHAPTER 4 THE EFFECTS OF MATRIC POTENTIAL IRRGATION ON THE GROWTH AND WATER STRESS IN FRASER FIR (ABIES FRASERI) PRODUCTION 106 Introduction Soil matric potential (SMP) is a main factor determining water availability to plants (Mullins, 2001) because it reflects on how tightly water is held within the soil matrix (Marshall et al., 1996). SMP has been the basis for irrigation scheduling for a variety of crops (Bower et al., 1975; Smajstrla and Locascio, 1996; Thompson et al., 2007; Wang et al., 2005). Irrigation scheduling using SMP has shown merit in increasing yields (Hoppula and Salo, 2007; Smajstrla and Locascio, 1996). The scheduling process typically specifies an upper SMP target or soil moisture depletion level before an irrigation event is initiated. This upper target is often determined by relating yields to various SMP levels (Bower et al., 1975; Smajstrla and Locascio, 1996). Yield, however, is a difficult criterion to use for perennial ornamental trees such as Fraser fir (Abies frasen) grown for Christmas tree production because annual shearing is practiced to maintain quality (Hinesley and Derby, 2004). Production rotations of A. frasen' often span 8 years or greater (Nzokou et al., 2006) and increases in growth, associated with irrigation, may not be noticed until the following growing season or later. Several studies have focused on plant water status as a possible solution to determining SMP levels for irrigation scheduling on various crops (Fereres et al., 2003; Goldhamer et al., 2003; Thompson et al., 2007). Changes in trunk diameter have been shown to correlate with changes in plant water status (Cochard et al., 2001) with soil water availability directly affecting changes in trunk diameter (Ortufio et al., 2004). Several other research projects have shown stem water potential (SWP) as a reliable indicator of plant water status and have 107 merit in irrigation scheduling (Naor et al., 1999; Shackel et al., 1997). However, differences in transpiration status can affect plant water potential measurements (Turner and Long, 1980). Leaf and air temperature, associated with transpiration, are related to soil moisture depletion (Ehrler, 1973). Transpiration produces a cooling effect due to latent heat of vaporization. However, if moisture becomes limited, the plant reacts by closing its stomata in an attempt to preserve water, thereby losing the cumulative cooling effect and raising leaf surface temperature. The accumulation of the difference between air temperature (TA) and leaf canopy temperature (To), summed over a period of time, are related to yield and water requirements (Jackson et al., 1977). Tc and TA have also been used for the calculation of a plant stress status known as crop water stress index (CWSI) (ldso et al., 1981). CWSI has been used for evaluating plant stress and for making irrigation decisions for many agricultural crops (lrmak et al., 2000; Orta et al., 2003; Payero and lrmak, 2006; Payero et al., 2005). However, CWSI baselines tend to be site-specific, and this lack of transferability limits the usefulness of CWSI for irrigation scheduling (Alves and Pereira, 2000). Only limited anecdotal information is available on water management and irrigation scheduling practices for A. fraseri (Nzokou et al., 2007). Determining optimum irrigation ‘ scheduling criteria based on SMP requires a good understanding of the effect of SMP on crop water stress and its subsequent impact on tree growth. Irrigating based on optimal SMP thresholds would lead to greater water use efficiency (WUE) and a decrease in costs associated with 108 irrigation. The goal of this study was to determine irrigation scheduling guidelines for A. fraseri based on SMP. Using predetermined SMP irrigation thresholds, change in height growth and basal area, along with changes in CWSI and plant water status were evaluated as the basis for determining optimum SMP for A. frasen’. The specific objectives were to (1) evaluate the effects of SMP A. fraseri growth and survival at various stages of the production rotation and (2) evaluate the effects of SMP on the SWP and CWSI. Materials and Methods The study was conducted at two sites for two years starting in 2006. Horton Site A Christmas tree farm in Horton, MI was the first location selected for research. The site selected for this research was a choose-and-cut Fraser fir plot, approximately 0.4 ha in size, with tree sizes ranging from new seedlings to mature trees ready for harvest. The existing drip line irrigation system was modified to allow the set up of irrigation treatments. The site was divided into 5 plots, representing of 4 irrigation treatments and a control. Electronic valves were installed allowing each irrigation treatment to be independent. The valves were connected to a control station which included a CR1000 data logger, AM 16/32 multiplexer, and an SDM-CD16AC ac/dc controller allowing modulation of the electronic valves (Campbell Scientific Inc. Logan, UT). To control irrigation and monitor changes in soil matric potential, 30 cm and 60 cm length tensiometers were used. The tensiometers, models R-RSU (lrrometer Company 109 Riverside, CA), were equipped with a remote sensing unit (RSU) to allow connection to the data logger. The tensiometers were placed in a central location within each plot and were spaced from emitter's representative to how the emitters were spaced from each tree. Also within each plot was an infrared temperature sensor (IRT) model lRR-PN from the Apogee Instruments Company Logan, UT. Sensors were placed in a central location in each plot, representative of the whole plot and mounted above a median sized randomly selected tree, allowing the field of view to focus on the tree canopy. Sensors were connected to the data logging station. Of the five plots, four plots were irrigation treatments and one was a non- irrigated control. The four irrigation treatments corresponding to 4 target SMP levels were 15 kPa, 25 kPa, 35 kPa, and 45 kPa. The decision to irrigate each plot was based on the 30 cm depth tensiometer and a 1:2 kPa tolerance (ex. 15 kPa treatment initiates irrigation 217 kPa and stops irrigation 513 kPa). Irrigation was fully automated via custom written data logger programming. A survey was done to quantify the size distribution of all of the Fraser fir present in each of the plots. Six sizes classes were designated which included < 0.6 m, 0.6-0.9 m, 0.9- 1.2 m, 1.2-1.5 m, 1.5-1.8 m, 1.8-2.1 m in height (<2 ft, 2-3 ft, 3-4 ft, 4-5 ft, 5-6 ft, 6-7 ft). Within each plot 20 trees from each of the 6 heights classes were randomly selected for measurement and data collection, totaling 600 trees in all. Measurements were made in the spring, prior to the start of growth, and the in fall, before shearing, as to the overall heights and the stem diameter. Differences in height and stem growth was quantified by dividing the change in 110 growth by the overall size of the individual trees (cm), resulting in a relative mean height growth (cm/cm) and relative basal area growth (mm2/mm2) for each size class. Sidney Site A setup similar to Horton was established in Sidney in 2006. The site selected for research was an existing Fraser fir plantation, with similar in age and size (approximately 4 years), with an area of approximately 0.16 ha. There were approximately 600 trees present of which were randomly assigned to 25 tree blocks, 5 x 5 in size, and assigned to one of three treatments that were replicated 4 times. The treatments were irrigation treatments designated by soil moisture tension and included a 15 kpa and 25 kPa treatment and a control. Tensiometers were placed at a 30 cm and 60 cm depth in each plot and connected to a data logger, similar to the Horton site. Irrigation in 2006 was manually controlled, but based on the same conditions of the automated system in Horton. In 2007 an automated drip system was installed, similar to the Horton site. An additional research plot was added to include research on Fraser fir seedlings. The field was prepared and planted in the spring of 2007 and an automated drip line irrigation system was put into place shortly after planting. The irrigation treatments specified were the same as in Horton, and the site was divided into 25 tree blocks, replicated 4 times, resulting in 600 seedlings across the 5 irrigation treatments and control. Measurements were made in the spring 111 and fall to determine relative height growth (cm/cm) and relative basal area (mmzlmmz), similar to Horton. Beginning and end of season height measurements were done in 2007 for the seedling plot to determine relative height growth (cm/cm). Seedling mortality was assessed visually and the number of dead seedlings was recorded in the fall. Crop Water Stress Index To determine the crop water stress index for Horton the procedures defined by ldso et al., 1981 were used. This uses water stressed and non-water stressed baselines to compute the CWSI. The CWSI can be defined as: CW 3| ___ (Tc 'Ta )M '(Tc 'Ta )max (1 ) (Tc "Ta )min '(Tc 'Ta )max Where To is the temperature of the canopy, Ta is the air temperature, M, min, and max represent measured, lower limit, and upper limit values, respectively. Upper and lower limit baselines were determined by plotting daily 12:00pm Tc-Ta measurements against the vapor pressure deficit (VPD). The lower limit is the non-water stressed baseline having a negative slope when plotted against the VPD. The upper limit typically has a zero slope when plotted against the VPD since a fully water stressed plant will not be transpiring and the TC will be unchanged. The VPD was calculated as the difference between the saturation vapor pressure es and the actual vapor pressure ea: VPD = eS - ea (2) Where es is the saturation vapor pressure kPa defined by: 112 = e°(Tmax) "’ e°(Tmin) (3) es 2 Where e°(T) is the function to calculate saturation vapor pressure defined by: (4) e°(T)=0.6108 exp[:r177é%j Where T is the temperature (°C) measured at the time for which the es is being calculated. The actual vapor pressure was calculated from measurements of relative humidity (RH) as defined: _ 511 11 ea - 100e (T) (5) Where RH is the relative humidity (%) for the given time period of measurement (ASCE, 2005). Stem Water Potential Stem water potential (SWP) was measured at the Horton research location in 2007 starting May 30. Measurements were taken two times weekly on Monday and Thursday between the hours of 12:00 noon and 2:00. The measurements continued until July 30. Three different height classes of trees were designated for measurement to develop an understanding how stem water potential is affected on different sizes of trees subjected irrigation. The sizes classes designated were small (<.9 m), medium (.9-1.8 m), and large (>1.8 m) trees. Fifteen trees from each of the height classes were randomly selected throughout each of the 4 irrigation treatments and control for a total SWP sample size of 75 trees. Measurements were done using a field plant water status 113 console, model 3115, from the Soilmoisture Equiptment Corp. Santa Barbara, CA. Procedures for measurement followed those outlined by Scholander et al., 1965. Cuttings approximately 5-7 cm were taken from each tree, mid height. Cuts were made to ensure that each clipping include woody tissue in comparison to the more herbaceous current year growth. Each clipping was promptly placed in the pressure chamber after removal, preventing additional dehydration of the sample. Statistical Analysis Treatment effects including change in relative height, relative basal area, and mortality were tested using ANOVA f-test procedure using a significance value of 0.05. Where mean significance was reported, mean separation was done with protected Fishers LSD = 0.05. Crop water stress index water stressed and non-water stressed values were related to VPD using regression analysis to establish and upper and lower baseline for the CWSI calculation. Stem water potential data was compared among treatments and size classes significance was reported using ANOVA when treatment means p< 0.05 and mean separation was done with protected Fishers LSD = 0.05. Results Horton site Frequent precipitation events led to overall lower SMP values in 2006 compared to 2007 (Figure. 4.1). In 2006 there was no discernable difference between irrigation treatments until day 219. Time following resulted in elevated 114 SMP values allowing a differentiation to be observed between irrigation treatments and the SMP values that they reached (Figure. 4.1A). Problems with the irrigation system on day 221 resulted in a failure to irrigate properly and the 25 kPa SMP treatment reached a high of 48 kPa on day 225 before irrigation resumed, returning the 25 kPa treatment into the desired range. Other than this single instance of system malfunction, treatments irrigated properly and post day 239, precipitation was frequent enough that all treatments were maintained near a field capacity SMP. In 2007, precipitation was less frequent and resulted in more desirable conditions to allow clear separations between SMP irrigation treatments (Figure. 4.18). Although a data gap existed between days 177 to 184, irrigation treatments functioned properly within the 12 kPa tolerance, indicated by decreasing SMP measurements following a peak (e.g. day 192, 198 for 45 kPa treatment, Figure 4.18). The conditions observed for 2007 SMP data are more ideal and likely to result in more differentiation of growth between trees in each treatment. Among all observations of difference in relative height growth, differences between treatments (P<0.01) was only observed for the B size class in 2006 (Table. 4.1). Over all there was a positive interaction between decreasing SMP irrigation treatment and increase relative height growth, with exception to the 15 kPa treatment similarly for the A size class in 2006 and the B and E size class in 2007 with exception to the 35 kPa and 45 kPa treatments, respectively. 115 Additionally, in comparison, relative height growth was generally greater in 2007 for size classes A and B, and greater in 2006 for size classes C through F. The response of mean change in relative basal area (mmzlmmz) due to irrigation treatments, were observed in 2006 and 2007 (Table. 4.2). The effect of irrigation treatments had little effect, in terms of significance between treatments, on mean differences in basal area for all height classes in 2006. Overall the increase in basal area was greater in 2006 than 2007, likely due to frequent precipitation. Difference between SMP irrigation treatments and increase in basal area were observed for size classes A and B in 2007 ([A] P<0.001 and [B] P=0.001) and collectively over the 2 years ([A] P<0.005 and [B] P=0.001). Trends followed a positive relationship between increasing relative basal area and decreasing SMP, with the 15 and 25 kPa treatments generally resulting in more change in basal area compared to the 35 kPa, 45 kPa, and control. This trend was not evident for any of the other size classes, regardless of treatment. The significance between size classes and the change in all growth parameters measured, for 2006 and 2007, and collectively for the 2 years, can be seen in table 4.3. The effects of SMP irrigation treatments, as determined by the ANOVA F-test procedure (P<0.05), were significant for the the A size class relative basal area in 2007 (P<0.001), and collectively over the 2 years (P=0.005), and for the B size class relative height 2006 (P=0.002), basal area 2007 (P=0.001), and collectively over the 2 years (P=0.001). Sidney site 116 Precipitation was frequent enough in 2006 resulting in only one period where SMP reached a high level of approximately 45 kPa on day 251 for the control plot (Figure. 4.2A). Issues associated with the control tensiometer resulted in sporadic measurements between days 206 and 250, although these problems were not an issue with the 15 and 25 kPa treatments. Overall irrigation treatments operated properly, maintain the SMP within the 12 kPa tolerance. In 2007 the research project was expanded to include the seedling plot and 4 irrigation treatments, in addition to the 2 existing treatments on the single aged stand. Precipitation was less frequent allowing treatments to reach their upper limits on several occasions (Figure. 4.23). Problems with the irrigation system resulted in SMP values greater than their :2 kPa threshold for the 15 kPa treatment on days 179 and 235. With exception to these two occasions the system operated properly and frequent wetting and drying cycles were observed (Figure. 4.23). Irrigation did not have an affect on relative height growth (cm/cm) in 2006 (Figure 4.3). Results obtained for relative height growth showed a significant difference between treatments (P<0.01) in 2007. Although there were no difference between the control and 25 kPa treatment for relative height growth, differences were observed (LSD 0.05) between the control (0.31 cm/cm) and the 15 kPa treatment (0.37 cm/cm). Results for mean change in relative basal area were similar for those of mean change in relative height growth. In 2006 there was no significant difference between treatments, although the 15 kPa treatment produced a higher 117 mean than the 25 kPa treatment and control (Figure. 4.4). In 2007 there was a significant difference between treatments (P<0.001). No significant differences were observed between control and the 25 kPa treatment mean height growth (LSD 0.05) although there was significance between mean height growth for the control (0.36 mmzlmmz) and the 15 kPa treatment (0.51 mm2/mm2). These difference were further emphasized showing a greater difference in 2 year relative basal area growth (P<0.0001). Similar to 2007, the 15 kPa irrigation treatment (2.2 mm2/mm2) resulted in more growth (LSD 0.05) compared to the non-irrigated control (1.82 mmzlmmz) collectively for the 2 years. Seedling survival generally decreased with increased SMP levels in 2007 (Figure. 4.5). Irrigation increased survival (P<0.05) compared to the control. The 25 kPa treatment produced the fewest dead seedlings (0 trees) followed by 15, 35, 45 kPa, and control plots. Although the 25 kPa treatment resulted in the fewest dead seedlings significance difference (LSD 0.05) was only apparent in the 45 kPa and control plots with 6 and 7 dead seedlings, respectively. Differences (P<0.05) were observed for relative mean height growth among the treatments (Figure. 4.6). The 25 kPa and 15 kPa treatments relative height growth (0.27 cm/cm and 0.28 cm/cm, respectively) was greater (LSD 0.05) than the 35 kPa treatment (0.23 cm/cm). Relative height growth values, although highest for the 15 kPa and 25 kPa treatments, were not significantly different from the 45 kPa and control, therefore making it difficult to associate a trend. Crop water stress index 118 A regression of calculated vapor pressure deficit values were plotted with Tc-Ta values to determine the upper and lower limits for CWSI calculations. For 2006 the upper limit was calculated to be 1.3° C with the lower limit defined by Y: -11.455x + 0.3618. Using these limits, the CWSI was calculated and plotted from Julian days 202 to 273 (Figure. 4.7). In 2006, measurements for CWSI were made for the control and 15 kPa plots. Overall the trends followed a similar pattern of gradualling increasing over the data collection period. The control plot CWSI values appeared to increase slightly more, overall, in comparison to the 15 kPa treatment (Figure. 4.7A). The mean CWSI values, over the collection period for the control and 15 kPa treatment were 0.51 and 0.53 respectively. Although the control, overall, had a lower mean CWSI, the variance in the control was greater (0.078) compared to the 15 kPa treatment (0.074). Overall CWSI values followed a similar trend and for both the control and 15 kPa plots, and no association were made between increasing SMP and increasing CWSI values for 2006. In 2007, the upper limit for calculating a CWSI was determined to be 13.3° C with the lower limit defined by Y=-19.25x — 0.6167. Using these limits, the CWSI was calculated and plotted from Julian days 152 to 273 (Figure. 4.8). Slight differences were noticed between overall CWSI values when compared to the different treatments. These values generally, increased as the SMP treatment increased in the order of 15 kPa, 25 kPa (not shown), and 35 kPa. Malfunctions with the 45 kPa IRT sensor made it difficult to calculate accurate CWSI estimates; therefore this treatment was not included. Overall the control 119 had the lowest CWSI values (Figure. 4.8A) with a mean CWSI over the data collection period of 0.2 and a variance of 0.025. The 15 kPa treatment was slightly higher (Figure 4.80) with an average CWSI value of 0.4 of 0.01 with the 35 kPa treatment having the highest overall CWSI values (Figure. 4.88) with a mean CWSI 0.5 and a variance 0.018. The lower CWSI index values observed in the control may have been partially due to difference in the tree measure. With exception to the control, the trend of increasing CWSI with increasing SMP is what would be expected as water availability decreases. Stem water potential Stem water potential measurements were based on size classifications small, medium, and large trees, within the various irrigation treatments. For small trees (<0.9 m), generally the highest stern water potential (Illsgem) was observed for the control plots (Figure. 4.9). Measurements between treatments were highly variable making it difficult to associate trends with the various irrigation treatments. The least variable treatment, for change in WStem was the 15 kPa treatment, although there were several instance where it exceeded the other treatments and the control (eg. Julian days 155, 165, 176, 179, 211). Similar variable results were obtained for Wetem for medium (0.9-1.8 m) tall trees (Figure. 4.10). Overall the control plots appeared to have the highest WStem. Similar to the small height class trees, the 15 kPa treatment had the highest 4’5th value on one occasion, Julian day 176. The general trend over the measurement period was a slight increase in values until approximately day 179 120 when values began to gradually decrease for the remaining time. The W313,“ measurements for all treatments for large trees (>1.8 m) overall were closely related for data collection period (Figure. 4.11). The control plot had showed an increasing trend until approximately Julian day 172 before it began to fall and follow a similar trend as the irrigation treatments. Associating trends with the difference in treatments with the difference in WStem is difficult due to the similarities of measurements with exception to the overall slightly higher measurements for the control plot. Although there were multiple occasions of significant difference (*indicates P<0.05) between treatments for all three size classes, mean separation (LSD=0.05) resulted in no clear trend (Table. 4.4) between change in matric potential and WStem, consistent to what is observed in figures 4.9, 4.10, and 4.11. Regression analysis between change in matric potential and W519," further supported the lack of trends resulting poor correlations (R2<0.139) for the majority of the comparisons (data not shown) and only two occasions where correlations were greater (R2=0.213 and R2=0.378), although still indicating a weak relationship. A summary of the three size classes WStem, regardless of irrigation treatment, is show in figure 4.12. Although this does not taking into consideration the effect of irrigation treatments, it is shown that l~l13tem decreases based on tree size. These decreases are evident from the beginning of data collection until Julian day 197 when the trend is no longer apparent. Prior to Julian day 197, there are several observations where the differences are significant (P<0.05) 121 between the size classes (Julian days: 150, 158, 162, 165, 169, and 186). These differences suggest a possible difference in water stress, among trees sizes, likely due to the difference in root system development. Discussion The results of this study demonstrate that irrigating based on SMP affects the growth of A. frasen' in smaller trees (<0.9 m in Horton, and 4 yrs in Sidney) when subjected to various SMP irrigation regimes. The difference in growth would likely be more apparent if precipitation was lesser in 2006, allowing SMP values to reach the upper threshold of their 12 kPa irrigation tolerance, similar to what was observed in 2007 for Horton and Sidney (Figures. 418 and 4.28, respectively). The effect of the various SMP irrigation treatments only resulted in significant differences (P<0.05) when the lowest SMP irrigation treatment, 15 kPa, was compared to the control for both Horton (Table. 4.1) and Sidney (Figure. 4.3). Quantifying change in height is difficult to do since annual shearing is practiced to maintain esthetics and a uniform shape (Hinesley and Derby, 2004). Although not considered for this study, the effects of shearing could have played a major role in the increase in height growth due to terminal bud vigor, and the quality of the cut that was made. From a production stand point, increased height growth could be a negative result since leader branches are typically pruned to a predetermined length to maintain canopy density. This increase in growth could result in additional labor required to remove and possibly have a negative physiological impact on the tree. 122 Most changes in stem diameter occurs in the living part of the bark (Molz and Klepper, 1973) suggesting it is a good indicator of water stress within the plant. Similar to the changes in tree height, results obtained for changes in basal area showed that lower SMP irrigation treatments resulted in a significant increase (P<0.05) in basal area compared to the control treatments on smaller trees (<0.9 m, Horton and 4 yrs Sidney). These results were only observed in 2007 and collectively for the 2 years of the study for Horton (Table. 4.2) and Sidney (Figure. 4.4). These results are in agreement to other studies that have used SMP based irrigation (lntrigliolo and Castel, 2004). Furthermore, the lack of significant difference for change in basal area for trees >0.9 m (Table. 4.2) suggests that increased SMP levels have a lesser effect on the larger trees. This is possibly due to more developed root system, reaching deeper depths where SMP might be lower due to higher water content. The effects of SMP irrigation treatments were more pronounced in the newly planted seedlings. Although the significance among SMP irrigation treatments did not suggest any trends (Figure. 4.6) the effects of increasing SMP on seedling survival was notable (Figure. 4.5). The seedling survival, generally was poorer when SMP increased with the control yielding 7 dead seedlings or 7% of the total sample (n=100). Although there was no significant differences between the 35, 25, and 15 kPa treatment means (LSD 0.05), the 25 kPa resulted in the fewest dead seedlings. The single dead seedling in the 15 kPa treatment may have been a result of a poor quality tree or improper planting, although this was not confirmed. The cost of A. fraseri planting stock, for a 2-0 123 seedling is approximately $0.70/tree and the labor associated with the planting these seedlings is approximately $0.17/tree (Nzokou et al., 2006). A 7% loss (Figure. 4.5, ctrl based on n=100) of plating stock could be a considerable expense for Michigan growers, and this expense is further emphasized as growers tend towards larger planting stock (2-2) in attempt to reduce harvest time and increase survival rate. Crop water stress index Little observable difference was seen on the effects of the various SMP irrigation treatment on the CWSI in 2006 (Figure. 4.7) and 2007 (Figure. 4.8). This in part may have been due to the fact that subject trees, were median sized trees and since changes in growth were not impacted on trees >0.9 m, water stress may not have been as much of a factor on these larger specimen trees than their smaller counterparts. CWSI values did follow increasing and decreasing trends in 2006 and 2007, due to the differences in TC-TA. Although a positive difference (canopy temperature greater than ambient) has been show to be an indicator of reduced transpiration and the onset of water stress (Ehrler, 1973) CWSI values did not show any discernable relationship to varying SMP irrigation treatments. Furthermore, CWSI studies typically involve annual crops (Jackson et al., 1977; Orta et al., 2003; Payero and lrmak, 2006; Payero et al., 2005) and A. fraseri being a perennial woody plant, may respond differently to water stress, making it difficult to quantify CWSI values among varying SMP irrigation treatments. Stem water potential 124 Stem water potential W519", values, for all tree sizes, followed similar trends, regardless of SMP irrigation treatment, with exception to the non-irrigated control which generally produced higher SWP values. The difference in SWP measured values may have been in part, due to the difference of transpiration between irrigation treatments (Melcher et al., 1998). It is likely that transpiration was greater among trees subjected to the lower SMP irrigation treatments. This maybe have been a possible explanation to the small trees subject to 15 kPa SMP treatment having higher SWP values than the control trees on various occasions (Figure. 4.9) similar to the results observed in other research (Melcher et al., 1998). These difference were less apparent in the medium (Figure. 4.10) and large trees (Figure. 4.11) suggesting that due to a more developed root system, water stress was lesser of an issue, agreeing with the results for changes in growth. Furthermore, figure 4.12 supports this idea of the larger trees exhibiting less water stress than smaller trees, indicated by lower overall SWP values. The similar SWP values observed post Julian day 197 could be due to seasonal variations (Nortes et al., 2005) possibly a result of physiological changes within the trees in preparation for winter dormancy. Although care was taken with the excised cuttings, this is another possible explanation for the lack of clear tends in the different size classes, since measurement errors can be of concern with improper handling (Turner and Long, 1980). Conclusion Our results showed that 15 kPa SMP irrigation treatment generally produced the most growth for trees less than 0.9 m in Horton, and in the 125 established stand in Sidney. These difference, typically, were more apparent in basal area change in comparison to change in height suggesting that stem growth is more affected by varying SMP irrigation levels than height growth. Similarly, lower SMP irrigation treatments results in higher seedling survival although decreasing SMP did not necessarily relate to an increase in height growth. Assessing water stress based on IRT measurements and CWSI calculations did not provide results consistent to what one would expect based on changes in tree growth with the various SMP irrigation treatments. Similarly, SWP results were also inconsistent with changes in growth, although results did show as tree size increased SWP decreased. Although water stress measurements (CWSI and SWP) did not produce results consistent with SMP irrigation treatments, all tree size classes were not taken into consideration. Based on the results obtained from this study, it is believed that a more extensive A. fraseri water stress study using CWSI and SWP, taking into account all tree sizes, would produce results consistent to the changes in growth associated with the various SMP irrigation treatments. Further research is needed to support this hypothesis. 126 Literature Cited Alves, I., and LS. Pereira. 2000. Non-water-stressed baselines for irrigation scheduling with infrared thermometers: a new approach. Irrigation science 19:101-106. ASCE. 2005. The ASCE standardized reference evapotranspiration equation. ASCE-EWEI Task Committee Report, January, 2005 Bower, C.A., B.A. Krathky, and N. lkeda. 1975. Growth of tomato on a tropical soil under plastic cover as influenced by irrigation practice and soil salinity. Journal of American Society of Horticultural Science 100:519-521. Cochard, H., S. Forestier, and T. Améglio. 2001. A new validation of the Scholander pressure chamber technique based on stem diameter variations. Journal of experimental botany 52:1361-1365. Ehrler, W.L. 1973. Cotton leaf temperatures as related to soil water depletion and meteorological factors. Agronomy Journal 65:404-409. Fereres, E., D.A. Goldhamer, and LR. Parsons. 2003. Irrigation water management of horticultural crops. Hortscience 38:1036-1042. Goldhamer, D.A., E. Fereres, and M. Salinas. 2003. Can almond trees directly dictate their irrigation needs? California Agriculture 57:138-144. Hinesley, LE, and SA. Derby. 2004. Growth of Fraser fir Christmas trees in response to annual shearing. Hortscience 39:1644-1646. Hoppula, KL, and T.J. Salo. 2007. Tensiometer-based irrigation scheduling in perennial strawberry cultivation. Irrigation Science 25:401-409. ldso, S.B., R.D. Jackson, P.J. Pinter, R.J. Reginato, and J.L. Hatfield. 1981. Normalizing the stress-degree-day parameter for environmental variability. Agricultural Meteorology 24:45-55. Intrigliolo, 0.8., and JR. Castel. 2004. Continuous measurement of plant and soil water status for irrigation scheduling plum. Irrigation science 23:93-102. lrmak, 8., DZ. Haman, and H. Bastug. 2000. Determination of crop water stress index for irrigation timing and yield estimation of corn. Agronomy Journal 92:1221-1227. Jackson, RD, R.J. Reginato, and SB. ldso. 1977. Wheat canopy temperature: A practical tool for evaluating water requirements. Water Resources Research 13:651-656. 127 Marshall, T.J., J.W. Holmes, and CW. Rose. 1996. Soil Physics, 3rd ed. Cambridge Univ. Press, Cambridge, UK. Melcher, P.J., F.C. Meinzer, D.E. Yount, G. Goldstein, and U. Zimmermann. 1998. Comparative measurement of xylem pressure in transpiring and non- transpiring leaves by means of the pressure chamber and the xylem pressure probe. Journal of experimental botany 49:1757-1760. Molz, F.J., and B. Klepper. 1973. On the mechanism of water-stress-induced stem deformation. Agronomy Journal 65. Mullins, CE. 2001. Matric Potential, p. 65-93, In K. A. Smith and C. E. Mullins, eds. Soil and Environmental Analysis. Marcel Dekker, Inc, New York, Basel. Naor, A., l. Klein, H. Hupert, Y. Greenblat, M. Peres, and A. Kaufman. 1999. Water stress and crop level interactions in relation to nectarine yield, fruit size distribution and water potentials. Journal of American Society of Horticultural Science 124:189-193. Nortes, PA, A. Perez-Pastor, G. Egea, W. Conejero, and R. Domingo. 2005. Comparison of changes in stem diameter and water potential values for detecting water stress in young almond trees. Agricultural Water Management 77:296-307. Nzokou, P., L.A. Leefers, and DE. Keathley. 2006. Costs and returns in Michigan Christmas tree production. MAES special report. Michigan State University, East Lansing, Michigan. Nzokou, P., N. Gooch, P. Nikiema, and B. Cregg. 2007. The “One-inch Rainfall per Week” Rule for Irrigation of Fraser fir: Assessing the rule using data collected at two tree farms in Michigan. Great Lakes Christmas Tree Journal 2:16-28. Orta, A.H., Y. Erdem, and T. Erdem. 2003. Crop water stress index for watermelon. Scientia Horticultu rae 98:121-130. Ortufio, M.F., J.J. Alarcén, E. Nicolas, and A. Torrecillas. 2004. Comparison of continuously recorded plant-based water stress indicators for young lemon trees. Plant and Soil 267:263-270. Payero, J.O., and S. lrmak. 2006. Variable upper and lower crop water stress index baselines for com and soybean. Irrigation science 25:21-32. Payero, J.O., C.M.U. Neale, and J.L. Wright. 2005. Non-water stress baselines for calculating crop water stress index (CWSI) for alfalfa and tall fescue grass. Transactions of the ASAE 48:653—661. 128 Shackel, K.A., H. Ahmadi, W. Biasi, R. Buchner, D.A. Goldhamer, S. Gurusinghe, J. Hasey, D. Kester, B. Krueger, B.B. Lampinen, G. McGourty, W. Micke, E. Mitcham, B. Olsen, K. Pelletrau, H. Philips, D. Ramos, L. Scheankl, S. Sibbert, R. Snyder, S. Southwick, M. Stevenson, M. Thorpe, S. Weinbaum, and J. Yeager. 1997. Plant water status as an index of irrigation need in deciduous fruit trees. HortTechnology 7:23-29. Smajstrla, AG, and SJ. Locascio. 1996. Tensiometer-controlled, drip-irrigation scheduling of tomato. Applied Engineering in Agriculture 12(3):315-319. Thompson, R.B., M. Gallardo, L.C. Valdez, and MD. Fernandez. 2007. Using plant water status to define threshold values for irrigation management of vegetable crops using soil moisture sensors. Agricultural Water Management 88:147-158. Turner, NC, and M.J. Long. 1980. Errors arising from rapid water loss in the measurements of leaf water potential by the pressure chamber technique. Aust. J. Plant Physiol 7:527-537. Wang, 0., W. Klassen, Y. Li, M. Codallo, and AA. Abdul-Baki. 2005. Influence of cover crops and irrigation rates on tomato yields and quality in a subtropical region. Hortscience 40:2125-2131 . 129 Table 4.1. Mean relative height growth and standard error (cm/cm) for all size classes and treatments in 2006 and 2007 for Horton. Similar letters indicate no significance between treatments means (LSD 0.05). Size class 2006 2007 Treatment (m) Mean SE Mean SE Control 0.164 0.024 0.302 0.054 A 45 kPa 0.172 0.023 0.246 0.031 (<0.6) 35 kPa 0.196 0.025 0.283 0.033 25 kPa 0.197 0.024 0.325 0.050 15 kPa 0.117 0.027 0.207 0.060 Control 0.205 a 0.019 0.281 0.031 B 45 kPa 0.214 a 0.032 0.283 0.032 (0.6 _ 0.9) 35 kPa 0.241 a 0.048 0.265 0.027 25 kPa 0.288 a 0.033 0.293 0.037 15 kPa 0.436 b 0.068 0.334 0.021 Control 0.346 0.026 0.261 0.029 C 45 kPa 0.293 0.024 0.274 0.028 (0 9 _ 1 2) 35 kPa 0.303 0.023 0.263 0.023 ' ' 25 kPa 0.250 0.019 0.248 0.019 15 kPa 0.275 0.023 0.282 0.025 Control 0.275 0.024 0.207 0.015 D 45 kPa 0.294 0.016 0.242 0.025 (1.2 _ 1.5) 35 kPa 0.276 0.024 0.288 0.024 25 kPa 0.278 0.043 0.229 0.027 15 kPa 0.282 0.021 0.224 0.019 Control 0.265 0.018 0.181 0.020 E 45 kPa 0.280 0.015 0.147 0.014 (15 _1 8) 35 kPa 0.199 0.011 0.185 0.019 ' 25 kPa 0.220 0.017 0.186 0.026 15 kPa 0.241 0.016 0.212 0.027 Control 0.237 0.013 0.138 0.017 F 45 kPa 0.246 0.015 0.157 0.019 (1.8 _ 2.0) 35 kPa 0.200 0.013 0.104 0.011 25 kPa 0.171 0.011 0.159 0.023 15 kPa 0.182 0.012 0.158 0.015 130 Table 4.2. Mean change in relative basal area and standard error (mmzlmmz) for all size classes and treatments in 2006, 2007, and total change in basal area for the 2 years at Horton. Similar letters indicate no significance between treatments means (LSD 0.05). Size class Treatment 2006 2007 2 Year (m) Mean SE Mean SE Mean SE Control 0.443 0.078 0.353 c 0.074 1.461 bd 0.236 A 45 kPa 0.666 0.172 0.335c 0.065 1.435 bc 0.171 (<05) 35 kPa 0.507 0.067 0.500 c 0.044 1.718 cd 0.146 25 kPa 0.632 0.081 0.755 b 0.100 2.850 a 0.383 15 kPa 0.365 0.091 0.808 a 0.192 2.045 a 0.667 Control 0.797 0.265 0.338 b 0.051 1.650 c 0.181 B 45 kPa 0.735 0.085 0.367 b 0.053 1.900 bc 0.184 (0.6-0.9) 35 kPa 0.675 0.086 0.433b 0.076 1.864 bc 0.206 25 kPa 0.847 0.092 0.685a 0.060 2.819 a 0.284 15 kPa 1.099 0.295 0.689 a 0.107 2.797 a 0.339 Control 1.135 0.476 0.313 0.042 1.925 0.212 C 45 kPa 0.728 0.152 0.281 0.026 1.548 0.120 (0.942) 35 kPa 0.691 0.071 0.323 0.030 1.650 0.137 25 kPa 0.577 0.061 0.327 0.037 1.756 0.161 15 kPa 0.679 0.057 0.569 0.176 1.840 0.132 Control 0.409 0.046 0.217 0.029 1.001 0.100 D 45 kPa 0.703 0.175 0.300 0.034 1.627 0.314 (124.5) 35 kPa 0.611 0.072 0.299 0.026 1.492 0.137 25 kPa 0.626 0.069 0.265 0.040 1.346 0.129 15 kPa 0.529 0.077 0.345 0.037 1.422 0.118 Control 0.507 0.103 0.198 0.022 1.045 0.110 E 45 kPa 0.453 0.038 0.148 0.028 0.927 0.061 (1.5-1.8) 35 kPa 0.337 0.031 0.245 0.032 0.977 0.048 25 kPa 0.430 0.036 0.222 0.041 1.060 0.107 15 kPa 0.475 0.047 0.250 0.030 1.100 0.087 Control 0.289 0.035 0.165 0.026 0.702 0.074 F 45 kPa 0.407 0.054 0.133 0.019 0.717 0.064 (1.8-2.0) 35 kPa 0.281 0.042 0.180 0.025 0.697 0.062 25 kPa 0.253 0.026 0.163 0.022 0.675 0.053 15 kPa 0.290 0.035 0.173 0.016 0.766 0.078 131 Table 4.3. Significance of treatments indicated by ANOVA F-test p-values of various growth measurements for different size classes for 2006 and 2007 at Horton research location. Size class (m) <0.6 0.6 - 0.9 0.9 - 1.2 1.2 - 1.5 1.5 -1.8 1.8 - 2.1 Height 2006 0.224 0.002 0.415 0.988 0.560 0.319 Height 2007 0.433 0.571 0.896 0.128 0.347 0.137 Area 2006 0.317 0.557 0.473 0.283 0.290 0.073 Area 2007 <0.001 0.001 0.116 0.096 0.140 0.631 Area 2 year 0.005 0.001 0.343 0.146 0.620 0.912 Measurement 132 Table 4.4. Mean separation (LSD=0.05) for stem water potential (bar) where significance (P<0.05) was observed for small (top), medium (middle), and large (bottom) size class trees for each of the matric potential irrigation treatments. Means with similar letters are not significantly different (P<0.05). Small Trees Julian Day Treatment 155 165 169 172 176 179 204 Control 7.8 b 14.6 c 14.3 c 14.3 ac 6.2 c 14.7 b 14.7 b 45 kPa 7.7 b 12.6 ac 12.8 ac 13.7 ac 6.8 c 13.7 bc 20.5 bc 35 kPa 7.9 b 15.4 c 7.7 b 13.0 bc 9.7 b 14.1 b 14.1 b 24 kPa 7.1 b 12.8 ab 9.2 b 12.0 b 11.2 a 11.4 a 11.4 a 15 kPa 11 a 13.6 abi 11.8a 14.8a 11.9a 12.3 ac 8.7a Medium Trees Julian Day Treatment 165 169 172 176 204 Control 13.7 b 13.7 c 13.0 b 6.6 c 14.3 b 45 kPa 12.8 ab 12.8 c 13.4 ab 6.3 c 12.7 ac 35 kPa 12.9 ab 6.8 b 13.2 b 8.5 be 13.4 bc 24 kPa 11.5 a 8.5 ab 12.3 b 9.0 ab 11.0 a 15 kPa 13.2b 9.7a 14.9a 11.0a 11.4a Large Trees Julian Day Treatment 150 162 169 1 76 204 21 1 Control 9.9 ac 12.6 b 13.7 c 6.0 b 13.0 b 6.8 b 45 kPa 9.9 ac 11.0 be 12.2 c 5.8 b 9.6 ac 8.1 b 35 kPa 11.2 be 11.5 bc 7.0 b 6.9 b 9.1 c 6.6 b 24 kPa 13.2 b 10.3 ac 7.7 ab 9.4 a 8.3 c 11.3 a 15 kPa 11.4 bc 10.3 ac 9.1a 9.8a 11.7 ab 11.1 a 133 80 60 _ precipitation (A) Q"? —o—ctrl 7 50 A 92; 50 7 +45 kPa E E ---A---35kPa ” ”40 z 25 9:3 40 . ——a—25 kPa 1 30 3% 0 mm. 15 kPa 1 .- .- Q- ,1 1. .9- :g r " 1‘ 2. -. _ 20 E (U 20 ‘ 1. 1.3° ‘ ... Q_ 2 H r: 1I ,, [12:1 7:53“PI - '5' " 10 ’ '2' a J“ '1 [J _ I C- ' L. ”.4 . . 0 h “In! 1 1 1 ‘Ilmlllfl‘lfl 0 179 189 199 209 219 229 239 249 259 269 Julian day 80 60 _ precipitation (B) A —o—ctrl ° - 50 (U o A g 60 — +45 kPa E : ---&--35kPa . “405, .59 0 ° c E +25 kPa ' .9 .940“ --e--15kPa "3033; 8 1.5. 8 . -2° (U " O. 2 z. 1‘1. z _ 1O 1 1 O 182 Julian day Figure 4.1. Daily precipitation and soil matric potential at 30 cm depth for Horton in 2006 (A) and 2007 (B). 134 7O 80 _ Precipitation (A) 75‘ —o—ctrl “ 60 A 3" 50 ‘ —o—25 kPa _ 50 E E - - O - - 15 kPa V *5 4o _ :3 «5 .3 °- .9- .g 8 0 l- 188 1 98 208 218 228 238 248 258 268 Julian day 80 7O _ Precipitation (3) a +Ctrl .' . F 60 o. 60 4 +45 kPa _ E E“: --¢--35kPa 50E, 5% +25 kPa — 40 g C ... 9340* --O--15kPa ‘ .5 a - . - 30 :5; .0 a, ‘ f 4, _ o *3 20 5 ', .' {A's} 20 (9;) 5 " 4&1 410 0 _ l‘fl' 0 137 147 157 167 177 187 197 207 217 227 237 Julianday Figure 4.2. Daily precipitation and soil matric potential at 30 cm depth for Sidney in 2006 (A) and 2007 (B). 135 F’ V 0 6 5 Control 5 I25 kPa ‘~(15 g E: 15 kPa E 0.4 1% a 0.3 g, 0.2 Q) I 0.1 O 2006 2007* Figure 4.3. Mean relative height growth (cm/cm) in 2006 and 2007 for Sidney. Similar letters indicate no significance between treatments means (LSD 0.05). *' **' ***Significance P < 0.01, 0.001, or 0.0001 136 00 5 Control 25 kPa [1315 kPa 7‘ N ammo: .0 <11 Basal Area Change mm2/mm2 O 2006 2007" 2 Year*** Figure 4.4. Mean change in relative basal area (mmzlmmz) for 2006, 2007, and 2 year total change for Sidney. Similar letters indicate no significance between treatments means (LSD 0.05). *' **' ***Signiflcance P < 0.01, 0.001, or 0.0001 137 8 7 _ 5 6 i 5 5- a 4- % 3 ‘ o 2 ~ 1 _ 0 _ ctrl 45 kPa 35 kPa 25 kPa 15 kPa Treatment Figure 4.5. Seedling mortality for irrigation treatments for Sidney in 2007. Similar letters indicate no significance between treatments (LSD 0.05). 138 0.4 E ’g‘ : E 0.3 — 8, i g _ 9 0.2 j 0 _ ff» 0.1 — a, _ I 0 - . ctrl 45 kPa 35 kPa 25 kPa 15 kPa Treatments Figure 4.6. Seedling mean relative height growth (cm/cm) for Sidney in 2007. Similar letters indicate no significance between treatments means (LSD 0.05). 139 2 _ (A) . 1.6 4 -- 1.2 5 a) 3 ’ .- 0 0 3 _ o i '0 0 o. o O . .0 . . ' 0'0 O ' ' ' o 041. ‘. '0... o. ... . 0. O l l l l l l f l T T T l l Fl— 202 212 222 232 242 252 262 272 Julian day 2 _ (B) 1.6 5 o . a 1.2 4 E 0.8 — . 0 0 ' on. ° ' o- o ‘ O ’ 0 'b ' 0. fl 0. ' . ' . o o. 04‘- ’..o... 0 .0 .' ... '0 .o'. ' o 0 i i l T r l l r l 7 l i I I 202 212 222 232 242 252 262 272 Julian day Figure 4.7. Crop water stress index (CWSI) for control (A) and 15 kPa treatments (8) for Horton in 2006. 140 2.0 — 1.64 (A) a 1.2 ~ (3) 0.8 —‘ ‘ 0.4 - ‘l- 5 'q I If“... 0.0 Mfldfifl l T l .l l l 152 166 180 194 208 222 236 250 264 Julian day 2.0 5 B 1.6— ( ) (7) 1.2 - 3 08 .‘fi . ° 0 - 4 . .. C.‘ 0.4:. .n O ~‘W’A‘o :0“. 0.0 l l T T T l l 7 l —i l l l j— T l T 152 166 180 194 208 222 236 250 264 Julian day 2.0 — C 1.61 ( ) c'r3 1.2 — 50.8- . 0-4 1‘57.” W'o‘fiwm 0.0 r j l l l l l T l T T f T l l l l i 152 166 180 194 208 222 236 250 264 Julian day Figure 4.8. Crop water stress index (CWSI) for control (A), 25 kPa (B), and 15 kPa treatments (C) for Horton in 2007. 141 _x a) 14 - A 12 — g 10 4 E 8 - .03 (9’3 6 ‘ — a- 35 kPa 4 ‘ +25 kPa 2 ~ : 15 kP a * Significance (P<0.05) O T T T T l l T l T T T T l T T 150 158 165 172 179 186 200 207 Julian day Figure 4.9. Pressure bomb mean stem water potential (“133,") measurements (bar) for small height class trees (<0.9 m), subjected to various irrigation treatments, for Horton in 2007. 142 _s C) .- a a _x _x _x O N -h J 1 l \ —O— Control \I --D--45kPa ‘ WStem (bar) CD 6 _. 4 - - -A- 35 kPa +25 kPa 2 - * . . < . O --o-- 15 kPa Significance (P 0 O5) 150 158 165 172 179 186 200 207 Julian day Figure 4.10. Pressure bomb mean stem water potential (Ll-ism") measurements (bar) for medium height class trees (0.9-1.8 m), subjected to various irrigation treatments, for Horton in 2007. 143 _l O) 14~ 12~ 3105 E 8-1 .93 a) 6— 9 4_ --A—35kPa 2 , +25 ”’3 * Significance (P<0.05) --O--15kPa O W T T T T T T T T T T T T T T 150 158 155 172 179 186 200 207 Julian day Figure 4.11. Pressure bomb mean stem water potential (l-PStem) measurements (bar) for large (>1.8 m) height class trees, subjected to various irrigation treatments, for Horton in 2007. 144 ..L C) _\ .h 1 L"’Stem (bar) 00 + Small 3 6 a 4 - - a - - Medium - a - Large 2 ‘ * Significance (p<0.05) 0 T T T T T T T T T T T T T T T 150155158162165169172176179183186197200204207211 Julian Day Figure 4.12. Summary of mean pressure bomb stem water potential (tPStem) measurement (bar) for all treatments for tree height classes, small (<0.9 m), medium (0.9 — 1.8 m), and large (>1.8 m). 145 CONCLUSION Irrigation of Fraser fir (Abies frasen') is increasing in Michigan due to challenges posed by growing conditions, and there is need to develop and implement sustainable irrigation system for this commodity. This project investigated methods for efficient irrigation of A. fraseri. From experiences found during the design and implementation of our automated irrigation system, we concluded that it is imperative to understand site conditions and irrigation needs before delineating a farm into irrigation zones. Furthermore, change in site elevation, pump output, pipe size, and emitter spacing need to be taken into consideration to ensure the irrigation system operates efficiently and delivers uniform coverage throughout the field. Additionally, a method of assessing irrigation needs is required to make education decisions on irrigation scheduling. We also concluded that, although its initial cost is greater, an automated irrigation system results in an overall reduction in operating costs after 2 years of operation. This reduction is attributed to a reduction in labor costs compared to a manually controlled irrigation system. in order to develop a model for prediction changes in soil matric potential (SMP) based on environmental conditions, potential evapotranspiration (PET) losses were combined with precipitation inputs to develop an adjusted evapotranspiration (AET) parameter. The AET was related to SMP measured at a 30 and 60 cm depth. Results obtained showed that AET had an exponential 146 relation with SMP during extended periods of drought. However, following a large precipitation event, the relation was no longer valid. Our study also showed that the relationship between AET and SMP was linear on sandy soils when SMP was close to or above field capacity. Results also showed that irrigation positively impacted the growth of A. fraseri notably during the early stages of the rotation when trees were less than 0.9 m in height. Seedling survival was generally greater with increased irrigation. Results of crop water stress index (CWSI) and stem water potential (SWP) showed no difference between irrigated and non-irrigated treatments. Overall the results from this study provide a good base for developing irrigation scheduling guidelines for Fraser fir in Michigan. However, several key factors need to be further investigated. For example, there are many limitations assessing the effectiveness of irrigation solely on growth. Most irrigation events normally occur in July and August when the trees height growth is nearing completion for the season but other factor such as physiology and reproductive development are still ongoing and may be impacted by irrigation. Additional research is needed to examine the effects of irrigation on these factors and test the application of CWSI and SWP to assess the irrigation needs of A. fraseri. Understanding how the various SMP irrigation treatments relate to these water stress measurements could then direct the focus into developing a more general model for prediction SMP based on changes in environmental conditions. 147 M lljllljjljjl “j Till ljlljlllll 33 T T 0 956