AN ANALYSSS OF THE HYEDROLOGSC BUDGET IN
{RACIAL SAfiéDS U‘NQER PKNE AND HARDWOOD FORESTS

Yhmis for flu Degree of Ph. D.
MMHIGAN SYA'H UNWERSIYY
{Sean Hawaré Urie
196:5

     

 
 

LIBRAR Y
Micki State
‘ Unlsvfisity

This is to certify that the

thesis entitled

AN ANALYSIS OF THE HYDROLOGIC BUDGET IN GLACIAL SANDS
UNDER PINE AND HARDWOOD FORESTS

presented by

Dean Howard Urie

has been accepted towards fulfillment
of the requirements for

__Eh__D_.__ degree in W

C" .

\_ M ) /
. l ' w
(j, k. /(__ 1 “L (L; L / . Z‘ 7. (_ 4--

Major professor

 

 

Date December 17‘ 1964

0.169

 

 

74w?

 

 

 

 

ABSTRACT
AN ANALYSIS OF THE HYDROLOGIC BUDGuf
IN GLACIAL SANDS UNDER PINE AND HARDWOOD FORESTS

by Dean Howard Urie

A method of analysis was developed for determining the water
yield as net ground—water recharge and the evapotranspiration from areas
of homogeneous vegetation. The technique usei is an unconfined lysimeter
approach which is applicable where there is no surface runoff and the
water yield is reflected in changes in storage in watermtable aquifers.
The tests of the method were made on the Udell Experimental Forest of
the Lake States Forest Experiment Station (U.S. Forest Service) located
in the northwestern part of Michigan's Southern Peninsula.

The evaluation of changes in groundwwater storage required
separating the water-table fluctuations into recharge, seepage flow
recession and evapotranspiration components. The seepage flow recession
rate was predicted from the conformation of the water table in the
vicinity of each local study area. Positive and negative deviations
from the predicted recession rate due to seepage were attributed to
recharge or evaporation losses, respecti Fly. The individual components
were then weighted by the specific yield of the aquifer layer in which
the fluctuation occurred. The products of this operation were accumuW
lated to obtain the volumes of gross recharge, seepage flow, and
evapotranspiration from local groundmwater storage.

The studies were conducted in medium textured outwash plain
sand soils. Two categories of water—table levels were sampled, 15-18
feet and O~8 feet. In the shallower areas, a portion of the root zone

was saturated during the early part of the growing season. Comparisons

Dean Howard Urie

were made of the evaporation and water yield for a tw0nyear period

beneath: (1) a 34nyear—old jack pine (Pinus banksiana Lamb.) stand,

 

(2) a 20~year-old red pine (E. resinosa Ait.) plantation with an oak
(Quercus sp. L.) overstory and (3) mixed hardwoods of pole and sawtimber
size. The hardwoods were composed of upland oaks on a well-drained
(15m18 foot water table) area and lowland species consisting mainly of

red maple (Acer rubrum L.), white birch (Betula papyrifera Marsh.) and

 

 

American elm (Ulmus americana L.) on a poorlywdrained soil.

 

Net water yields were greatest under the deciduous forests where
the average for the two years was 15.3 inches. In the pine plantations,
the average water yield for the same period was only 12.4 inches. These
differences in net groundmwater recharge were caused by the greater
evapotranspiration losses in the conifers, 20.7 inches versus 17.1 inches
in the hardwoods. The annual patterns of recharge and evapotranspiration
showed these differences to occur while the hardwoods were dormant. The
water content of the snowpack was least under the densest conifer stands
and greatest under hardwoods. These snowpack differences were reflected
in the inputs of groundmwater recharge following snow melt.

The depth of the water table was inversely related to the ground»
water losses due to evaporation. Analysis of diurnal well level fluctuaw
tions showed that evaporation losses ceased in poorlywdrained soils when
ground~water was below 4.5 feet under hardwoods and below 5.5 feet under
jack pine. Evaporation losses under red pine were still evident, though
slight, when the water table was 8 feet below the mean land surface.
These differences in water~tab1e depth effects were partially explained

by corresponding differences in the rooting depths of the three species.

AN ANALYSIS OF THE HYDROLOGIC BUDGET

IN GLACIAL SANDS UNDER PINE AND HARDWOOD FORESTS

By

Dean Howard Urie

A THESIS

Submitted to
Michigan State Univers16y
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Forestry

1965

ACKNOWLEDGMENTS

The author wishes to acknowledge the assistance and encourage-
ment of the members of the Guidance Committee: Drs. A. E. Erickson,
W. J. Hinze, S. O. Serata, and D. P. White. To Sidney Weitzman and
John Arend of the Lake States Forest Experiment Station, and especially
to W. D. Striffler, formerly of this station and now with the Central
States Experiment Station, thanks are due for their patient help over
the past few years.

Special thanks are due to all the past and present co~workers
at the Cadillac Field Office who have helped with field measurements
and participated in seemingly unending discussions of the merits of

alternative approaches to the problems of analysis.

ii

VITA
Dean Howard Urie

Candidate for the Degree of
Doctor of Philosophy

Final Examination: December 17, 1964

Guidance Committee: J. L. Arend, A. E. Erickson, W. J. Hinze, S. O. Serata
and D. P. White (Major Professor)

Dissertation: An Analysis of the Hydrologic Budget in Glacial Sands
Under Pine and Hardwood Forests

Outline of Studies:
Major subjects: Forestry, Watershed Management
Minor subjects: Soil Science, Geology

Biographical Items:
Born February 12, 1929, Craftsbury, Vermont

Undergraduate Studies:
University of Vermont, 1947~1949
University of Michigan, 1949~1951
B.S. Forestry, 1951

Graduate Studies:
Oregon State University, l956w1958
M.S. Agriculture (Soils), 1959

Michigan State University, l961e1964
Ph.D. Forestry, 1965

Experience:
Forestry Research Technician, Department of Forestry, University
of Vermont (Burlington, Vermont), 1954~1956; Graduate Research
Assistant, Soils Department, Oregon State University (Corvallis,
Oregon), 1956~1958; Research Forester (Hydrologist), Lake States
Forest Experiment Station (Cadillac, Michigan), 1958 to date.

Member:
Society of American Foresters
Soil Science Society of America
National Waterwell Association (Technical Division)
Sigma Xi
Xi Sigma Pi
Michigan Association of Conservation Ecologists
Michigan Academy of Science, Arts, and Letters

iii

TABLE OF CONTENTS

Chapter Page
I. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . 1
II. OBJECTIVES OF THE STUDY . . . , . . . . . . . . . . . . 5
III. REVIEW OF LITERATURE . . . . . . . . . . . . . . . . . . 7

Estimation of Groundeafer Buigets Using

Surface Runoff MeasUremen‘s . . . . . . . . . . . . 7
Estimation of Ground-Water Budgets Where

Surface Runoff Data is Not Applicable . . . . . . . 9

IV. DESCRIPTI'N Of GENEFAL STVSY AREA . . . . . . . . . . . 11

Soils and Vegetation . . . . . .
Distribution of V getafion Types . . . . . . . . . . . 16
Drainage Basins and Waterriaile Slopes . . . . . . . . 18

V. DESCRIPTION OF LOCAL SJLDI'ArhAS . . . . . . . . . . . . 20
VI. STUDYME’I‘HOIB.......‘.............. 24

Delineation of Basin Boundaries . . . . . . . . . . . 24
Instrumentation of Legal Study Areas . . . . . . . . . 24
Derivation of Preiitfej Water Table

Recession Rates . . . . . . . . . . . . . . . . . . 27
Separation of Compunrr‘a of Wa‘er~Table

Fluctuations . . . . . . . . . . . . . . . . . . . . 32
Determination of Speéific YiVId for

Aquifer Layers . . . . . . . . . . .
Computation of Daily Ground WE‘FI budgets . . . . . . 38
Evaluation of SnOavVil‘ Re.h»‘ge . . . . . . . . . . . 40

VII. RESULTS . . . . . . . . . . . . . , . . . . . . . . . . 43

Gross Ground Water R pharge . . . . . . . . . . . . . 46
Evapotranspiration from Gro Hi Wa‘er Supplies . . . . 55
Net Ground~Water Recharge . . . . . . . . . . . . . . 59
Snowpack Accumulation ani S ring Recharge . . . . . . 61
Evapotranspiration , . . . . . . . . .
Ground~Water Yield to seepage Flow . . . . . . . . . . 78

VIII. DISCUSSION OF WATER BFDOET RES LIS . . . . . . . . . . . 83
IX. SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . 89

LITERATURE CITED . . . . . . . . . . . . . , . . , . . . . . . . 95

LIST OF TABLES

Table Page
1. Stand conditions on selected study areas . . . . . . . . . 22
2. Specific yield values of subsoil layers . . . . . . . . . 37
3. An example of ground-water budget computations . . . . . . 39
4. Mean monthly precipitation, Udell Experimental

Forest and monthly totals on local study areas . . . . . . 44
5. Theoretical water balance by months . . . . . . . . . . . 45
6. Gross ground-water recharge (area inches) . . . . . . . . 47
7. Evapotranspiration drain from groundwwater in

shallow waterutable areas (area inches) . . . . . . . . . 48
8. Net ground-water recharge (area inches) . . . . . . . . . 60
9. Winter and snow~melt recharge, maximum snowpack and

winter precipitation by cover type in the shallow

water-table zone . . . . . . . . . . . . . . . . . . . . . 67
10. Total annual evapotranspiration by forest cover

type and water—table depth condition . . . . . . . . . . . 7O
11. Monthly hydrologic budgets for six local study

areas utilizing changes in soil moisture status
predicted by Thornthwaite’s formula, 1961-62
water year . . . . . . . . . . . . . . . . . . . . . . . . 71

12. Monthly hydrologic budgets for six local study
areas utilizing changes in soil moisture status
predicted by Thorthaaite’s formula, l962~63

water year . . . . . . . . . . . . . . . . . . . . . . . . 72
13. Monthly losses from aquifer storage to seepage
flow by cover type and depth . . . . . . . . . . . . . . . 79

Figure

10.

ll.

12.

1:5" or F gems

The location of the Udell Experimental Forest
in Michigan’s Southern Peninsula between the
Manistee and Little Manistee Rivers . . . . . . .

Topographic Map, Udell Hills, Manistee
County, Michigan . . . . . . . . . . . . . . . .

Profile of water table eleva'iona in relation
to land surface and surface wa‘er featureS.
Udell Experimental Forest ard vicini'y . . . . .

Forest cover types, Udell Experimental Forest
and vicinity . . . . . . . . . . . . . . . . . .

Topography of the wafer ‘able and approximate
groundeater basin boundarififi, Uiell Experimental
Forest and vicinity. July 26, 1963 , . . . . . .

Plot and instrument locations, hydrologic budget
study, Udell Experimental Forest . . . . . . . .

Six~inch diameter well equipped ai‘h water level
recorder, Oak (deep) study area. Udell Experiment

Forest . . . . . . . . . . . . . . . . . . . .

Onewfoot watermtable file? Finn (untoucs in the

vicinity of three local Stud} er 19, Tune 12, 196

and September 30, 1963 . . . . . . . . . . . . .

Daily waterstable rere-sizué‘v its Jue ‘o seepage
flow in relation ‘n fhr HI""tr«e :n ua'erhteble

slope above and belcv th+ eiv‘. -‘ll . . . . . .

Separation of the componenis of wafer~*able
level fluctuations, Hfirixmol (shalloW) study
a 116’ £1 0 1 v a I o c r O a 9 a V 9 I 0 e

O O 2 0 0

Specific yield of shallow aquifer layers for
six local study areas in nutmash sands . . . . .
Water-table level flurtuafiions during two«year
study period, mean of five sells in each

local study area . . . . . . . . .

vi

al

\\J

Page

12

14

17

19

21

26

28

31

34

36

41

Figure

13.

14.

16.

17.

18,

19,

21.

(x
[Q

23.

25.

Monthly precipitation, gross and net recharge
and evapotranspiration from the saturated
zone for six local study areas, 1961—62

water year

Monthly precipitation, gross and net recharge
and evapotranspiration from the saturated
zone for six local study areas, 1962-63

water year . . . . . . . .

Weekly increments of gross recharge in six
local study areas . . . . . . . . . . . . . .

Cumulative gross recharge in six local
study areas by water year

Root distribution in relation to depth in six
local study areas

Monthly evapotranspiration from the saturated
zone as a percentage of potential evapotranspira—
tion in relation to water~table depth

Average snOWpack water equivalent values
by forest cover type . . .

Cumulative ground—water recharge during
winter and snow=me1t periods in six local
study areas

Monthly evapotranspiration for six local
study areas

Daily evapotranspiration for three shallow
water—table study areas, average rates for
periods between ground-water recharge events

Cumulative evapotranspiration for three shallow
water-table areas based on computed evapotranspira-
tion between groundwwater recharge events

Monthly losses from aquifer storage to seepage
flow for six local study areas

Water-table elevation on eastern outwash
plain, Udell Experimental Forest, in
relation to cumulative departures from
normal precipitation

vii

Page

49

50

53

58

63

74

75

77

81

88

CHAPTER I
INTRODUCTION

One of the more complete, yet concise, descriptions of the
hydrologic cycle has been given by McGuinness (1963) who wrote of "that
endless circulation of water from the primary reservoir, the ocean, to
the atmosphere, the land, and back to the ocean over 25 beneath the
land surface." As ground-water is a part of this cycle, it must be in
motion, however slow.

The relative importance of this subsurface routing of water from
the land toward the sea is dependent on the infiltration capacity of
the soil surface. »Where these rates are high, a large portion of rain-
fall moves into and through the ground-water aquifers. In those areas
of the Lake States Region of the United States which are covered with
coarse textured glacial drift, the portion of streamflow which is
derived from such aquifers is the major part of the total flow. Here
the ground-water route forms the connecting link between the land and
the surface water features.

The portion of total precipitation which flows by surface or
underground routes from a land area is dependent on the amount of water
evaporated from that land and from the vegetation which is growing upon
it. This evaporation loss may occur when rain or snow is intercepted
by the surface parts of the vegetation or it may cycle through the soil
to the roots of the plants and be transpired. The nature of the vegeta—

tion, its surface area, its depth of rooting and its physiological

characteristics influence the amount of this evapotranspiration recycling
within the greater cycle. In forest land management, considerable study
has been given to the effects of various vegetative conditions on the
division of total precipitation into the evaporation and runoff cate-
gories. Traditionally these studies have relied on streamnflow measure—
ments to obtain the runoff quantity. In such high infiltration areas

as are found in much of the northern part of Michigan's Southern
Peninsula, the direct relationship between small areas of surface and
the flow of the widely spaced streams is so masked by the mosaic of
vegetative conditions that the influence of a single cover type is
obscure. In order to reduce the size of the surface area to be studied
to that represented by a definable cover type, it is necessary to
evaluate the water yield at a more immediate point. The ground-water
flow which represents this water yield occurs at that immediate point
and, despite the many difficulties involved in its determination, pro-
vides the only practical in EEEH method by which the water balance for
such an area can be resolved.

The deep mantle of coarse textured glacial drift which covers
the Southern Peninsula of Michigan represents a giant mound of saturated
unconsolidated sediments. Subsurface drainage from this mound seeps
into the streams along most of their course. Many of the large and
small lakes which dot the landscape are but surface emergences of the
ground-water body. Thus, the ground—water resource is the source of
streams, the continuation of lakes; the water supply of the region in
both its used and potentially useful forms.

Precipitation in the northern part of Michigan's Southern

Peninsula ranges from 28 to 32 inches per year. This rainfall maintains

the saturated layer at its normal level about which fluctuations occur
with seasonal and longer term variations in rainfall. Surface and sub-
surface flow constantly drain water from the saturated zone toward the
local base level of the Great Lakes. That portion of the total precipi-
tation which represents true ground-water recharge must replace this
drain to maintain the normal level. When vegetative demands for water
alter the evaporation vector of the hydrologic equation, the remaining
recharge portion must also be altered.

Such variables of vegetative cover as density, dormant season
interception, depth of rooting, periodicity of moisture utilization
and physiological adaptation to poorly drained soils may have measureable
effects on the amount of evaporation from the land surface. These same
factors are likely to be altered by the land management practices of
forestry. Planting conifers with deep root systems, dense winter crowns
and long transpiration periods in place of grasses or deciduous forests,
is likely to affect the water balance.

Information from studies in the Lake States and elsewhere in the
northeastern part of the United States has documented the smaller amounts
of snow accumulation in dense conifer stands (Weitzman and Bay, 1958;
D115 and Arend, 1956; Striffler, 1959; Hart, 1963). Soil moisture levels
have also been shown to remain at low levels under conifers for longer
periods than under hardwood forests (Urie, 1959). Water in the snOWpack
is a major part of the total annual ground-water recharge (McGuinness,
1941). When soil moisture deficits exist it is impossible for precipi-
tation to recharge the ground-water supply. These two facts point to
an expected alteration of the ground-water balance following the

establishment of conifer plantations. This is indirect evidence. Actual

measurements of volume of ground-water recharge and the attendant effects
on the hydrologic balance are needed.

This study has sought to evaluate the hydrologic budget under
coniferous and deciduous forest cover in a relatively uniform sand
aquifer situation where surface runoff is practically nonexistent. Under
this situation the ground-water balance is the hydrologic balance. These
forest types and watershed conditions are prevalent throughout central
and northern Michigan and Wisconsin. Accordingly, the results of this

study should have regional application.

CHAPTER II
OBJECTIVES OF THE STUDY

The present study represents the first comprehensive investi-
gation of the effects of forest cover and reforestation on hydrology
under northern Lower Michigan conditions. This study is a part of the
research program in forest watershed management problems which is being
conducted by the Lake States Forest Experiment Station.

The sample watershed area selected for intensive study is located
in southern Manistee County about ten miles east of the Lake Michigan
shoreline in the northwestern part of Michigan's Southern Peninsula
(Figure 1). This area, the Udell Experimental Forest, is representa—
tive of much of the publicly owned forest lands in Michigan and adjoining
states. It is comprised of 3800 acres of federal lands on the Huron-
Manistee National Forest which were set aside in 1960 for research in
forest-ground-water relations. The overall project is designed to
compare the water economy of the various forest conditions as they now
exist and to determine what changes are produced by the forest manage-
ment practices currently in use in the region.

The specific objectives of the study reported here were:

1. To develop methods for deriving the hydrologic budget

for local sectors within a broad ground-water basin,
using well data.

9. To use these methods for deriving the groundwwater

recharge, seepage flow losses, and evapotranspira-

tion losses from the history of water-table fluctua-
tions beneath three common forest cover types.

3. To determine the effect of ground—water depth on the
hydrologic budget under these three cover types.

5

Figure l.

The location of the Udell Experimental Forest in Michigan's

Southern Peninsula between the Manistee and Little Manistee
Rivers.

 

   
 

Manistee River

 

 

. Udell
Experimental
Forest

 

 

 

Udell L.o.

 

 

 

 

 

 

 

 

CHAPTER III

REVIEW OF LITERATURE

The estimation of the evaporation portion of the hydrologic
budget of ground-water basin has been attempted in two general ways.
Where a well defined basin is drained by a stream, the water yield of
the basin is computed from the discharge of this stream. Evapotranspira-
tion is then computed from the difference between total precipitation
and runoff over a period which begins and ends with equal water storage.

Where a surface stream is not available to provide a measure of
the water leaving the basin, an alternative approach is necessary in
which the changes in ground-water storage are determined directly from

well measurements.

Estimation of Ground-Water Budgets Using Surface Runoff Measurements

 

Ground-water levels measured in wells over the entire basin in
Pomperaug, Connecticut were related to the level of runoff during periods
when the stream-flow was assumed to be coming from seepage flow (Meinzer
and Stearns, 1929). This relationship changed during the growing season
when evaporation losses from vegetation reduced the portion of the
seepage flow which left the basin. The authors found only a minimal
measure of ground-water losses to evapotranspiration when they sub-
tracted the ground-water runoff in the summer from that of the winter
months, using periods when water-table levels were similar. Rasmussen

and Andreasen (1959) utilized the same type of instrumentation in

Maryland. They solved the equation:

P=R+ET+ASW+ASM+AGW

where P is precipitation, R is runoff, ET is evapotranspiration,‘ASW is
the change in surface water storage,IBSM is the change in soil moisture
and.AGW is the change in ground-water storage. Weekly solutions of this
equation for ET were plotted against calender weeks. A smoothed curve
through these plots was used to solve for a convergent storage co-
efficient for the aquifer materials. These convergent solutions for

the gravity yield of the sediments permitted a solution which was in
agreement with the theoretical curve of seasonal ET.

Olmsted and Hely (1962) applied an average gravity yield
determined from the volume of aquifer dewatered by measured amounts
of dormant season stream—flow to obtain measures of ground—water dis—
charge from summer well recession. The evapotranspiration loss in a
western Pennsylvania drainage basin was found to be about one-fourth
of the total ground-water discharge.

In Illinois, Schicht and Walton (1961) separated the gross ET
derived from P-R determinations into soil and ground—water losses. They
constructed rating curves relating groundwwater runoff to mean ground-
water stage for both dormant and growing seasons. The difference between
the two curves was attributed to ET from ground-water supplies. They
then computed the expected loss for the annual pattern of ground—water

stages to obtain a ground-water ET value.

Estimation of Ground-Water Budgets Where
Surface Runoff Data is Not Applicable

 

 

White (1932) used the diurnal fluctuations of the water table to
obtain a measure of ET in a closed basin in Utah. By computing a
recharge rate for the 24-hour period and adding on the net daily reduction
in storage, he computed daily and seasonal ET rates for various types of
vegetative cover.

Ferris (1959) has suggested a similar approach to that of White
for the evaluation of evaporation effects on Michigan ground-water
levels. Where the rate of water-table change without ET is predictable,
he showed that a measure of the evaporation loss can be derived from
the accumulated deviation of the actual water level from this predicted
curve.

Holstener-Jorgensen (1961) compared the seasonal drawdown of
ground-water levels in Danish clay soils. In an area where surface runw
off was negligible during the growing season, the amount of precipitation
required to restore water-table levels to their spring stage was con“
sidered equal to the total ET for the period between equal high waterw
table stages. By this method, the author was able to demonstrate dif-
ferences in moisture use between forest types and to show seasonal
patterns of moisture use.

The direct evaluation of evapotranspiration drawdown from the
accelerated rate of water-table drawdown during daylight hours assumes
that the area of uniform water-table depth and uniform vegetation with
even moisture use is sufficiently large so that the rate of drawdown is
similar over the area of the aquifer which contributes to groundwwater

flow beneath the study well.

10

Troxell (1936) has shown a graphical method for solving the
problem of changing ground—water inflow due to increased head differences
during periods of rapid ET drawdown under shallow water—table vegetation.
This requires an estimation of the changing head due to the inflow—outflow
balance in the absence of ET losses. The net ET loss is then calculated
from the accumulated difference between the predicted head change and
the total head change occurring when evapotranspiration is included.

This is, in effect, a separation of the subsurface flow balance which
occurs due to the actual history of head change and the net loss in head
due to ET. In the situation where no inflow increase occurs, the two
vectors of total head change during a measurement period are multiplied
by the drainable porosity or specific yield to obtain a volumetric
measure of the two categories of ground water loss.

In essence the two methods differ in that, where streamwflow
data is used an entire drainage basin is evaluated as a unit. Where
stream~flow data is lacking, or impossible to relate to a sufficiently
localized area, the changes in groundwwater storage which occur in a
localized segment of the aquifer are analyzed. As was indicated earlier,
the geologic condition in the deep drift areas of Michigan necessitate

the use of the latter method.

CHAPTER IV

DESCRIPTION OF GENERAL STUDY AREA

The Udell Experimental Forest includes 3800 acres of National
Forest land in the southeastern part of Manistee County, Michigan. The
Forest includes parts of two townships, Township 21 North, Range 14 West,
and Township 21 North, Range 15 West (Figure 2). The experimental area
includes large portions of an isolated moraine. This upland feature is
approximately two miles in diameter and is characterized by a broken
ring of sandy ridges along the northwest and western portion as well as
along the southeastern edge.

Elevations range from 700 feet above sea level at the lowest
points on the western outwash plains to 1030 feet at the highest ridge—
tops. Most of the hillsides have gentle slopes between five and 15 perm
cent. A few short slopes exceed 25 percent.

The upland portion is situated on a broad outwash plain which
slopes gently westward, except as dissected by the major west flowing
rivers. The Manistee River, with a water surface at about 600 feet
elevation, flows about three miles north of the Udell Forest. The
Little Manistee River borders the south edge of the morainal feature at
a distance of only one—half mile. Pine Creek, tributary to the Manistee
River, flows within one-fourth mile of the northeast corner of the
research area. The only other perennial stream, Claybank Creek, arises
from an extensive swamp area along the north border of the Udell Hills.

A concentration of surface flow in the extensive swamp area in

the southeastern portion of the experimental area produces a discharge

ll

Figure 2. Topographic Map, Udell Hills, Manistee County, Michigan.

12

..0m ””55. mamezoo &%V
as). oifimooaoe Amarafllg a

m4 <
/\/
.:o__2 .Cznoo H.524: 8... coo
{Us
mji jug:
“We“
a , (we?
91
rev
9 VN

.>> E .m >> 9.x

   

 

 

 

 

.2_NH

 

 

13

of two—three cubic feet per second during periods of high ground—water
levels. This runoff flows westward for three—fourths of a mile along
the ditch associated with a raised woods road.

During early periods of agricultural use, a drainage ditch was
constructed to conduct high water from the vicinity of Timmerman Lake
north-northeast to Pine Creek. The effects of this ditch on the shape
of the ground—water surface along the east border of the study area are
discussed below.

All of these surface flows are at, or along, the boundaries of
the experimental area. The only surface waters which occur within the
heart of the study area are small bogs which contain surface water in
the spring. Observations over the entire study area during the four
years in which instruments have been installed, have failed to detect
any surface runoff from the highly permeable soils.

The depth of the sand mantle in the ridge areas is shown in two
vertical profiles (Figure 3). The depths shown by well logs are con-
nected by straight lines. The moraine is shown to be predominantly sand
with till clay and sandy clay underlying the western and northern ridges
at depths of 200 feet or more. Four wells, located west of transect
B-B; indicate a higher clay lens in the center of the western ridge at
about 790 feet above sea level. The presence of numerous small bogs in
depressions in the center bowl between the surrounding ridges suggests
that slowly permeable layers are present near the surface. Well G-56
penetrates such a layer at a depth of ten feet. This layer of sandy
clay is underlain by permeable sands from 17 feet downward.

A partially penetrating well at the Manistee Ski Area, located

near the base of the eastern face of the moraine, showed clay layers

Figure 3. Profile of water-table elevations in relation to land surface
and surface water features, Udell Experimental Forest and
vicinity.

l4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m
.m luv... . a

8832.. {to

.3 9.... .5226
I. a
. m
\ < P K

n a 4. \\\ 5;:

'nI"- Ille"- -|-ll'- .I'III '-""" """"""" .:.m=£

. . . 0.0 a on J 3.3 waddh «med; .9 2:...

/ a“... 3 as...“
- .00m
.801—
ceEBEEF 00
O . >>\.v .
\/ x
s \l/&
\ o) ,
\s l\1\ Ev
\ new“), \
l\l.o /

 

 

 

 

. / films: _

J9 a” 3:65 .3 340m 4323on

 

 

 

 

 

 

 

15

which extend between elevations of 710 feet to 620 feet. These clays are
underlain by permeable sand aquifers. A partially penetrating well at
the east one-fourth corner of Section 18, T21N, R15W, shows permeable
sands to a depth of 52 feet. Below this depth, alternating thin zones

of reddish till materials interrupted the sand to a depth of 120 feet.
Indirect evidence from seismic and resistivity surveys-i/suggests that,
under the outwash plain on the east side of the uplands, the saturated
sand aquifer is at least 100 feet thick with no continuous slowly perme-

able layers above this depth.

Soils and Vegetation

 

Most of the soils of the Udell Experimental Forest, on both
morainal and outwash areas, are formed from medium sand parent materials.
The major portion of these soils is within the limits of the Grayling
sand series. The soils range from regosols, with minimal profile
development, to incipient podzols having a light brown colored B with
no visible structure. In interior valleys, a higher content of silt in
the surface horizon results in visible improvement of the site quality
even though the textural change is insufficient to alter the soil type
classification.

Imperfectly drained soils cover approximately ten percent of the
land area. These soils, in which the water table is within the developed

solum during a part of the year, are a complex of Saugatuck and AuGres

 

‘l/Hinze, W. J'!.§E'.§l' 1964. A geophysical investigation of
hydrogeologic characteristics of the Udell Hills area, Manistee County,
Michigan. Unpublished file report. Dept. of Geology, Michigan State
University, East Lansing, Michigan.

lb

loamy sands. The Saugatuck profile is characterized by a well~developei
ortstein layer in the lower B horizon. The AuGres soils, which are
irregularly intermixed, have strong cementation in the B horizon but
have no continuous cemented layer. These imperfectly drained soils are
located along boundary strips near swamp areas of Rifle peat of varying
thickness over mineral soil. Soils of the Maumee series occur where the
organic matter layers are only 6 to 12 inches deep over gray mottled
sands.

The water table in these poorly drained soils varies from the
surface to 24 inches below the surface during the growing season. Followe
ing the cessation of moisture use by vegetation, the water table returns
quickly to near the surface where it remains at a relatively high level

until spring.

Distribution of Vegetation Types

‘. ”A———-

 

Native vegetation follows the pattern of the soil differences
(Figure 4). On the Grayling sands of the uplands and wellwdraintd HUIr
wash plains, the northern pin oak type (Society of American Foresters,
1954) is the major forest cover. In this type northern pin oak (Querggi

ellipsoidallis E. J. Hill), white oak (Q. alba L.), northern rei oak

 

(Q. borealis Michx. f.), and black oak (Q. velutina Lam.) are the

principal species. Bigtooth aspen (Populus grandidentata Michx.) and

 

quaking aspen (P. tremuloides Michx.), which form a small component of

 

these stands on the exposed ridges and flat slopes, become the dominant
species in the sheltered valleys and on the lower slopes. Red maple

(Acer rubrum L.) and paper birch (Betula papyrifera Marsh.) are prevalent

 

 

where the water table is high enough for groundmwater to suppliment the

Figure 4.

Forest cover types, Udell Experimental Forest and vicinity.

17

V\A.. ."

 

 

 

 

 

 

 

 

rum ‘4
:3. .3. .8. L gown... Ema :OJA 3.0.0.“.
~: 0* u = . a

. X . Eam—uOEI: mmo :0 wt..0 muo
é.“ hOOQI. mmn .uO-VIOI£

 

.3755 1.3%

 

 

 

 

 

 

 

 

1558 .1:
“cum .£

 

 

 

nouns—U unmm 9.3m

05m 0.53 .0

05m to: -w

oil moi. um

ans—m u:£304-m4
5:32??? am
mvookmzam 95.25-34
nvoozéua: Eva—£02 -2
M30. v—

425 Bank -:on_m< .<
mom; $230
Dzmomx—

 

 

 

    

 

:.o(.:ux :WN.

.:u<

fl
a
20x 2. ( :.u<

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

   

52:22
5:500 033—32
32 a: 2 722.

Beach 13352 uoumEaZ

amazon iezmzimmxm jun: #

 

 

 

 

 

 

 

 

 

 

18

soil moisture supply. In the forested swamps American elm (Ulmus

 

americana L.), black ash (Fraxinus nigra Marsh.), red maple, northern

white cedar (Thuja occidentalis L.) and eastern hemlock (Tsuga canadensis

 

 

L.) are the major species. Eastern white pine (Pinus strobus L.) and

 

red pine (P. resinosa Ait.) were formerly important constituents as is
evidenced by the frequency of stumps on uncleared lands. These species
now occur only as occasional stems or in isolated small groves. Jack
pine (3'.2325§1323 Lamb.) in natural stands is locally prevalent,
especially on the outwash plains.

Pine plantations of jack, red and white pine, established since
1934, (over approximately 1100 acres or nearly onerthird of the experim
mental area. The oldest plantations were installed on cleared lands.
Plantings since 1940 have had varying degrees of hardwood overstory.
Timber stand improvement operations in 1955 and 1956 resulted in a
partial release of the underplanted pine. Approximately 200 acres of

the red pine plantation still has a considerable degree of hardwood overr

story.

Drainage Basins and Wateerable Slopes

 

The Udell Experimental Forest is situated on the ground-water
divide between the Manistee and Little Manistee Rivers. Drainage to the
surface flow outlets of the Pine Creek and Claybank Creek produces a con«
formation of the waterutable surface which divides the north side of the
area into two subnbasins. Waterntable contours on the south side of the
area are less affected by surface drainage than by the relative permeabil~

ity of the saturated layer (Figure 5). Profiles of surface elevations
and groundeater levels show a generally high watermtable level in the

upland portion with evidence of perched water tables in the interior basin.

Figure 5. Topography of the water table and approximate ground-water
basin boundaries, Udell Experimental Forest and vicinity,
July 26, 1963.

19

CHAPTER V
DESCRIPTION OF LOCAL STUDY AREAS

Six study areas were selected on outwash plain sites for intenw
sive measurements of the ground-water balance under three representative
forest cover conditions (Figure 6). The study areas selected were
typical of many pine plantations, pine plantations with a hardwood over-
story and mixed hardwood forests characteristic of the site conditions
represented. Under each forest type, an area was selected where ground~
water was within the rooting zone during at least a portion of the grOW~
ing season. As closely as possible, the same forest cover conditions were
replicated in deeper water-table areas where the saturated zone was well
below depths at which its water would be available for transpiration.

Stand conditions on the six study areas are summarized in
Table l. A 100~percent tally of all trees over 2.5 inches in diameter
was obtained on 3 one-acre plots in each location. Crown density
measurements were made at 25 systematically selected points in cach
area using a spherical densiometer (Lemmon, 1956),

The jack pine plantations, which were established in 1934 at a
spacing of six by six feet, have most trees and the densest basal area
stocking per acre. Crown closure is uniform in comparison to that of
the red pine plantations with an oak overstory, as is shown by the lower
variance of the individual density observations. This uniformity is

even greater by comparison during the dormant season when the overstory

oak provide effectively no crown cover.

20

Figure 6. Plot and instrument locations, hydrologic budget study, Udell
Experimental Forest.

21

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00:06 50m 0

 

 

 

 

f

 

 

mZO_._.<oO.._ FZMZDmEbZ. 024. ._.On_n_

8003 m2» R v a n s... \_\\H/t,z,_ N t:
to UX‘J 9 \ \I // \\\\\ \ n / III/I \ /
mm «m _n W \‘wx \ u / \x /
.‘ e o\\\ - I’ll
m he: 2.53%mequ ............. fi-uanl.n!-n-fiuun\.nu* \/
,. / \x/z / m \ \X
/{ \ Ill/l/LII‘\\\\\\ /l\\
b. 8:. 32509.... .< :N 111/ __ \\\\
om aw on _ 2.. one .55 _ .33 3E 5:.
:0) 0533.: gm 0
Q
9.4.. 9.4.. mzo_._.<uo._ 44w; “.0 n__<._.wo
ma... z<zcg£v . F
l\.\ 233 a
1E .82.
32.25. g A
2E .32. 0
N ON _ 9 0. cu m N NN .N
32.26 . fie -
2.: .5. 2.: 3m _
as. n , w a... e
as _ i
o. n. 1 $12 2
826.8.
Q N- _- O.

 

 

 

 

TABLE 1

STAND CONDITIONS ON SELECTED STUDY AREAS

 

1/
2/ Crown Crown Variance-
Trees Basal Mean Density Density of
Per AEea Diameter Summer Winter Density
Study Area Acre (Ft /A) (in) (%) (%) (%)
2/
Jack Pine(S) 766 5/ 94.7 4.59 81.8 fso 8.1
(729)- (88.5) (4 60)
4/
Jack Pine(D) 809 103.4 4.56 84.6 t83 5.4
(688) (85.3) (4.66)
Red Pine (S) 708 86.5 4.49 80.6 f70 14.2
(580) (59.7) (4.25)
Red Pine (D) 578 74.8 4.39 70.0 f60 18.4
(458) (40.5) (3.98)
No. Hdwds(S) 322 72.7 5.88 83.8 £10 12.6
Oak (D) 461 78.6 5.11 81.3 iio 6.8

Variance of 25 systematically located crown density measurements.
— Trees 2.5 inches at DBH and larger.
_ Shallow water-table area.
— Deep waterwtable area.

- Figures in parentheses refer to pine only.

22

The lowland hardwood area has the fewest number of trees, the
largest average diameter and a high variability in crown density during

the growing season.

23

CHAPTER VI

STUDY METHODS

Delineation of Basin Boundaries

 

The initial investigations of the groundwwater situation in the
outwash plain portions of the Udell Experimental Forest were directed
toward defining the boundaries of the ground-water basins. Under a
separate study, 56 wells were installed at one-half mile intervals
along a rectangular grid system. A leveling survey related datum
elevations at each well site to established U.S. Geological Survey
bench marks. From water-table measurements taken over this grid, it
has been possible to locate the perimeter of ground-water basins and to
define the approximate surface topography of the saturated layer throughw

out the year and over the period of the present study (Figure 5).

Instrumentation of Local Study Areas

 

Three study plots were located in each of the three forest cover
type water-table depth conditions. Each five—chain (330 foot) square
plot was located in an area of uniform topography and characteristic
forest stand conditions. In most instances, it was possible to also
place these plots adjacent to one another (Figure 6). The jack pine
areas were chosen near the water—table divide between the Manistee and
the Little Manistee Rivers. The red pine areas with the oak overstory
were nearer to the Pine Creek seepage line. Both the hardwood study
areas were located on the northwest lower slope of the uplands section.

24

25

The six study areas were assigned abbreviated cover-depth labels
which will be utilized throughout this report. These labels were: Jack
Pine (deep) and (shallow), Red Pine (deep) and (shallow), Oak (deep) and
Hardwood (shallow). The general term "Hardwood" was used in this instance
because of the mixture of such species as red maple, paper birch, American
elm, bigtooth aspen, and northern red oak on the imperfectly drained soils
of this local study area.

A one-acre circular plot was delineated around each of the three
plot centers. This central plot wasused for measurements of stand
characteristics. A partially penetrating well was installed at the
center of each plot. The key well for each local study area was located
in the central plot. This was a six—inch diameter well equipped with a
water level recorder (Figure 7). All other wells were constructed with
1 1/4 inch galvanized steel pipe attached to a sand drive well point.

Datum elevations for each well, of both types, were established
and checked against the datum of the one-half mile grid well network.
Water levels in all non-recording wells on the local study areas were
measured at weekly intervals. The key recording wells were also measured
manually at this same frequency. The water level shown on the recorder
chart was corrected to the beginning and ending levels determined by
direct tape measurements for each weekly period. All water—table
depths were recorded to the nearest 0.01 foot.

A rain gage network has been in operation on the Udell Experi-
mental Forest since November, 1959. The gages which were used for the
present study are shown in Figure 6. Local precipitation amounts were

measured at Station 1 for the Oak (deep) and Hardwood (shallow) areas,

from Stations 14 and 16 for the two Jack Pine areas, and from Stations 15

Figure 7. Six-inch diameter well equipped with water level recorder,
Oak (deep) study area, Udell Experimental Forest.

26

‘___._3___..

.. 1131...: .

 

27

and 16 for the two Red Pine areas. A total of 12 raingaages were operated
throughout the tw0wyear study period on the entire experimental forest.
Snow measurements were taken with a Mt. Rose snow sampling tube
along five-point snow courses under timber stand and slope conditions
representative of each raingauge station. Supplementary snow sampling
was carried out on the local study areas.
Data from the tWOeyear period beginning October 1, 1961, and

ending September 30, 1963, were used in the present study.

Derivation of Predicted Wa‘erefable ReceQSIOn Rates

 

The changes in contour sparing of the ground water surface are
shown in figure 8 for the makimum ani minimum stages occurring during
the tWOWyear study period. The June 13, 1962, conditions represent the
water-table surface at a time when the majori') of rh» grid wells were
at their highest levels. At this time of year the gradient of the water
table steepens rapidly with increasing diq*an e from The wa*er fable
divide. The lowest stages were reached at the end of the 150 year study
period. The maximum gradient of 19 feet per mile occu:a near the Pine
Creek seepage line. Near 1‘he (reef of ihe basin the gradient approaches
zero. The shape of the water table profile remains relatively unchanged.
the low stage profile representing a uniform 10wering of the ground-water
level of the entire slope. Minor changes in the gradient in the immediate
vicinity of individual wells were found to be related to the rate of
groundwwater subsidence during nonwrecharge periods. Even though these
gradient changes were small, the differences be‘ween the approximate

tangent to the wateretable surface up the flow line from a well and the

tangent below the well was related to the average daily recession rate.

Figure 8. Onevfoot wateretable level contours in the vicinity of three
local study areas, June 12: 1962, and September 30, 1963.

28

 

 

WATER TABLE ELEVATION \\

-- - - June l2, |962
Sept.30,l963

 

 

’/ x’ O
/ / am:
I Jack P'ne
// O O (Dacia) // O O O
I O O / I
h
4" O ’I O a
x / 1‘5

 

 

29

The change in storage in the vertical sub-section of the aquifer
represented by a well is equal to the difference between the rate of
ground-water inflow and outflow during periods of negligible recharge
or evaporation loss. The flow through a vertical section of the aquifer

may‘be-expressed:

Q = PAI (VI-1)

Where P is the average or effective permeability, A is the cross-
-sectional area of a unit width, and I is the water-table slope.
At a given point along a flow line, the change in storage is

equal to:

AS = PuAqu-PdAdId (VI-2)

Where the subscripts B and 2 represent conditions up—slope and down-slope
from-the well.

Over an arbitrarily short interval along a flow line the thick-
ness of the aquifer and its permeability may be assumed to be constant
as long as the thickness is not significantly affected by the raising

and lowering of the water table. Equation (VI-2) then becomes:

AS

PA (In-Id) (VI-3)

The change in storage is directly proportional to the change in
pieziometric head as long as the specific yield of the aquifer materials

which are drained remains constant.

‘AH PA (Iu-Id) (VIm4)

SY

 

and

AH

C (Iu-Id) (VIwS)

30

From equation (VIwS) the relationship between the rate of well
recession due to gravity flow should be linearly related to the difference
between the water-table slopes above and below the well. This relationm
ship was studied for the six recording wells used for determining groundm
water budgets. Only periods of negligible recharge and low evapotranspir-
ation drain were considered.

Local ground-water level contours were plotted from weekly and
monthly well measurements in the vicinity of each of the six study areas.
From these contours, the slope above and below the central recording
well was computed from the respective distances to the nearest contour
line along the flow line passing through the well, The average daily
recession in the well level was then plotted against this computed dif-
ference in waterutable gradient (Figu'e 9).

Four of the areas showed slope relationships in which the
gradient steepened along the flow line. In the Jack Plne (deep) area
the location of the plots near the water table divide resulted in very
small slopes which could only be roughly defined by the contouring
method. There appeared to be a consistent concavity to ‘he slope in
this area which suggests variability in 'he *ransnissihtlity. The rate
of water-table decline was linearly related to the uztference in slopes,
however.

In the Oak (deep) plots, the up slope gradient was also conw
sistantly larger than the slope below the well. The up~slope area,
which is known to be underlain with till clays at the foot of the
adjoining moraine, probably has a lower permeability than the outwash

sediments below the well.
The rates of waterwtable recession in the Jack Pine (shallow)

and Hardwood (shallow) wells could be related linearly to the slope

Figure 9. Daily waterwtable recession rates due to seepage flow in
relation to the difference in watermtable slope above and
below the study well.

31

 

 

(ft/day)

WATER TABLE RECESSION RATE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-3ng \‘ Hardwood (S) -—
\ Jack Pine (8) -----
. Red Pine (8) - ---
Oak (D) _ '-
-.025# ,L X \ Jack Pine (0) — '-
‘\ \g \ Red Pine (D) —--
‘\ \ \\ \'\
_ \
0W x‘ \ ‘1 O\H
\\ ‘ \ .
\\ \ . \
\ z. .
-.ons \a‘ l '.
\‘a:~. ‘-. \
- x
t: \ ~ \
l \\ .
-.OIOE V\ .1" \ “‘b‘ \‘
= ‘ ‘\
\, \ '\
-.005k \. .. \\
o o
£5005F
-3 -2 -| 0 +3 +2 + 3

SLOPE DIFFERENCE ( Iu - I a xuo’4 n/n)

 

32

difference only after adjusting for the specific yield of the layer in
which the recession occurred. In the other four areas, the variations
in specific yield were too small to affect the relationship of volume of

seepage to the change in well level.

Separation of Components of Wateerable Fluctuations

 

The principal analytic technique utilized for the evaluation of
the ground-water budgets was the separation of the various components
of fluctuation in the waterwtable movements. The method used by White
(1932) for fluctuations within a closed basin was modified to consider
the recession rate due to seepage outflow (Figure 9). In general, the
equation for the change in watermtable elevation for a given time period

can be written:

AH = AHRg - AHRog — AHETg (VIIu-l.)

where AHRg change in head due to recharge from precipitation
AHRog change in head due to net seepage flow
AHETg = change in head due to evapotranspiration drain

The rate of watermtable change due to seepage flow was determined
from the relationship between recession ra‘es during nonurecharge periods
and the differential groundwwater gradients along the flow line in which
the observation well is located. These recession rates were found to be
relatively constant in the deeper waterwtable areas. In the Hardwood
(shallow) and the Jack Pine (shallow) areas, the recession rates were
influenced by the changes in the specific yield of the aquifer when the
water-table fluctuations occurred within the developed solum. Seepage
flow recession rates in these conditions were estimated from the rate

of waterwtable recesSion during the late night hours (12 midnight to

33

6 A.M.), (Figure 10).

The total waterwtable change during a 24~hour period was deferw
mined from the recording well data. The rate of seepage flow recession
was subtracted from this total. When the difference between these
amounts was positive, a net daily head change due to recharge was
indicated. When the difference was negative, that is, when the total
daily recession exceeded the expected seepage loss, an evapotranspira»
tion loss was posted. Conversion of these three values; AH due to
seepage flow, AH due to recharge, andlfiH due to evapotranspiration; to
volumetric units required the application of appropriate specific yield

values.

Determination of Specific Yield for Aquifer Layers

 

Theoretically, the term specific yield, when applied to unconfinei
aquifer conditions, refers to the ratio of the total volume of aquifer to
the volume of water which will drain by gravity from the aquifer (Todd,
1959). This ratio is usually expressei as a percentage, but tor the
purposes of this study where the ratio was used dIIECYIV 1n decimal form,
the data were so expressed. Over short time periods, the passage of the
maximum capillary rise miniscus through a layer of uniform sand has been
shown by Smith (1961) to remove about 90 percent of the total drairahle
water from sands with effective grain sizes of 0.25 to 0,30 millimeters
diameter such as are found in the Udell aquifers. Early experimen.s by
King (1898) demonstrated that even after 2 1/2 years complete equilibrium
was not yet established in five foot vertical columns of sand. Smith s

theoretical analysis of King's results, as well as tests of nonwuniform

sands conducted by Hazen (1892), showed that initial drainage gives a

Figure 10. Separation of the components of watermtable level fluctua»
tions, Hardwood (shallow) study area,

34

 

_. 23..

 

@ 233

n 23...

 

 

 

Aomxa

 

WEI?

 

 

 

 

 

4"

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

=NO.

 

 

(‘3‘!) WflLVO M0138 HLdSO

 

 

 

 

good approximation of the specific yield. This fact becomes particularly
important when the daily fluctuations of ihe water i hie ave to be
evaluated. Childs (1960) characterized the concept of a specific yield
as a rough approximation necessary for progress.

Specific yield values in the shallow ma erwtable plots were
estimated by measurement of the rise in well levels shows *he expected
recession during periods of recharge. Recharge eyen’: were selected
during periods when soil moisture deficits were neg1;gir1e, mostly during
the dormant season. Gross precipitation measurei in the open was adjusted
for interception losses by use of published curves of net interception
versus storm intensity (Kittridge, 1949). A scatter-diagram of the ratio
of throughfall to wateretable rise was plo“ed agaxnc‘ The depth within
the aquifer. The Hardwood and Jack Pine (shallo~) areas showei definite
zonation which corresponded to the degree of s01] profile d£\e10pmcnt
(Figure 11).

Specific yield values in the Hei Pine (shallo~) area and in all
deep wateretable plots were less variable and no consisfan* pattern could
be found in the subsoil layers. Specific yzelj values for these jecpcr
strata were derived by draining 24vinth (olumhe n1 H841;”U*b€j sedimcn‘s
removed from the aquifer layer Just above ‘he 10 e~f :i‘erw‘ible s*agr,
These columns were drained for 48 hours with the *0p of the column pro
tected against evaporation loss. The difference in muls’UfP cont‘nt
between the upper three inches of the column and the lover three inches
was weighted by the bulk density of subsOil sands to obtain a spec1fic

yield estimate. Specific yield values for subsoil samples from all 51K

study areas are listed in Table 2.

Figure 11. Specific yield of shallow aquifer layers for six local study
areas in outwash sands.

36

 

 

(feet)

DEPTH BELOW MEAN GROUND SURFACE

SPECIFIC YIELD By)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0 .l0 .l5 .20 .25 .30
l .o’
9"
0"
..’.-’~H01Pwood (S)
O./
I _
,/|
2 "
3
4 \
Red Pine '-
I
I
I
5 I
I
I
I
I
- u:
I
I
I
1‘
7
I
I
I cum)
I .
I6 Jack Pine w’\. /Red Pine(D)
" |'/
I :
:7 i 'I
I ':
I8 I I.

 

 

 

 

 

 

 

 

 

 

Study Area

 

Hardwood (s)
Oak (D)
Red Pine (5)
Red Pine (D)
JaCk Pine(s)

Jack Pine(D)

1/
— Expressed

sediments.

SPECIFIC YIELD V

’TABLE 2

ALUES OF SUBSOIL LAYERS

 

 

Range of Waterefable Depths Specific Yield l/
(feet)

2.0 e 7 0 0.08 — 17
13.4 m 16.8 245

4.0 e 7 3 20 - 256
15.1 ' 18.5 . .25

2.8 . 7 3 10 - 18
13.1 - 16.6 235

as the decimal ratio

of volume of water to volume of

37

38

Computations of Daily Groundeater Budgets

 

From the fluctuations of the water table as recorded in each
cover-depth situation, a daily recharge discharge balance sheet was com»
puted for the three shallow study areas. An example of this bookkeeping
technique is shown in Table 3 for June 5-20, in the shallow Jack Pine
area.

In the three deep watermtable areas. the fluctuations of the
water table were much more gradual. Recharge from a given storm was
found to be distributed over a period of several days to two weeks. in
these plots there was no direct evapotranspiration from the grounjwwater.
The budgeting process was simplified accorfiingly. The rate of seepage

C
~l

in

flow recession changes very slowly in these plots, therefore, it w
possible to compute the recharge and seepage flow volumes by weekly
periods.

The daily anj weekly imput and drain volumes were totalei for
each month and the net change in storage compared to that computer frwm
the periodic change in water level. This can be eupvessed in equation

form as follows:

As = AH - Sy -_-. Rg — ROg rig (wit 2)

The total ET for the month was computed from the g"oss p"czip3*

tion minus the net ground~water recharge, where the net recharge was

equal to the gross minus ET losses from the saturated zone.

ET = p - Rg + EI'g (VIP-3‘;

This analysis is valid only when there is negligible change in soil

moisture storage between the beginning and end of the month. In the

39

 

vo. moo. o o omo.a so. moo. omo.- oeo.: ea. Ho.v ma
Ho. Hoo. o o oao.u so. moo. omo.a omo.a «a. om.v «a
Ho. Hoo. o o oao.- so. moo. omo.1 omo.- «a. mm.e ma
o o o o ooo. «o. moo. omo.- omo.n ea. am.v ma
o o vo. moo. omo.+. «o. moo. omo.a ooo.n ea. om.e Ha
o o Ho. Hoo. oao.i. eo. moo. omo.a oao.a «a. ms.v oa
eo. moo. o o omo.- eo. moo. omo.g ovo.u ea. se.m m
«o. moo. o o mao.a mo. moo. mmo.! omo.n ea. ms.e m
Ho. Hoo. o o moo.a mo. moo. omo.a mno.; ma. mo.v s
Ho. Hoo. o o oao.: so. moo. omo.n omo.£ ma. mo.e o
o o o o o so. moo. omo.- omo.a ma. mm.e m
Acme Apmo Acao Auwo pro Acao ApmV Apmo atoo oamaw Apmo memo
was mmgmsomm oommaILiq mom momma aid .omom spoon magma
>Hflma poem; 2mm:

mmma

Ame mafia some
.maum mass
mzoas<som2oo smooom mms<auozoomo mo mom2<xm z<

m mam<9

40

shallow water-table aquifers, this condition was usually met since
capillary moisture rise above the water table kept the unsaturated

layers near field capacity. Evidence of this situation was seen in the
prompt recharge response of the shallow wells in these aquifers to almost
every rainfall (Figure 12).

Evapotranspiration in the deeper aquifer areas was estimated for
periods between recharge events. The assumption that recharge can only
occur when soil moisture deficits are satisfied was not completely
justified since low recharge totals were measured from storms which
occurred when soils were assumed to be at field capacity by this critew

rion.

Evaluation of Snow~Melt Recharge

 

Groundnwater recharge from the water released during melting of
winter snow made up 40 to 80 percent of the total annual recharge during
the two-year period. Since the high permeability of the surface 50115
on all six study areas permitted infiltration of all melt waters. the
recharge of ground-water might be expected to closely parallel the
water content of the snOWpack. Cumulative recharge during the March-
April period was computed for each study area. The water content of
the snowpack in each forest cover type was measured during the winter
buildup and melting period. During the winter and spring of 1962, these
measurements were confined to the five-point snow courses associated
with the precipitation network (Figure 6). In the second year, supplew
mental snow measurements were taken on the study plots where 25 snow
cores were weighed, five in the vicinity of each of the five wells in

each coVer-depth area.

Figure 12. Water-table level fluctuations during two-year study period,
mean of five wells in each local study area.

41

 

 

mom. «mm. .mm.
8m 84 .2. :2. so: .332 no“. :2. So 82 to 9m 2.4 .2. :2. so: .362 no... :2. 08 >02 to

 

NOILVA313 318%]. 831W

 

 

The total amount of moisture available for recharge during the
March-April periods included the snOWpack water at the start of the melt,
plus rain and snow received during the melt period. Evapotranspiration

losses in the early weeks of spring were assumed to be negligible.

CHAPTER VII

RESULTS

Precipitation during the two-year period of this study was very
close to the long—term mean of 32.10 inches as measured at Wellston,
five miles east of the study area (U.S. Dept. of Commerce, 1962, 1963,
1964). The average at the 12 gauges on the Udell Experimental Forest
was 33.11 inches during the 1961-62 water year (Table 4). For the
second year this average was only 30.10 inches. The entire study period
began with soils near field capacity following 8.75 inches of rainfall
in September, 1961. For the remainder of that autumn period, rainfall
was above normal. Winter precipitation was above normal until March.
Deficient spring rain was followed by abundant June and July precipitam
tion. The water year ended with slightly below normal precipitation.

The 1962-63 water year was marked by deficient rainfall in late
fall and late spring. December and March were the only months in Uthh
average precipitation was more than an inch above the longeterm norm.
This second year began with a condition of soil moisture deficiency and
ended with an even greater deficiency (Table 5). The two years were
characterized by the theoretical computation of the water budget using
Thornthwaite's potential evapotranspiration (Thornthwaite and Mather,
1957). The obvious difference was in the autumn recharge period when
a water yield of nearly 4.5 inches was indicated during the first year.

In the fall of 1962, the excess precipitation above evapotranspiration

demands was sufficient only to restore soil moisture to field capacity

43

TABLE 4

MEAN MONTHLY PRECIPITATION, UDELL EXPERIMENTAL FOREST AND
MONTHLY TOTALS ON LOCAL STUDY AREAS

1/ Average Average Average
Normal" Average Hdwd.(S) Red Pine(S) Jack Pine(S)
ppt. 12 Udell and and and
Wellston Stations Oak(D) Red Pine(D) Jack Pine(D)

 

 

1961—62 Water Year

 

 

 

Oct. 3.07 3.37 3.42 3.32 3.26
Nov. 2.97 3.54 3.56 3.44 3.57
Dec. 2.01 2.65 2.79 2.38 2.36
Jan. 2.06 3.49 3.64 3.34 3.54
Feb. 1.55 2.62 2.52 2.38 2.64
Mar. 1.83 1.37 1.49 1.20 1.29
Apr. 2.63 0.92 1.24 0.88 0.88
May 3.08 1.64 1.19 1.92 1.90
Jun. 3.18 4.40 4.36 4.32 3.80
Jul. 2.69 2.99 3.14 2.66 2.94
Aug. 3.33 2.85 2.91 2.76 2.82
Sept. 3.70 3.27 3.40 3.32 3.24

Annual 32.10 33.11 33.66 31.92 32.24

1962-63 Water Year

Oct. 3.51 3.42 3.70 3.70
Nov. 0.57 0.56 0.56 0.58
Dec. 3.48 3.37 3.55 3.26
Jan. 2.76 2.63 2.78 2.61
Feb. 1.42 1.43 1.57 1.45
Mar. 2.91 3.00 3,30 3.28
Apr. 2.58 2.60 2.54 2.33
May 2.19 2.19 2.16 2.21
Jun. 1.22 1.25 1.07 1.12
Jul. 3.70 3.67 3.60 3.75
Aug. 3.52 3.56 3.56 3.39
Sept. 2.24 2.06 2.37 2.19
Annual 30.10 29.74 30.76 29.87

1
_/Climatologica1 normal based on period 1931-1960.

44

.mmLQCH mm «o swamp

 

ucfluoon can unmouoa v no mucmu musumwoe manuafiw>m cues .nHHOm ucmm usedmmpu how xawumeaxouaam I 6052mmm :o.m we zufiomomo Uaofiw d I
E
mm.- m®.m b®.m~ vn.©+ mm.mm w~.om o~.om m.mv Hmscc<
o vn. >N.N mo.OI no.0 hw.® hm.ol Ax.m mo. mm.v vm.m w.hm .oovm
0 Av. mm.m No.01 wo.o 0H.® vv.ol om.m AH. Nh.m mm.m h.m® .m:<
o mm.~ Hw.m HA.OI o~.o mm.m hm.al bo.m Ma. vn.w Ob.m H.mm .an
o wv.~ mm.m mm.wl Hm.o mm.v Hw.ml mo.m ma. mw.b NN.H o.>m .:3w
0 o >®.N wv.o: vn.m mv.o mv.01 hm.m no. Ha.m m~.m o.~m mm:
vN.H o vm.a o oo.m o VN.H+ vm.H v0. wn.H mn.m H.vv .n;<
HH.OH o o o oo.m o Hm.m+ o o o Hm.m b.0m .Lmz
e o c o oo.m o mv.a+ o o o NV.H m.mH .20w
0 o o o Gown 9 wh.m+ o o o mh.m m.m~ .cmw
o o o ®V.O+ oo.m o wv.m+ o o o xv.m w.mm .omo
o o mv.o mo.©+ vm.m o mo.o+ me. No. Av.o hm.o 0.5m .>cz
e c co.m Hm.~+ mv.m o Am.~+ co.m no. wo.m Hm.m m.em .uoo
mflowlflueew... werw we:
Nw.v~ Av.m m>.om mm.w+ w~.vm ww.mm .-.nm , m.mv Hmscc<
o o ~m.m mv.$+ vx.o 3 mv.3+ gm.m me. on.v hm.m m.wm .unom
o ~m.~ bd.m mm.ou mm.o mo.o mm.d| ww.v md. no.5 mw.m m.w® .u:¢
o 24 and one: 35 8.6 3.7 $6 3. $1.4. mod. «.3 43.
0 mm. ~n.v Hm.01 mm.~ an.m no.0: no.0 m~. mv.h o¢.v w.mw .::w
0 av. mm m vm.~l hm.fi ww.~ m>.~| mv.m mo. aw.v vw.~ 0.5m xx:
0 c HO.H mo.ou ~m.m mo. mo.on Ho.fi me. rm.o mm.c m.:v .La<
mH.oH 3 o o oo.m : nm.~+ c o o hm.~ m.mm .Lms
o o e o 0:.m 2 m®.m+ o o o mw.m m.nd .29;
c o c c oo.m c mv.m+ a 3 o av.m m.>~ .::w
o c o o oc.m 2 mw.m+ a 3 o mw.m u.vn .cva
om.m : vm. : oo.m e om.m+ vm. do. mm.o vm.m m.®m .>:z
mo.H : mm.m o oo.m c mo.~+ mm.m mo. mm.m hm.m m.om .300
36.2.8623 3.5%
113.31-- 14:37.. Q3: 119133. 1-9.4.1-.. -ilmeflis 9737-, 1115.33: .943: .wru i E: ,1. .23
Uflwwy ewcemoc km mumqum mumpcom 21:; area: maid zazucofi ma xmocm and >Hzecc:
poem; .umao: .uo< EmaV .umHos ~.uo; we afies: use: xdzucoz cams
130m .5664. .23.... .325 coo:
\H

hhmmdv mmzk<£ a mFH<3IszOIF I sz202 >m MUZ<J<m mmb<3 4<QHFEKCEZH

A. $424?

46

levels.
Computed actual evapotranspiration was more than one inch greater
in the first year, while estimated water yield was more than three inches

greater.

Gross Ground-Water Recharge

 

The total volume of water which was added to ground-water by
percolation from rain and melting snow was computed from the daily and
weekly aquifer storage changes. The monthly and annual totals for each
of the six study areas are listed in Table 6. In every area, the total
amount in 1961-62 exceeded that in 1962-63. In general, these annual
differences were similar to the predicted difference obtained from the
Thornthwaite calculations. A large portion of this difference occurred
as autumn recharge.

Total recharge in the Hardwood (shallow) area exceeded that of
all other cover-depth conditions. During a major portion of the year,
the water table in this plot area is within 18 inches of the soil surface
(Figure 12). Capillary moisture rising above the water table maintains
this soil at or near field capacity. Consequently, rainfall even in
moderate amounts, produces a marked rise in the water-table level. Since
the evapotranspiration withdrawals (Table 7) are greatest from these
shallow water-table areas, the net recharge differs less than the gross
recharge between the deep and shallow counterparts.

Figures 13 and 14 show the monthly gross recharge, net recharge

and evapotranspiration loss from ground-water in comparison to the total

monthly precipitation. These histograms show that the greatest amount

of recharge occurs during the dormant season. In 1962, autumn rainfall

TABLE 6

GROSS GROUND-WATER RECHARGE (Area Inches)

Hardwood Oak Jack Pine Jack Pine Red Pine Red Pine
(Shallow) (Deep) (Shallow) (Deep) (Shallow) (Deep)

 

 

1961-62 Water Year

 

 

 

Oct. 2.99 0.62 1.87 0.23 1.37 0.16
Nov. 2.64 3.83 2.84 2.93 3.10 2.84
Dec. 0.68 0.91 0.65 1.34 0.72 1.38
Jan. 0.20 0.21 0.24 0.30 0.29 0.36
Feb. 0.60 0.35 0.08 0.16 0.30 0.23
Mar. 6.94 6.86 5.08 3.26 5.43 3.05
Apr. 1.07 1.73 1.33 4.32 1.82 4.87
May 1.07 1.61 0.65 0.32 0.48 0.68
Jun. 2.39 0.79 1.85 0.84 0.88 0.34
Jul. 1.67 0.29 0.71 0.37 0.07 0.12
Aug. 0.44 0.02 0.37 0.35 0.19 0.24
Sept. 0.89 0.20 0.18 0.44 0.18 0.24
Annual 21.58 17.41 15.85 14.90 14.83 14.51
1962—63 Water Year
Oct. 1.92 0.07 0.49 0.18 0.26 0.16
Nov. 0.41 0.82 0.46 0.41 0.34 0.49
Dec. 0.85 0.28 0.09 0.10 0.35 0.12
Jan. 0.71 0.34 0.18 0.25 0.14 0.08
Feb. 0.08 0 0.02 0 0.12 0
Mar. 7.46 5 36 5.12 1.42 5.44 1.32
Apr. 2.33 2.86 3.19 6.20 3.01 6.55
May 1.80 2.39 1.49 1.48 0.86 2.16
Jun. 0.66 0.24 0.01 0 0.02 0.28
Jul. 1.78 0 29 1.25 0.42 0.68 0.53
Aug. 0.43 0 24 0.41 0.34 0.10 0.12
Sept. 0.31 0 0.02 0.18 0.01 0.08
Annual 18.74 12.89 12.73 10.98 11.33 11.89

47

TABLE 7

EVAPOTRANSPIRATION DRAIN FROM GROUND-WATER IN
SHALLOW WATER-TABLE AREAS (Area Inches)

 

 

 

 

 

 

Month Hardwood(S) Jack Pine(S) Red Pine(S)
1961-62 Water Year

October 0.39 0.16 0.12

November 0.07 0.04 0.01

December

January

February

March

April 0.04 0.24 0.13

May 0.82 0.56 0.64

June 2.35 0.50 0.65

July 1.32 0.42 0.71

August 0.23 0.06 0.35

September 0.07 0.22
Annual 5.29 1.98 2.83

1962—63 Water Year

October 0.01 0.18

November 0.16 0.01

December

January

February

March

April 0.02 0.06 0.05

May 0.34 0.53 0.32

June 1.38 0.74 0.15

July 1.78 0.64 0.29

August 0.28 0.08 0.20

September 0.17 0.07
Annual 4.14 2.05 1.27

48

Figure 13. Monthly precipitation, gross and net recharge and evapo-
transpiration from the saturated zone for six local study
areas, 1961—62 water year.

49

I96I-2 WATER YEAR

RED PINE (S)

S

   

\I'A

   

HON

    

  
   
  

Proci

ONDJFMAMJ

éc'nli-cbnbétbdl;

JACK PINE (S)

HARDWOOD ( S)

 

:IIIIIIQIIIIII
I
I
I
I

 
     

 
 

I
SIIIIIIII

 
 

Gross Recharge

 

ONDJFHAMJJAS

uIlnIIIIII
I

I
50‘
I

    

’III
I
I

    
   
   

ONDJFMAHJJAS

   

Nat Rochar- o

méééééfiofi;

SEHDNI VEHV

 
 

RED PINE (D)

QCNIDDV’NIOJ

JACK PINE (D)

    

016866613361;

.....

ONDJFNAMJJAS

 

   
   

ONDJFMAHJJAS

 

:IIIIIIIIIIIIIII

 

696?;(0'065303
canon: Hana

o'uoaruAuaaas

    

Figure 14. Monthly precipitation, gross and net recharge and evapo-
transpiration from the saturated zone for six local study
areas, 1962-63 water year.

50

l962-3 WATER YEAR

RED PINE (SI

 
    

ooeonéwba;

JACK PINE (S)

   
   
  
       

.IIIIE m
E <
3'

- '5
’IIIIIIIII'.

 

Pnci Ration

I
I
I
IIIIII
I
I
I
.IIII
I
I
SIIIIgoggggg. .. ~IIIIIIIIIII

o N I) J F’Iu A M if’

  

Gross Recharge

 

,Q\&
0 $5 $5:
. \ 1 . .“ >
:IIIIIII.$R$\“>
I
I
I
‘Illl'
I
I
: I
III! :0.
= .
I I
IIIIIIIIIIIIr 'IIIOIOIIIII:§?

s?
IIIIIIIIIIII 1“

ONDJFHAMJJAS

monomcntfi;

HARDWOOD (SI

JAS

     
    

.

   

ET from So '.

ONDJFMAM

Not Rochar.

RED PINE (DI

    

compound-.55.;

JACK PINE (D)

OAK (D)

 
  

o'tcbl'soiuéubél;

  
   
     
   

I
I
I ‘4
IIIIIIIIo

ONDJFHAMJJAS

IIIIIIIII

   
  
  
     

   

«E §§D§§X

as \ kwfixxfim 3

“\§§Nw#a§wai§$%§w
I

'1
I

<> u I) J F'Il A .1 a J .A s

   

O
I
I
\
IIII
I
I
I

IIIIIIIIII “a
q-
\u
f-g. ..
‘xwy
IIIIIIIIIII

ONDJFMAMJJAS

a'té: I; énéfioi _'.
SBHDNI VEHV

éosooena;
SBHDNI vaav

51

was not sufficient to produce recharge before winter precipitation began
to accumulate in the snowpack.

During both winter periods, continued low temperatures maintained
the snowpack until March. At this time, the shallow water-table plots
received most of their annual increment of ground—water. Earlier snow
melt under the oak plots allowed a larger proportion of this snow water
to reach the water table before the end of the month. Insulation of the
crowns in two pine areas delayed snow melt until April. The recharge
timing differences and the comparative increments from growing season
precipition are shown in Figure 15.

The cumulative recharge curves (Figure 16) show these differences
in recharge timing in more detail. In the Hardwood (shallow) area, the
combination of high water-table levels and low evapotranspiration after
leaf fall, produced high rates of recharge in 1961. In the comparative
deciduous forests with a deep water table, the recharge was lower by
nearly one inch. A part of this difference is compensated for by the
evapotranspiration drain of nearly one-half inch on the grounduwater
beneath the Hardwood (shallow) stand.

During the winter months, recharge from snow melt was slight
but both hardwood areas showed greater recharge than the pine areas
where brief periods of warm weather had less effect on the snOWpack.
Differences in the mean water content of the snOWpack and the recharge

during and following the snow period are discussed below.
Heavy rainfall in mid—June and mid-July produced marked recharge
in both shallow Hardwood and Jack Pine areas. At this time, the water

table in the Red Pine (shallow) was 4 1/2 to 5 feet below mean ground

level. The recharge pattern was very similar to the three deep water table

Figure 15. Weekly increments of gross recharge in six local study areas.

52

Inches

AI‘DO

GROSS RECHARGE (weekly)

999TN‘9‘?

90-wqeset'99999789

9799999???“

:LaLhi

HARDWOOD (S)

 

OAK (D)

JACK PINE (S)

 

JACK PI NE (D)

       
    

 

RED PINE (S)

   

RE D PINE (D)

 

   

 

I96I-2 WY 1 I962'3 WY

 

 

Figure 16. Cumulative gross recharge in six local study areas by water
year.

53

 

 

Inches)

(A no

CUMULATIVE GROSS RECHARGE

 

o-Nu-buaiflmco

I96l-2 ,r-’

..
O...

      

3 . ..... Hardwoed(S)

. .-'.! .__._.00k (0)
..........Jock Pine (5)
-.._.....Jack Pine ID)
.......... Red Pine (5)
-........- Red Pine(D)

 

 

OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP

I962-3 Pd,’

 

 

   

OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP

 

  

 

54

areas which indicates the building up of a soil moisture deficiency above
the capillary zone. At the same time, there continued to be a marked
diurnal fluctuation in the rate of water-table recession which showed
that ET moisture was still being obtained from the ground-water source.
Excavated profiles in the three shallow water-table areas provided an
explanation for this anamalous response. Fine roots (less than 0.2 inch
diameter) were distributed more evenly through the B horizon in the Red
Pine (shallow) stand than in the other two shallow water—table types
(Figure 17). There is also five-six feet of microtopographic variation
in the Red Pine (shallow) area which places approximately one-third of
the stand area in a well-drained condition soon after water—table levels
drop below their annual peak.

The Oak (deep) area received markedly greater recharge during
April and May than was measured in the two deep water-table pine areas.
Full hardwood leaf development was not reached until the last week of
May. Presumably lower transpiration rates in the Oak accounted for the
recharge differences.

In the fall of 1962, only the Hardwood (shallow) area received
measureable recharge before the end of November. By that date, the
water table in the Jack Pine (shallow) and Red Pine (shallow) areas was
well below the root zone.

Both deciduous forest areas showed greater winter recharge during
brief winter melting periods. Snow-melt recharge began and ended one to
two weeks earlier in these two areas than in the pine areas with compar~
able ground—water depths. Recharge continued at higher rate under Hard~

Wood stands until leaf development was complete in the first week of

June.

Figure 17. Root distribution in relation to depth in six local study
areas.

55

 

 

 

 

 

 

 

 

 

«v-00

 

 

 

 

 

 

 

 

 

 

8. 56 3.8 E 68324...
SIN
euro. m
- 7 2.9 M
W «To a H
_ 0'0 8
3
.l
m
a: mzi 204.. as uza 53. w
onéu 3
can. w
«Tu.
«Tu _ w
6-0 0
n
N
O
S
m
.9 mzi cum omen a: qu cum w.
emum. 9
2-9 3
«To
_ two _
co co ow 0.. 8

mmw><4 Ikmmo cw>m A.Eo_v-N.Ovv whoom w:.m.m_> no mufikzmummm

 

 

56

Following the prolonged drought period from late May to midoJuly,
only the Hardwood and Jack Pine (shallow) areas produced heavy recharge
from the drought-ending storms with a gross rainfall of 3.3 inches.

In both water years, the two hardwood forest areas produced more
gross recharge than any of the pine types. The Jack Pine (shallow) area
had the greatest gross recharge of the conifer study areas. During a
portion of each growing season, the Red Pine (shallow) area did not have
water-table levels sufficiently high to prevent the development of soil
moisture deficits. Except during the early part of the growing season,
ground-water recharge under this stand was similar to the two deep waterw

table pine stands.

Evapotranspiration From Ground-Water Supplies

 

In the shallow ground-water areas, the accelerated rate of water»
table recession during the day-light hours indicated evapotranspiration
losses (Figure 10). The sum of these daily recessions multiplied by the
specific yield of the appropriate aquifer layer is shown for each month
and year in Table 7. The Hardwood (shallow) area with the highest
average water level during the growing season exhibited the greatest
diurnal fluctuations and the greatest annual ground—water losses to
evapotranspiration. The rate of ET loss was greatly accelerated after
June 1, when the forest was in full leaf.

ET losses in the Jack Pine area were minor after July when the
water—table level fell below the B horizon. In April, 1962, moisture

use was greatest in this area during a low rainfall period. In the
following April,cold temperatures and abundant rainfall limited grounde

water use for ET.

57

ET losses were most even throughout the growing season in the Red
Pine area. The effect of lower average ground—water levels during the
1962-63 water year was shown in a much lower total water loss (1.26 in

1962-63 vs. 2.82 in 1961-62) under this forest cover.

The comparative effects of ground-water depth on the amount of
ET drain from the aquifer is shown in Figure 18. The monthly water
losses are plotted as a percentage of the total potential evapotranspira—
tion for that month (See Table 5). By making this conversion, it was
possible to compare all monthly periods during the growing season.
Separate regression lines were fitted to the monthly observations for
each of the three shallow water-table areas. April, May, and November
data were excluded from the Hardwood (shallow) analysis. At these
times, the principal species were dormant and the ET rates did not fit
the trends which were evident for the remainder of the snow-free period.
These data are indicated by the subscript "d" on Figure 18.

It is evident from this figure that ground-water losses to ET
cease when water—table levels drop below 4.5 feet in the Hardwood and
below 5.5 feet in the Jack Pine. In the Red Pine area, the slope of
the regression line is less steep. No clear cutoff point was evident

from the data obtained during the two years of study. Obviously ground-
water losses to ET become negligible at water—table depths below eight

feet. Due to the microtopographic conditions mentioned earlier, only
a portion of the trees on this area ever obtained ground-water during
the study period. Conversely, some low lying sections may be able to

derive capillary moisture from the saturated zone when the average depth

is greater than eight feet.

Figure 18. Monthly evapotranspiration from the saturated zone as a
percentage of potential evapotranspiration in relation to
water-table depth.

58

 

368:3. 6:305 .565 30.3 Coon:

Ihmmo medF mmk<3 Z<m2

 

 

 

m h m m e m N _
I,
III I
“I u I
I‘ll
I III
I
I
comomm Eoctoolu
whine Amy mzi ommno
emf". “my sz x0336
wbru. Ame 0003931-.“

9

O
N

O
M

O
‘1’

O
I!)

 

("/o) UBLVM’ONI'IOHD W085 GBNIVLSO (5.1.3 :IO NOIJJJOd

 

 

59

Net Ground-Water Recharge

 

A comparison of the actual effectiveness of the six study areas
as water producers is shown in Table 8. The amounts of ground-water
utilized for ET has been subtracted from the gross recharge figures. It
is evident that both deciduous forest areas yielded more water for
seepage flow than did any of the pine areas. The greatest difference
was 5.4 inches (45%) between the Red Pine (shallow) and the Oak (deep)
during the first year. Both deep pine areas produced 2.5 - 3 inches
less than the deep Hardwood area.

Among the shallow water-table types, the Hardwood area yielded
2.4 inches (17%) more than the Jack Pine and 4.3 inches (35%) more than
the Red Pine in the first year.

All areas showed reductions in net recharge during the second
year. These differences, ranging from a 1.7 inch reduction in the
Hardwood (shallow) to a 4.5 inch reduction under the Oak (deep), were
doubtless affected by the soil moisture deficit which existed at the
start of the second water year. In the shallow plots this deficit was
less severe and the reduction in net recharge, over that due to the
difference in precipitation, was not as large as the reduction in their
deep water-table counterparts.

The negative net recharge values in the shallow water—table
areas during June, 1963, indicated a net loss from storage during this
low rainfall period. The Red Pine (shallow) area showed a net loss
during several other growing season months resulting from low gross
recharge and a continual, though small, utilization of ground-water.

In this respect, the deeper rooted red pine was consistantly the highest

water use area. Although water-table levels were higher, on the average,

TABLE 8

NET GROUND-WATER RECHARGE (Area Inches)

 

 

Hardwood Oak Jack Pine Jack Pine Red Pine Red Pine
(Shallow) (Deep) (Shallow) (Deep) (Shallow) (Deep)
1961-62 Water Year
Oct. 2.60 0.62 1.71 0.23 1.25 0.16
Nov. 2.57 3.83 2.80 2.93 3.09 2.84
Dec. 0.68“ 0.91 0.65 1.34 0.72 1.38
Jan. 0.20 0.21 0.24 0.30 0.29 0.36
Feb. 0.60 0.35 0.08 0.16 0.30 0.23
Mar. 6.94 6.86 5.08 3.26 5.43 3.05
Apr. 1.03 1.73 1.09 4.32 1.69 4.87
May 0.25 1.61 0.09 0.32 -0.16 0.68
Jun. 0.04 0.79 1.35 0.84 0.23 0.34
Jul. 0.35 0.29 0.29 0.37 -0.64 0.12
Aug. 0.21 0.02 0.31 0.35 -0.16 0.24
Sept. 0.82 0.20 0.18 0.44 —0.04 0.24
Annual 16.29 17.41 13.87 14.90 12.00 14.51
1962-63 Water Year
Oct. 1.91 0.07 0.49 0.18 0.08 0.16
Nov. 0.25 0.82 0.46 0.41 0.33 0.49
Dec. 0.85 0.28 0.09 0.10 0.35 0.12
Jan. 0.71 0.34 0.18 0.25 0.14 0.08
Feb. 0.08 0 0.02 0 0.12 0
Mar. 7.46 5.36 5.12 1.42 5.44 1.32
Apr. 2.31 2.86 3.13 6.20 2.96 6.55
May 1.46 2.39 0.96 1.48 0.54 2.16
Jun. 0.72 0.24 -0.73 0 —0.13 0.28
Jul. 0 0.29 0.61 0.42 0.39 0.53
Aug. 0.15 0.24 0.33 0.34 -0.10 0.12
Sept. 0.14 0 0.02 0.18 -0.06 0.08
Annual 14.60 12.89 10.68 10.98 10.06 11.89
Difference
al.69 —4.52 -3.19 —3.92 -l.94 -2.62

 

60

 

 

61

in the Jack Pine (shallow), averaging 4.04 feet in comparison to 6.17
feet below mean ground surface during the two growing seasons, the more
restricted rooting depth of the Jack Pine limited moisture use and
allowed for greater net recharge.

One of the principal factors in the low net 1962—63 recharge
of all types, was the lack of autumn rainfall. Only the Hardwood
(shallow) area received a considerable amount of recharge above the
ET usage during October. Rainfall late in October resulted in 0.8
inch of recharge under the Oak in November, while producing only one-
half this amount in the pine types.

There was clear evidence of earlier transpiration demands in the
conifers in the relative amounts of May recharge in both years. Delayed
drainage of snow-melt waters in the deep water-table plots obscured this
relationship in April in the monthly totals, but was evident in the

cumulative recharge curves (Figure 16).

Snowpack Accumulation and Spring Recharge

 

McGuinness (1941) has stressed the important role of snow-melt
waters in the recharge of ground-water in the Lower Peninsula of Michigan.
Coming during a period of low evapotranspiration losses and at a time
when soil moisture is at or near field capacity, the water in the melt-
ing snOWpack is available for runoff or recharge depending on the
infiltration capacity of the soil, topography and soil frost conditions.
In sand soils of the Udell Experimental Forest, surface runoff from snow
melt is negligible even when the soils are frozen. Probably this is due

to a combination of spotty soil frost patterns (Striffler, 1959) and

gentle relief. It is obvious from the monthly and the cumulative annual

62

recharge patterns that a large portion of the total recharge increment
resulted from snow melt in both years of this study.

The winters of 1961-62 and 1962—63 were periods of continual
snowpack accumulation with only minor melting periods prior to the spring
breakup. The maximum water content of the snowpack during each winter
was determined for the three cover types (Figure 19). Consistant dif—
ferences in the amounts of water stored on the ground were inversely
related to the dormant season crown densities. The highest snowpack
water equivalent values were found under deciduous stands which have
low (10%) crown densities in the dormant season. This was true in the
well-drained sections of the forest. In shallow water-table hardwoods,
the snOWpack was melted at the ground—line and the maximum water equiva-
lent never reached the 7.6 and 7.4 inch values which were measured under
upland oaks and aspen in 1962 and 1963, respectively. The early winter
recharge which resulted from this mid-winter melting was most obvious
in the winter of 1963 as is shown in the Hardwood (shallow) cumulative
recharge curve (Figure 16).

The mean of 25 snow samples taken in this area on six dates in
the later half of the winter, are shown in Figure 19. From January 28,
to February 18, during a period of very low temperatures, the snowpack

in the high water-table hardwoods accumulated at the same rate as that
in the well-drained portions of the hardwood forest. Between the next
two sampling dates, the water equivalent of the upland snOWpack increased
0.8 inch, while that in the Hardwood (shallow) area declined 0.1 inch.

During this time, the wells in the shallow water-table portions showed
a recharge of 0.5 inches, while the Oak (deep) received no measureable

input.

Figure 19. Average snOWpack water equivalent values by forest cover
type.

63

 

mam.
mn_< Id! mm... 24..

 

 

 

   

¥<Ousz o m

m0003om<1
QZ<JQD

  

 

    

\.r\
5:6sz cum

m0003om41
024.35

 

('3 'M ‘OIIOUD MDVdMONS

 

 

64

The Jack Pine areas, with a uniform crown density of approximately
82 percent, accumulated the lowest snowpack. The maximum water equivalent
in both deep and shallow water-table zone was 5.8 inches in 1962 and
5.6 inches in 1963. Apparently the water-table levels in the shallow
pine areas were sufficiently deep so that the latent-heat of the ground-
water did not induce melting. There also may be less opportunity for
mid-winter melting due to slower ripening of the snowpack beneath the
conifer crowns.

The Red Pine areas, with an intermediate crown density of about
65% in winter, contained a maximum snOWpack of 6.5 inches in 1962 and
6.7 inches in 1963. These values, intermediate to the two conditions
mentioned previously, were identical in both deep and shallow water-
table lands in 1963. During this winter in which intensive snow sampling
was carried out in the six study areas, the snowpack was found to be very
irregular, having patches and bands with water equivalent values as high
as 8.5 inches in the holes created by the deciduous oak component of the
stands. In comparison, the snOWpack beneath Hardwoods and Jack Pine was

regularly distributed except for major aspect differences.

The portion of gross annual recharge which resulted from winter
precipitation is shown on an expanded time scale in Figure 20. The
earlier melting which occurs under the open Hardwood cover is best
illustrated by comparing the recharge timing of the three shallow water-
table areas. Recharge in the deep water-table zones shows the same
timing difference, but all recharge patterns are delayed by the time
required for percolation.

Snow-melt recharge had ended in all the shallow water-table

areas by April 12. The total recharge in these areas is compared in

Figure 20. Cumulative ground—water recharge during winter and snow—melt
periods in six local study areas.

65

and 31102-551:

 

 

(Area Inches)

CUMULATIVE RECHARGE

O

‘1

fi

0|

 

I96I-2

 

+’—--
” 7 ~—
I
1"
I,

I .."""" /

I / ’.

’ ' I

I 0 .l

-'
I _ —-

’ an— - —. ’1: O

’0’ "0"- ’l:
";" ”‘ .0 /

Ig" ‘ I -../
" 00—..-00_.O—OI—00-00’
,0
0’ .
1" ’ oe-

Dec Jan Feb
Hardwood (S) ——
Oak (0)
Jack Pine (5) ------
Jack Pine (0) --°—
Red Pine (
Red Pine (D) “W“

I962-3

..OIIIIIIIO
#

.......

 

’-..y-‘--.--’.-.T.- ----- - . .. .:

..’
J .-..-—I é o :0

a . .-’- 0-0 -0
. ’ ‘fiwazuat'z'fiz'uutu- . _. 2 . .—
'0

Dec Jan ‘Feb‘; March

 

 

 

66

Table 9.

These recharge values correspond to the differences in snowpack
accumulation. Total recharge in the Red Pine (shallow) area in 1962
appears anamolously high when the difference in precipitation is con-
sidered. As was explained earlier, the location of these plots near
a strip of sparsely wooded private land may have resulted in inflow due
to more rapid recharge in the open land. Short periods of altered
ground—water flow patterns were noted during both recharge periods in
the Red Pine (shallow) vicinity. Appropriate corrections were made in
the inflow-outflow balance to the extent permitted by weekly and monthly
measurements of surrounding wells. Nevertheless, the precision of the
gravity flow recession rate estimate, which was basic to this analysis,
was least accurately determined during this period of rapidly changing
water-table levels.

The total snow-melt recharge in the deep water-table areas is
not easily separated because of the longer delay in recharge response
in the wells. The percolation of autumn rainfall caused greater recharge
in all three deep areas during December of 1961 than in any of the
shallow areas. It is unlikely that a greater portion of the early snow-
fall melted on these better drained sites during this period. Recharge
from the spring melt continued throughout April and into early May. Rain-
fall during the spring months also contributed some recharge so that there
was no clear end point to the snow—melt contribution. Figure 20 shows
that, on April 12, of each year, the total recharge in these deep areas,

though still increasing, was of the same relation as that shown by the
shallow water-table areas. The Oak (deep) had received the greater

amount of recharge, partly due to the earlier snow melt. The Red Pine

TABLE 9

WINTER AND SNOW—MELT RECHARGE,
MAXIMUM SNOWPACK AND WINTER PRECIPITATION
BY COVER TYPE IN THE SHALLOW WATER-TABLE ZONE

1961—62

Max
Dec-Mar SP
ppt(in) WE(in)

1/
Hdwd (S) 10.46 7.6 -
Jack Pine(S) 9.83 5.8
Red Pine (S) 9.30 6.5

l
—/ Snowpack data from well—drained hardwood area used

all—winter buildup in conifers.

Total
Recharge

(in)

8.

67

73

.84

.95

Dec-Mar
ppt

(in)

10.43

10.60

11.20

1962—63
Max Total
SP Recharge
WE(in) (in)
7 4 9.14
5.6 7.70
6.7 8.79

for comparison to

68

(deep) was intermediate. The Jack Pine (deep) has the least accumulated
winter and spring recharge.

These six case histories show a consistant pattern in which
higher crown densities produce lower snOWpacks and lower ground—water
increments following snow melt. Except for the slightly greater winter
transpiration which may be expected in conifers, the principal cause for
these differences must be greater interception losses from the denser
crowns. Numerous reports have shown that less snow accumulates under
dense conifer crowns (Hart, 1963; Lull and Rushmore, 1960; Weitzman and
Bay, 1958). This analysis has demonstrated a measureable difference in
water yield associated with these snowpack differences.

In 1961—62, the recharge received from winter precipitation
accounted for from 49 to 66 percent of the total net recharge for the
entire water year. This was a year of abundant autumn rainfall. Recharge
during the 1962-63 water year was much more dependent on snow water.
Sixty-one to 88 percent of the net annual recharge came during the winter
and early spring. Because of their longer transpiration period, conifer
stands, in general, received less recharge during spring and autumn.
Winter recharge made up a larger portion of the annual total and in these
cover types the winter precipitation contributed less to ground-water

supplies than it did in the deciduous forests.

Evapotranspiration

 

Estimates of the total annual evapotranspiration in each of the

six cover-depth conditions were computed from the formula:

ET = p - Rg + ETg -ASM (VII-l)

69

where ASM is the change in soil moisture storage between the beginning
and the end of the water year. The 1961-62 period started with soils at
field capacity following heavy rainfall in late September, 1961. The
ground-water recharge increments in the autumn of 1962 showed that only
in the Hardwood (shallow) area were the soils near field capacity at the
end of September. The soil moisture deficit in the other five areas was
not measured directly. Using the accumulated soil moisture deficit which
was indicated by the Thornthwaite computation for September 30, the total
annual ET values were computed. For the l962a63 water year, the difference
between the calculated deficit on October 1, 1962,and September 30, 1963,
was used. The annual ET values computed by this method are shown in
Table 10.

The shallow pine areas had the highest evapotranspiration rates
in both years, corresponding to their low net recharge values (Table 8).
The deep water-table pines used from 0.3 to 2.5 inches less than their
shallow counterparts. The hardwood forests were consistantly lower in
total water use than the pines. Surprisingly, the shallow water—table
Hardwood area used less water for ET than the well-drained Oak area.
This can be explained partly as a function of rooting depth. The shallow
root development in the poorly-drained mineral soil must produce lower
ET rates when the water table is lowered below the two-foot level.

Monthly evapotranspiration during the growing season was computed
from the precipitation, net recharge and soil moisture storage changes

(Tables 11 and 12). In the absence of on-site measurements of soil
moisture storage, monthly changes in storage were computed from the
theoretical depletion indicated by Thornthwaite's water balance method.

In the deep water-table areas, it was also necessary to estimate the

70

mm.m~

hm.m~

ADV.Q Umm

 

 

me.am wn.oa mo.om «6.83 ea.ma moimoma
mo.mm om.ma mm.om He.ma nm.ea mouaoea
Amv.a pom Ame.a xooo Amv.a xoae onxeo Amvoooaoaam aaaa amen;
a

ZOHBHQZOU :Hme qu<filmm9<3 Qz< mm>e mm>OO
BmmmOh >m AmmnocH :HV ZOHB<mHmmZ<mHOQ<>m J<DZZ< A<EOE

Qa Mdm<k .

Table 11. Monthly hydrologic budgets for six local study areas utilizing
changes in soil moisture status predicted by Thornthwaite's
formula, 1961—62 water year.

71

 

 

NET TOTAL TOWAL
A0011
STUDY AREA Ppt Rg R0 81‘‘ R8 615 ST S‘ 5. SP lg Ed
(N’anrn 1061
0040(«) 3.12 3.00 3,20 0.00 2.60 0.13 0.62 >H1.31 11 '1 I) '
11.11. (11) 11.12 0.62 1.014 0 0.62 2.110 1.201 —0. 16 0 0 1.1311 '
1 VIHV( 1 3.20 1.x7 2,17 0.16 1.71 1.20 1.55 -0.76 U H 1L '
1 Plht(h) 3.26 0.23 0.01 0 0 20 3.03 2.28t -0.71 0 0 0,70 '
N. Pine(x1 3.32 1.37 2.36 0.12 1 25 1.03 1.71L -1.11 0 0 0.40 :
1:, 1*.unt-( 11) :1. .12 11. 1(3 1 .1115 11 11 113 2 .1111 2 -""1 —11 .1111 11 11 1 . 11:
Novrunuh HRH
Hdwd(s) 3.56 2.61 2.28 0,07 2.57 0,02 0.00 +0.20 0 0 0 -
05k (0) 3.56 3.83 1.06 0 3.83 -0.27 0.60t +2.77 0 0 0.73 -
.1. Pine(i) 3.57 2.81 1.83 0,01 2,50 0.73 0.77 -H).HN 11 0 0 -
J. 1'n1c(01 3.37 2.03 11.01 0 2.03 11.61 1.00, 12.02 0 0 (1.30 -
R. Pine(S) 3.41 3.10 1.37 0.01 3.00 0.31 0.71. +1.52 0 0 0 -
R. l'lnc(ln 15.1I 2.84 1,06 0 2.81 11.60 11.70, -+|.78 0 0 1.06 ’
DHCIAHHHifiURiI
Hdwd(s) 2,70 0.68 1,16 0 0,68 0 0 —0.48 0 2.11 0 '
Oak (D) 2.70 0.01 1.13 0 0.01 0 0 -0.22 0 2.61 0 -
J. Pine(S) 2.36 0.65 1.30 0 0.65 0 0 -0.74 0 1.71 0 -
J. Pine(D) 2.36 1.31 0.80 0 1.04 0 0 -H1.45 0 1.41 0 -
H. Pine(S) 2.18 0.72 1.37 0 0.72 0 0 -0,65 0 1.66 0 -
H. Pine(D) 2.18 1.38 1.15 0 1.38 0 0 +0.23 0 2.06 0 -
JANUARY 1062
Hdwu(s) 3.64 0.20 0.50 0 0.20 0 0 -0 30 0 5.55 0 -
Oak (0) 3.61 0.21 0.70 0 0.21 0 0 -0.58 0 6.04 0 -
J. Pine(S) 3.54 0 24 0.76 0 0.24 0 0 -0 52 0 5.01 0 -
J. Pine(D) 3.51 0.30 0.67 0 0.30 0 0 -0 37 0 4.65 0 -
R. Pine(S) 3.34 0.20 0.74 0 0.29 0 0 -0.45 0 4.71 0 '
R. Pine(D) 3.34 0.36 0.70 0 0.36 0 0 -0.43 0 5.04 o -
FEBRUARY 1962
0000(5) 2.52 0.60 0.70 0 0.60 0 o —0.10 0 7.47 0 '
Oak (D) 2.52 0.35 0.72 0 0.35 0 0 -0 37 0 8.21 0 -
J. Pine(S) 2.64 (IAIN (1.31 0 11.06 0 0 -0.26 0 7.57 0 -
J. Pine(D) 2.64 0.16 0.65 0 0.16 0 0 —0.49 0 7.13 O -
R. Pine(S) 2.38 0.30 0.71 0 0.30 0 0 -0.44 0 6.79 0 -
R. Pine(D) 2.18 0.23 0.72 0 0.23 0 0 -0.49 0 7.19 0 -
MARCH 1062
Hdud(S) 1.40 6.04 3.71 0 6 04 0 0 +3 23 0 ---2.02--— -
Oak (0) 1.49 6.86 1.12 0 6.86 0 0 +5 74 0 2.84 '
J. Pine(S) 1.29 3.08 U.HH 0 5.08 0 0 + 90 0 3.78 -
J. Pine(D) 1.29 3.26 0.84 0 3.26 0 0 +9 42 0 5.16 -
R. Pine(s) 1.20 5.10 -1.61 0 5.43 0 0 +7 07 o 2.56 -
R. Pine(D) 1.20 3.05 0.00 0 3.05 0 0 +9 15 0 5.34 -
ArMIL 1962
0000(5) 1.24 1.07 2.00 0.01 1 03 0.17 0.21 -0.97 0 0 0 2.02
Oak (0) 1.24 1.73 1.56 0 1.73 -0.40 0.20t +0.17 0 0 ---2.15---
J. Pine(S) 0,88 1.33 1.68 0.24 1.00 -0,15 0.74t -0.59 0 0 0 2.83
J. Pine(D) 0.88 4.32 0.78 0 4.32 —0.44 0.70t +3.54 -0.50 0 0 1.52
R. Pine(S) 0.88 1.82 0.53 0.13 1.69 —0.94 ”-631 +1.16 -0.25 0 0 1.37
R. Pine(D) 0.88 1.57 0.78 0 4.87 -3.09 0 7“1 +4.09 -0 50 o 0 1.15
MAY 1962
Hdwd(S) 1.19 1.07 1.00 0.82 0.25 0.12 0.94 -1.65 0 0 0 -
Oak (0) 1.19 1.61 2.11 0 1. 1 -0.42 0.90. -0.80 0 0 0 0.83
J. Pine(S) 1,00 0.65 1.52 0 36 0.09 1.25 3.00 -1.43 -1.19 0 0 -
J. Pine(D) 1.00 0.32 1.27 0 0.32 1.58 3.00 —0.05 —1.42 0 O -
R. Pine(S) 1.02 0.18 1.51 0.61 -0 16 1.44 3.00 -1.67 -0.92 0 0 -
R. Pine(D) 1.92 0,68 1.11 0 0.68 1.24 3.00 -0.46 -1.76 0 0 -

 

NET TOTAL TOTAL

 

 

A .ET
STUDY AREA Ppt RB 80 ET“ R8 ET. ET 8‘ 3. BF I' Ed nn
JUNE 1962
Hdwd(S) 4.36 2.39 1.25 2.35 0.04 1.97 4.32 -l.21 ' ' ' '
Oak (D) 4 36 0 79 3.07 0 0.79 3.57 3.88 -2.28 —0.31 ' “ '
J. Pine(S) 3 80 l 85 1.87 0.50 1.35 1.95 2.76 -0.52 —0.31 ' ' '
J. Pine(D) 3 80 0.81 2.11 0 0.84 2.96 3.27 -1.27 -0.31 ' ‘ ’
R. Pine(S) 4 32 0 88 1.87 0.65 0.23 3.44 4.40 -l.64 ~0.31 ' ‘ ‘
R. Pine(D) 4 :2 0.34 2.02 0 0.34 3.98 4.29 -1.68 -0.31 ’ ‘ ‘
JULY 1962
Hdwd(S) 3.14 1.67 1.72 1.32 0 1.47 2.79 -l.37 ‘ ‘ ' '
Oak (D) 3.14 0.29 2.82 O 0 )9 2.85 3.41 -2 53 -0.56 ' ' '
J. Pine(S) 2.94 0.71 2.23 0.42 0.29 2.23 2.93 -1.94 -0.28 ‘ ’ '
J. Pine(D) 2.94 0.37 2.96 0 0.37 2.57 3.13 -2.59 -0.56 ' ' '
R. Pine(S) 2.66 0.07 1.51 0.71 -0.64 2.59 3.86 -2.15 -0.56 ' ' '
R. Pine(D) 2.66 0.12 1.92 0 0 12 2.54 3.10 —l.80 -0.56 ‘ ’ '
AUGUST 1962
Hdwd(S) 2.91 0.44 1.10 0.23 0.21 2.47 2.70 -O.89 0 - - -
Oak (0) 2.91 0.02 2.08 0 0.02 2.89 3.21 -2.06 -O.32 - - -
J. Pine(S) 2.82 0.37 1.75 0.06 0.31 2.45 2.83 -l.44 -0.32 — - -
J. Pine(D) 2.82 0.35 1.87 0 0.35 2.47 2.79 -l.52 -0.32 - - -
R. Pine(S) 2.76 0.19 1.49 0.35 -0.16 2.57 3.24 -1.65 -0.32 - - -
R. Pine(D) 2.76 0.24 1.89 0 ).24 2.54 2.86 -l.65 -0.32 - - -
SEPTEMBER 1962
Hdwd(S) 3.40 0.89 1.07 0.07 0.82 2.51 2.58 -0,25 0 - _ _
oak (D) 3.40 0.20 1.13 0 0.20 3.20 2.74 -0.93 +0.46 - - -
J. Pine(S) 3.24 0.18 1.28 0 0.18 3.06 2.60 -l.10 +0.46 - _ _
J. Pine(D) 3.24 0.44 1.96 0 0.44 2.80 2.34 -l.52 +0.46 - - -
R. Pine(S) 3.32 0.18 1.38 0.22 -0.04 3.14 2.90 -l.42 +0.46 - _ -
R. Pine(D) 3.32 0.24 2.14 0 0.24 3.08 2.62 -l.90 +0.46 - - -
TOTAL (1961-62 Water Year)
Hdwd(S) 33.66 21.58 19.74 5.29 16.29 10.06 15.35 -3.45 0 - - 2,02 11,37
Oak (D) 33.66 17.41 18.97 0 17.41 15.41 16.14 -1.55 -0.73 - - 0,33 15,97
J, Pine(S) 32.24 15.85 18.00 1.98 13.87 13.56 17.18 -4.12 -l.64 - - 2.83 20.01
J. Pine(D) 32.24 14.86 15.85 0 14.86 16.36 18.51 -0.99 -2.65 - - 1.52 20.03
R. Pine(S) 31.92 14.83 13.43 2.83 12.00 16.33 20.45 -l.43 —l.90 - - 1.37 21 82
R. Pine(D) 31.92 14.51 15.55 0 14.51 16.78 19.27 -l.04 -2.99 — - 1,15 20,42
5g = periodic change in ground water storage
S5 = periodic change in soil moisture storage
SP = snow pack (8.8.)
WK = gravitational water in vndosc zone

Ed = dormant season evaporation 1055

t = estimated ET values from Thornthwaite formula

Table 12. Monthly hydrologic budgets for six local study areas utilizing
changes in soil moisture status predicted by Thornthwaite's
formula, 1962-63 water year.

72

 

NET . TOTAL 70741
as

 

 

 

STUDY AREA Pp. Rs 80 21‘ ET. ET 8‘ 5I sp 8‘ Ed 455.0
()CI‘UI 11;” "1163
Hd.d(s) 3.42 1.92 0.50 0.01 1.91 1.50 1.51 +1.41 0 0 o -
Oak (D) 3.42 0.07 0.65 0 0.07 3.35 1.28 -0.58 +1.51 0 0.56 -
J. Pine(S) 3.70 0.49 0.85 0 0.49 3.21 1.70 -o.36 +1.13 0 0.38 -
J_ pine(D) 3.70 0.18 1.06 0 0.18 3.52 2.01 —0.88 +1.18 0 0.33 -
R. Pine(S) 3.70 0.26 1.06 0.18 0.08 3.44 2.11 -0.98 +1.23 0 0.28 -
R Pine(D) 3.70 0.16 1.37 0 0.16 3.54 2.03 -1.21 +1.08 0 0.43 -
NOVEMBER 1962
Rd.d(s) 0.56 0.41 0.60 0.16 0.25 0.15 0.31 -0.35 o 0 0 -
Oak (D) 0.56 0.82 0.76 0 0.82 -0.26 0.30t +0.06 0 o o -
J. Pine(S) 0.58 0.46 1.00 0 0.46 0.12 0.50t -0.52 0 0 0 —
J. Pine(D) 0.58 0.41 1.06 0 0.41 0.17 0.50t -0.65 0 0 0 -
R. Pine(S) 0.56 0.34 1.00 0.01 0.33 0.22 0.50t -0.67 0 0 0 -
R Pine(D) 0.56 0.49 1.54 0 0.49 0.07 0.50t -1.05 0 o o -
DECRMUUR 1962
Hdwd(S) 3.37 0.85 0.65 0 0.85 - - +0.20 0 2.52 0 -
Oak (0) 3.37 0.28 0.41 0 0.28 - - -0.13 0 3.09 0 -
J. Pine(S) 3.26 0.09 0.65 o 0.09 - - -0.56 0 3.17 o -
J. Pine(D) 3.26 0.10 0.77 0 0.10 - - -0.67 0 3.16 o -
R. Pine(S) 3.55 0.35 1.02 0 0.35 - - -0.67 o 3.20 o -
R. Pine(D) 3.55 0.12 1.03 0 0.12 - - -0.91 0 3.43 0 -
JANUARY 1963
Hdwd(S) 2.63 0.71 0.72 0 0.71 - - —o.01 0 4.44 0 -
05k (D) 2.63 0.34 0.34 0 0.34 - - 0 o 5.38 o -
J. Pine(S) 2.61 0.18 0.54 0 0.18 - - -o.36 0 5.60 0 -
J. Pine(D) 2.61 0.25 0.65 0 0.25 - - -0.40 0 5.52 0 -
R. Pine(S) 2.78 0.14 0.68 o 0.14 - - -0.54 0 5.84 0 -
R. Pine(D) 2.78 0.08 0.76 0 0.08 - - -0.68 o 6.13 o -
FEBRUARY 1963
Hdwd(S) 1.43 0.08 0.56 o 0.08 — - -o.48 0 5.79 o -
Oak (0) 1.43 0 0.34 o 0 - - -0.34 0 6.81 0 -
J. Pine(S) 1.45 0.02 0.43 o 0.02 - - -0.41 0 7.03 0 -
J. Pine(D) 1.45 0 0.67 o 0 - - -0.67 0 6.97 o -
R. Pine(S) 1.57 0.12 1.02 0 0.12 - - -0.90 o 7.29 0 -
R. Pine(D) 1.57 0 0.72 o 0 — - -0.72 0 7.70 o -
MARCH 1963
Hdwd(S) 3.00 7.46 2.10 o 7.46 - - +5.36 0 0 o 1.33
Oak (0) 3.00 5.36 0.42 o 5.36 - — +4.94 0 o ---4.45---
J. Pine(S) 3.28 5.12 0.34 0 5.12 - - +4.78 0 ------------- 5.19--—
J. Pine(D) 3.28 1.42 0.85 0 1.42 - - +0.57 0 ------------- 8.83---
R. Pine(S) 3.30 5.44 0.70 0 5.44 - - +4.74 0 ------------- 5.15---
R. Pine(D) 3.30 1.32 0.94 o 1.32 - - +0.38 0 ------------- 9.68---
APRIL 1963
Hdwd(S) 2.60 2.33 3.20 0.02 2.31 0.27 0.29 -0.89 o 0 0
Oak (0) 2.60 2.86 0.79 0 2.86 -0.26 0.30 +2.07 0 o ---3.89-——
J. Pine(S) 2.33 3.19 1.12 0.06 3.13 -0.86 1.40. +2.01 0 0 o 2.93
J. Pine(D) 2.33 6.20 0.72 0 6.)0 -3.87 1.40. +5.48 0 0 ---3.56---
R. Pine(S) 2.54 3.01 1.13 0.05 2.96 -0.47 1.40. +1.83 0 0 0 3.28
R. Pine(D) 2.54 6.55 0.82 0 6.55 -4.01 1.40l +5.73 0 0 ---4.27---
MAY 1963
Hd‘d(S) 2.19 1.80 2.39 0.34 1.46 0.39 0.73 -0.93 0 - - 1.33
Oak (0) 2 19 2.39 1.74 0 2.39 -0.20 0-701 +0.65 0 - - 2.99
J. Pine(S) 2 21 1.49 1.10 0.53 0.96 0.72 1.65 -0.14 -0.40 - - 2.93
J. Pine(D) 2 21 1.48 0.96 0 1.48 0.73 1.70 +0.52 -0.40 - - 3.00
R. Pine(S) 2.16 0 86 0.29 0.32 0.54 1.30 1.62 +0.25 0 - - 3.28
R. Pine(D) 2.16 2.16 1.20 0 2.16 0 1.60 +0.96 -0.40 - - 3.07

 

NIT TOTAL TOTAL
“6

 

 

 

 

STUDY AREA apt R‘ 80 IT‘ IT. ET 8' 8. 89 v‘ Ed Ann.EI

JUNE 1963
Hded(S) 1.25 0.66 1.68 1.38 -0.72 0.59 2.97 -2.40 -1.00 - — -
Oak (D) 1.25 0.24 2.58 o 0.24 1.01 3.34 -2.34 -2.33 — - -
J. Pine(S) 1.12 0.01 1.49 0.74 -0.73 1.11 3.41 -2.22 -1.59 - — -
J. Pine(D) 1.12 o 1.56 0 0 1.12 3.45 -1.56 -2.33 - - -
R. Pine(S) 1.07 0.02 1.32 0.15 -0.13 1.05 3.53 -1.45 -2.33 - - -
R. Pine(D) 1.07 0.28 1.26 o 0.28 0.79 3.12 —0.98 -2.33 - - -

JULY 1963
HdId(S) 3.67 1.78 1.20 1.78 0 1 89 3.67 -1 20 0 - - -
Oak (D) 3.67 0.29 2.56 o 0.29 3 38 3.49 -2.27 -0.11 - - -
J. Pine(S) 3.75 1.25 1.90 0.64 0.61 2 50 3.14 -1.29 0 - - -
J. Pine(D) 3.75 0.42 1.72 0 0.42 3.33 3.44 -1 30 -0.11 - - -
R. 9156(5) 3.60 0.68 1.50 0.29 0.39 2.92 3.32 —1.11 -0.11 - - - ;
R. Pine(D) 3.60 0.53 1.82 0 0.53 3.09 3.18 -1.29 -0.11 - - - 3

AUGUST 1963 '
fldId(S) 3.56 0.43 1.12 0.28 0.15 3.13 3.41 -o.97 o - - - .
Oak (D) 3.56 0.24 2.21 o 0.24 3.32 3.32 -1.97 0 - - - “
J. Pine(S) 3.39 0.41 1.49 0.08 0.33 2.98 3.06 -1.16 o — - - -' ‘
J. Pine(D) 3.39 0.34 1.75 o 0.34 3.05 3.05 -1.41 o - - - -
R. 9154(8) 3.56 0.10 1.16 0.20 -0.10 3.46 3.66 -1.26 0 - - - hr
R. Pine(D) 3.56 0.12 1.69 0 0.12 3.44 3.44 -1.57 o - - -

SEPTEMBER 1963
HdId(S) 2.06 0.31 0.70 0.17 0.14 1.75 1.92 ~0.56 0 - — -
Oak (0) 2.06 0 1.30 0 o 2.06 2.06 —1.30 0 - - —
J. P1ne(8) 2.19 0.02 0.98 0 0.02 2.17 2.17 -0.96 0 - - -
J. Pine(D) 2 19 0.18 1.03 o 0.18 2 01 2.01 -O.85 0 - - —
R. Pine(S) 2.37 0.01 1.20 0.07 -0.06 2.36 2.43 -1.26 0 - - -
R. Pine(D) 2.37 0.08 1.25 o 0.08 2.29 2.29 -1.17 0 - - -
' TOTAL (1962—63 Water Year)
Hded(S) 29.74 18.74 15.42 4.14 14.60 - 14.81 -o.82 —1.0o - - 1.33 16.14
Oak (0) 29.74 12.89 - 14.10 0 12.89 - 14.79 -1.21 -o.93 - - 2.99 17.78
J. Pine(S) 29.87 12.73 11.89 2.05 10.68 - 17.06 —1.19 -0.89 - - 2.93 19_99
J Pine(D) 29.87 10.98 12.80 0 10.98 - 17.56 -1.82 -1.66 - - 3.00 20.56
R. Pine(S) 30.76 11.33 12.07 1.27 10.06 - 18.57 -2.02 -1.21 - - 3,23 2..35
R. Pine(D) 30.76 11.89 14.40 0 11.89 - 17.56 -2.51 -1.76 - - 3.07 20.63
S8 : periodic change in ground water storage
3 =

s periodic change in soil moisture storage

SP = anoe pack (I.E.)

‘
11

g gravitational water in vadose zone
Ed = dormant season evaporation 109a

estimated ET values from Thornthwaite formula

9
1|

73

volumes of percolating moisture which has not yet reached the water table.
This approach is admittedly arbitrary. However, the computed values for
the period from May to September in both water years was little affected
by the problem of delayed percolation. The patterns of evapotranspira-
tion shown in Figure 21 are illustrative of the early use of water by
conifers and the concentration of evapotranspiration into the three
summer months in the hardwood forests.

A more detailed picture of the comparative rates of moisture use
in wet and dry periods was obtained by analysing each period between
recharge events in the three shallow water-table areas. Theoretically,
the unsaturated soil moisture status is at the same field capacity level
at the end of such a recharge event. If this were actually the case, a
storm coming only one or two days after another recharge producing storm
should produce a recharge input nearly equal to the total throughfall.
Many such instances during the two growing seasons demonstrated that
this was not strictly true. Soil moisture is evidently depleted and
recharged in irregular horizontal patterns, so that portions of the area
produce recharge while others still exhibit soil moisture deficits.
Reports on the variations in soil moisture conditions during the growing
season have shown by direct measurement that the opportunity for ground-
water recharge varies over even a localized area (Striffler, 1961; Lull

and Axely, 1958).

In spite of these discrepancies, a pattern of evapotranspiration
in the three cover types emerged from this analysis (Figure 22). During
both growing seasons, the conifer types began rapid evapotranspiration

use in mid-April. The hardwood forest moisture use for ET was delayed

four to six weeks, until the time of leaf development. During late

Figure 21. Monthly evapotranspiration for six local study areas.

74

 

 

new and

N00.

www—
LO
2 com 25 .2.

5... .5... >62 :5 hWOL-w >62 80
1‘1 1

0

Ex. he: :7. P

 

«col

.4.

l a I ~
I II I
.. /\V.n. .

. 3.3
.../...
a. . ..../
\fi far/.0 .
.....v, 1..
J.)-

IV:...[.J.: \..( .
R: 65.". com 111111
“02:0 tom
A995. 3.02. I ..... .1

«92:1 xoeal.l.l

8v zoo-1:-

«9 30332..

 

//v

......2... t N.

1
1
’I
I
O I,
’1

.. I
... ....N /./...v
.. . / ...
.. ..M K,
.. ./’
§ 6
.. .3
.\.\:.4 .
.. ...M n

a:

. ..
.
..
I.

z

¢ N
(WWW/$811901) NOILVU ldSNV 8.1.06 VAB

 

 

 

Figure 22. Daily evapotranspiration for three shallow waterwtable study
areas, average rates for periods between grounduwater recharge
events.

75

 

 

 

 

 

 

 

 

 

 

 

! ‘
p

AUG

May Jun Jul

 

 

 

 

 

 

 

 

 

 

   

Hardwood (S)

—-—-— Jack Pane (S)
Red Pine (5)
r §1
1 2'1

-' 4.-

 

.....r...‘----- --

   

M9309

May Jun Jul

 

Apr

Feb Mar

 

 

 

 

 

.15

r r11114
p
n: A ................... r113
. r "m
i a ..
. M "n
... u .. .... u"
11 _ 2 u
6 n R 6 u L
9 - 9 .u
l u n l u n
u M "m x
.n
. c .u c
n e
u m “u 0
Pl P.
1— . W ..m W
m 111 .— N 1|111111111-1.». N
rwldL 61.22". ...-....
11 1. ....... .....U..." c u d
a O m n O
m 0 5 O O

.230 \ 62.0.5

zofi<mEmr<mkoa<>m

 

 

June and early July, the maximum ET rate was greatest in the Hardwood
stand. In the autumn of 1962, a higher rate of moisture use was evident
in the conifers, although the late November period showed use in the
Hardwood area to be slightly higher. This is but one example of the
irregularities in the calculated ET pattern which arise from the soil
moisture deficits which remained at the time the periodeending recharge
occurred. Many of the irregularities disappeared when the periodic ET

losses were accumulated to obtain the annual totals (Figure 23). The

 

apparent lag in ET under Jack Pine in July is another example of the

soil moisture deficit error. Both the lower dormant season ET rates "
and the shorter growing season are evident in the slopes of the cumulae

tive curves for the Hardwood cover type.

No estimate of soil moisture def1c1ts was included in the latter
periodic ET calculation. The greatest resulting error occurred at the
end of the first water year, and at the beginning of the second. Accord~
ing to the Thornthwaite calculation, the 1961~62 water year ended with
a soil moisture deficit of over two inches. If measurements of this
deficit were available, the rate of moisture use for late September,
1962, would be greater and those for October, 1962, would be lower.

The greatest effect would be in the Red Pine (shallow) area with the
deepest water table and the greatest opportunity for a soil moisture
deficit to develop. A somewhat lesser deficit would be expected in
the Jack Pine (shallow) area and very little deficit in the Hardwood

(shallow) area.
In the absence of in situ measurements of the soil moisture

status during the study period, the best statement which can be made

on evapotranspiration is: the patterns of evapotranspiration use

Figure 23. Cumulative evapotranspiration for three shallow water—table
areas based on computed evapotranspiration between ground-
water recharge events.

77

.........

 

 

20

10

(Area Inches)

CUMULATIVE EVAPOTRANSPIRATION

 

1961-2

.0
’0
..

    

 

 

1962-3

 

 

 

Nov Dec

Jan Feb

Mar Apr

May Jun Jul

M95013

 

 

78

follow those shown in Figures 21 and 22, although the total annual conw
sumptive use is probably in excess of the indicated amounts. These annual

totals closely approximate those shown in Table 10.

Ground-Water Yield to Seepage Flow

 

An estimate of the rate of waterwtable recession caused by the
drainage of the aquifer by seepage flow was basic to the computation of
all the previous water budget items. These recess1on rates were deter~
mined for a single point in the extensive water table aquifer, the
individual recording well location. In the budgeting operation, daily
and weekly changes in storage due to seepage flow were computed by multi~
plying the predicted recession rate by the appropriate specific yield
value. These HrunoffH values were accumulated for each month and for
each water year (Table 13).

Total annual seepage flow from the six local study areas was
generally in the same ratio as net recharge. The Hardwood (shallOW)
area contributed the greatest amount to seepage flow. In 1961 62, the
Oak area was the second largest contributor. The Jack Pine (shaIIOW)
area was highest of the conifer areas in 1961 62, but lowest in 1962 63.
This difference was due to the high wateratable conditions at the star:
of the 1961—62 water year. During the fall of 1961, this Jack PIHF area.
located near the water—table divide, was drained rapidly by seepage flow,
In the 1962—63 water year, the water table was lower in this area in the

dormant season and the seepage loss was much less.
All areas except the Red Pine (deep) showed a net loss from

storage during the period from October 1, 1961, to October 1, 1962

(Figure 12). During the second water year, this Red Pine (deep) area

TABLE 13

MONTHLY LOSSES FROM AQUIFER STORAGE TO SEEPAGE FLOW,
BY COVER TYPE AND DEPTH (Area Inches)

Hardwood Oak Jack Pine Jack Pine Red Pine Red Pine
(Shallow) (Deep) (Shallow) (Deep) (Shallow) (Deep)

 

 

1961~62 Water Year

 

 

 

 

Oct. 2.26 1.08 2.47 0.94 2.36 1.06
Nov. 2.28 1.06 1.82 0.91 1.57 1.06
Dec. 1.16 1.13 1.39 0.89 1.37 1.15
Jan. 0.59 0.79 0.76 0.67 0.74 0.79
Feb. 0.70 0.72 0.34 0.65 0.74 0.72
Mar. 3.71 1.12 0.88 0.84 ~1.64 0.90
Apr. 2.00 1.56 1.68 0.78 0.53 0.78
May 1.90 2.41 1.52 1.27 1.51 1.14
Jun. 1.25 3.07 1.87 2.11 1.87 2.02
Jul. 1.72 2.82 2.23 2.96 1.51 1.92
Aug. 1.10 2.08 1.75 1.87 1.49 1.87
Sept. 1.07 1.13 1.28 1.96 .£;2§ _2;11
Annual 19.74 18.97 18.00 15.85 13.43 15.53
1962w63 Water Year
Oct. 0.50 0.65 0.85 1.06 1.06 1.37
Nov. 0.60 0.76 1.00 1.06 1.00 1.54
Dec. 0.65 0.41 0.65 0.77 1,02 1.03
Jan. 0.72 0.34 0.54 0.65 0.68 0 7
Feb. 0.56 0.34 0.43 0.67 1.02 0.72
Mar. 2.10 0.42 0.34 0.85 0.70 0.94
Apr. 3.20 0.79 1.12 0.72 1.13 0.82
May 2.39 1.74 1.10 0.96 0.29 1.20
Jun. 1.68 2.58 1.49 1.5 1.32 1-26
Jul. 1.20 2.56 1.90 1.72 1.50 1.82
Aug. 1.12 2.21 1.49 1.75 1.16 l 69
Sept. 0.70 1.30 0.98 1.03 1.20 1.73
Annual 15.42 14.10 11.89 12 80 12.07 14.10

79

80

had a greater storage loss than the other areas. This storage loss, plus
the net recharge for the year, resulted in a greater seepage flow than
occurred in any area except the Hardwood (shallow).

The timing of maximum seepage flow losses in the six study areas
provides some measure of their relative contribution to streamwflow
during the year. The deeper water—table areas and those located most
distant from the streams have later maximum recession from seepage
(Figure 24). Continual drainage during the Winter months provides a
steady contribution to stream—flow during the snowpack period.

In the Red Pine (shallow) area, the proximity of open lands which
had earlier snow melt than the conifer area, produced subsurface inflow
during early spring, In March. 1962, this inflow was sufficient to more
than counterbalance the drainage from the area. The net seepage flow
for that month resulting in a gain in storage which was recorded as
negative seepage flow. As was explained in the description of the study
area, a drainage ditch which conducts high waterwtable flow northeastward
toward Pine Creek, passes by the eastern border of the Red Pine (shallow)
study area. There are numerous hardwood and bog~type swamps along the
groundmwater basin divide which runs southeast from the Udell Hills.

When these swamps are flooded in the spring, surface flow is produced

in this ditch to the extent required for complete influent seepage. In
1962, surface flow extended to the north section line of Section 20, I21N.
Rl4w. Influent seepage produced an addition to groundwwater storage

beneath the Red Pine (shallow) study area. The net seepage loss from
storage was reduced accordingly during the month of April.
All areas are affected by the evapotranspiration rates in shallow

water—table lands. The Oak (deep) area is particularly influenced by the

Figure 24. Monthly losses from aquifer storage to seepage flow for six
local study areas.

81

HARDWOOD (3)

new.

     

2

a z/

JACK PINE (S)

 

. .... + m. z
nose... no.3

  

 

RED PINE (S)

 

Loaf“;

3.5205

to...

4

1m.

  

RE D PINE (D)

/

hoaxeaz

304m mw<mmmm

   

O'ONDJFHAMJJASONDJFMA

I963

l962

accelerated water—table gradient which results when water levels in the
hardwood swamp to the north are lowered in June and July. Whereas rem
charge is apparently a function of cover type and waterutable depth,
seepage flow is dependent on the position of the particular area in the
basin, the recharge and discharge behavior in adjacent portions of the
aquifer and the timing of percolation recharge within the study area
itself. When these complexities are considered, seepage flow losses
from these six local sections of the aquifer cannot be simply added to
obtain a basin drainage volume. Use of these local analyses of seepage

loss is nevertheless valid for derivation of the local water budget.

CHAPTER VIII

DISCUSSION OF WATER BUDGET RESULTS

Analysis of the two water years, 1961m62 and 1962~63 has produced
measures of comparative recharge of ground—water supplies beneath three
contrasting forest cover types. In general, the deciduous forest types
produced greater amounts of net groundwwater recharge than the coniferous
types. In the first water year (1961w62), the mean addition to groundw
water supplies in the four pine areas was 18 percent less than the mean
for the two hardwood areas. In the l962~63 water year, the conifers
yielded 21 percent less than the hardwoods. The mean differences for
the two years was 2.9 inches (15.3 vs. 12.4).

The mean difference in winter recharge, attributable to snow melt.
was 0.9 inches. In the Jack Pine areas where crown densities remained
high throughout the year, the decrease was greater, 1.2 inches less for
the mean of the two water years. Under this forest type, winter and
spring melt recharge differences accounted for 46 percent of the differw
ence in total annual net recharge. In the Red Pine areas this portion
of the water year accounted for only 16 percent of the mean annual dif~
ference.

Cumulative curves of both annual recharge and annual evapotransw
piration illustrated the effects of longer transpiration periods in the

conifer stands (Figures 16 and 23). The greater use of moisture for

evapotranspiration in the Red Pine~0ak areas is presumeahly due to the
deeper rooting depths and, hence, larger soil moisture storage capacities

83

84

in the root zone. Investigations of the comparative rooting depths in the
Udell Experimental Forest (Figure 17) support the findings of DeByle and
Place (1959). In sand soils, jack pine roots are concentrated in the
surface six to eight inches of soil. Mixed Oak and Red Pine stands had
deeper rooting which help to explain both the high evapotranspiration
rates and the ability of these stands to obtain water from the saturated
zone at greater depths than the other types.

An earlier investigation by the author of s011 moisture depletion
patterns in red pine and oak forests near the Udell Experimental Forest,
showed that an opportunity existed for groundnwater recharge differences
during the spring and fall months (Urie, 1959). Using the soil moisture
levels which were measured during the 1958 growing season, precipitation
during the period was found to exceed field capacities on sufficient
occasions to produce 3.3 inches of recharge under oak forests and only
1.5 inches of recharge under red pine plantations. The greatest differ;
ence during the April to October study period occurred prior to leaf
development in the oak forests. The present study substantiates these
indications; recharge was greater in the hardwood forests in both spring
periods.

The mean annual difference in net recharge during the snow~free
periods was 2.7 inches between the two red pine areas and the two hard-
wood areas. Thus, 82 percent of the recharge effect occurred during the

rainfall months. A lesser difference, 1.5 inches, was measured between
hardwoods and jack pine. Since precipitation was reasonably well dis»
tributed throughout the two-year period and both years were near the
mean annual rainfall, these recharge effects of forest cover may also

approach the long—term norm. Continued measurements will be needed to

 

substantiate this.

The effects of groundwwater depths were illustrated by the timing
of recharge and the comparative rate of evapotranspiration drain. Because
of differences in soil profile development and water~table regimen in the
three shallow water—table plots, the observed differences in root penetram
tion cannot be explained simply as species differences. Regardless of
these interacting effects, there was a clear relationship between the
moisture drained and ground-water depths. Sharp cutoff levels were
found in the Hardwood and Jack Pine (shallow) areas. In the Red Pine
(shallow) area, the diurnal acceleration of the rate of water-table
decline continued at a very slow but measureable pace to the lowest well
levels occurring during the study period.

The short history of waterwtablewlevel trends which has been
accumulated in the Udell area (Figure 12) shows that the mean annual
ground-water levels in the Red Pine (shallow) were lower than in the
other shallow types. The deeper root development in this forest type
may be due to these lower levels. In the deep waterwtable areas, the
red pine rooting depth is also greater than in jack pine under similar
drainage conditions. Unfortunately, there was no available red pine
plantation on AuGres soils, such as existed in the extensive Jack Pine
(shallow) area. The explanation for different rechargewevapotranspiraw
tion balances between these two pine types is not at hand. It is
apparent, however, that the two areas are different in their hydrologic
behavior. The Jack Pine (shallow) pattern of ET drain on ground—water
is more like that of the Hardwood (shallow) area. Both have short

periods of rapid use while ground-water levels are high. The conifer

areas are alike in showing earlier ET drain than the Hardwoods.

86

Both pine types utilized more water than hardwoods for evapotrans—
piration in the shallow water—table position, 0.7 inches in Jack Pine and
2.2 inches in Red Pine. No definite effect was found between the two
hardwood study areas. It was apparent from observations of soil condi—
tions that soil moisture deficits were rare in the Hardwood (shallow)
area. The inter-recharge evapotranspiration patterns (Figure 22) show
that, for brief periods early in the summer, the daily rate of evapo~
transpiration is very high in this type. The seasonal decline in ground_
water levels must result in a drop in the withdrawals from the saturated
zone, for the diurnal fluctuations were almost nonedetectable by August
of both years. Satterlund (1960) found northern hardwood forests in
Michigan's Upper Peninsula could use groundewater only to 30—inch depths.
This study was also carried out in an area of shallow water tables where
root penetration was doubtless inhibited by saturated soils during much
of the year.

There are species, stocking level and distribution differences,
between the two hardwood study areas (Table 1). Since these differences
are also characteristic of the change from well—drained to poorlywdrained
sands on the entire Udell Forest, the water budgets, if true in them“
selves, are valid measures of these two drainage conditions under exist~

ing native forests.

The two water—year periods which were included in this analysis
did not differ greatly in total precipitation. Despite this similarity,
the water yield which was obtained as net ground water recharge was 20
percent less in the second water year (mean of all six areas). The

recharge difference is partly due to the difference in antecedent soil

moisture conditions. In 1961, October 1 was a date on which water tables

87

were high and soils were at field capacity levels. The following water
year began with a soil moisture deficit of two or three inches.

In 1961—62, there was normal or above normal precipitation during
the fall and winter months. In l962—63,there was no appreciable recharge
before the snOWpack began to accumulate. A higher proportion of the
annual precipitation fell during the growing season in 1962 63. Accord~
ingly, the rates of total evapotranspiration were similar in both years.
Summer rainfall produced little groundmwater recharge; this was reflected
in the low total for the second year.

Moisture removed from groundwwater by evapotranspiration reduced
the amount of net recharge. In 196l~62, wateretable levels were higher
than in the following year, especially in the autumn and spring. The
rate of ET drain was directly related to the height of the water table.
This tended to reduce the net recharge during the year of higher water
yields. The influence of waterwtable height on the evapotranspiration
vector can be predicted to the extent that annual watermtable trends are
predictable. Drecher (1957) showed a high correlation of waterwtable
levels with a five—year running average of precipitation. In deeper
water-table conditions, the trends may reflect longer term precipitation
conditions, (Wenzel, 1936). The relatively short history of Udell
ground-water conditions corresponds to cumulative departures from normal

precipitation (Figure 25).

Figure 25. Water-table elevation on eastern outwash plain, Udell Experie
mental Forest, in relation to cumulative departures from
normal precipitation.

88

 

m. _ E. 5.9 E E FE

 

 

 

 

t 7

 

</

 

 

/ / .7

F-
i3

 

h

 

 

 

no .33 (4
cozo>2m «Ba... .33)

 

 

 

IWI 035 OAOQD .1395

E

 

 

 

lll'l lllllljll|lll.I

 

 

 

 

 

 

 

522389;. .2502 En... 8.3.630 028.3830

 

 

 

 

_

 

 

 

 

 

CHAPTER IX

S UIVTA‘IAR Y

The objectives of this study were:

1. To develop methods for measuring the hydrologic
budget of definable cover type areas in highly infiltrable
sands.

2. To compare the net water yield (or net ground»
water recharge) and evapotranspiration for three forest
conditions common to the northern part of Michigan’s
Southern Peninsula.

3. To determine the effect of wateretable depth under

these forest cover types on the evaporation loss and, thus,
on the net water yield.

The first objective was met in that a method was developed for
obtaining the water budget for a localized sector of a broad wateretable
aquifer. Once such a method became available, from measurements of
ground—water recharge, seepage flow losses and evapotranspiration drain,
it was then possible to compare cover types now on the lands. 1: will
also be possible to evaluate the hydrologic effec:s of future iores:
management practices. The need for such a method is obVious from the
heterogeneous pattern of cover conditions within the large drainage
basins characteristic of this northern portion of Michigan.

The first step in developing the required analysis was an
objective separation of the components of water~table fluctuations.
Except for temporary rises due to barometric pressure changes, wind
effects (Parker and Stringfold, 1950), and entrapped air beneath a
percolating wetting front (Lee, 1934), recharge by percolated precipitaa
tion water was the only cause for positive changes in well levels.

89

90

Declining well levels resulted from one, and sometimes two,
processes. Seepage flow produced a continual change in storage which
could be predicted from indications of the inflowmoutflow balance in
the local study area. Secondary recessions due to evaporation losses
from the saturated zone occurred during the growing season in shallow
water-table areas.

The separation of these two recession vectors was possible
because of diurnal fluctuations in the rate of water—table decline.

The nocturnal rate of recession should be approximately that predicted
from the slopes of the water table (Figure 9). Agreements between these
independent measures of the recession rate due to seepage flow, indicated
that local recession because of evapotranspiration drain in shallow water»
table areas did not induce appreciably accelerated inflow during the
daylight hours.

On the other hand, low evapotranspiration rates in the Hardwood
(shallow) area, although partially explained by shorter growing seasons
and shallow root penetration, suggested that the seepage flow and evapo~
transpiration drain may not have been entirely separated. With this
exception, the techniques utilized in this analysis prOVided an adequate
method for the separation of the components of waterwtable fluctuations.

Conversion of these separate trends in well levels to a measure

of the volume of water was accomplished by multiplying the head changes
for the well area by the specific yield values.

The method for obtaining the specific yield values for the

shallow water—table aquifers was essentially that described by Olmsted

and Hely (1962) as Method "A." As was pointed out by these authors,

the short-term coefficient of storage, which they called the gravity

91

yield, is ordinarily less than the specific yield. Olmsted and Hely

solved the equation.

Yg : ASg (X-l)

where Yg is their gravity yield (a dimentionless ratio), ASg is the
change in ground-water storage in a given period in area inches, and
AHg is the corresponding increase in ground-water storage in inches, by
computing the integral of base-flow recession during the dormant season.
In the Udell Hills study, no stream—flow data was available. Recharge
inputs were used for the Sg term. The resulting gravity yield values

varied inversely with the degree of soil profile development.

In the deep water-table aquifers, specific yield values were
obtained by draining undisturbed sample cores. Following initial saturae
tion, the cores were drained for 48 hours with the base of the core main~
tained at water-table level. The height of capillary rise in medium
sands is 12-35 centimeters (4.7 to 13.8 inches) (Harr, 1962). Examina—
tion of the undisturbed cores after drainage showed capillary rise in
the Udell samples to be approximately 12 inches. Specific yield deter-
minations were made from the upper and lower three inches of these 24-
inch columns. The upper sample was well above the height of maXimum
capillary rise. The lower sample was almost completely within the lower
20% of the capillary zone where saturation is relatively high (Taylor,

1948).
To the extent which the two-inch diameter cores were actually

representative of undisturbed aquifer materials, acceptable gravity yield

values were determined in this manner.

In these deeper aquifers, where there was only one major cycle of
water-table fluctuations each year, the drainage period was sufficiently
long so that the gravity yield probably closely approximated the specific
yield. In the shallow water-table areas, there were more frequent alter-
nating recharge and drainage cycles. In these areas, the use of a con-
stant specific (or gravity) yield value for a given aquifer layer is less
justified. However, in view of the much greater variations due to the
degree of soil profile development, the estimated mean specific yield
for a daily recharge or recession event was probably little affected by
the minor changes due to time of drainage.

The second objective was the determination of the hydrologic
budget for three representative forest types. In the deep water-table
areas this was accomplished as soon as a method for measuring the volume
of recharge was developed. Periodic evapotranspiration was then deter-
mined from the difference between recharge and precipitation. In shallow
water-table areas, the net recharge remaining after evapotranspiration
from the saturated zone was equated with water yield. Total evapotranSM
piration was the sum of precipitation which did not reach the saturated
zone plus the volume of water removed from the saturated zone.

The precision of the gross and net recharge determinations could
not be checked directly. Comparisons with the periodic water yield and
evapotranspiration predicted by Thornthwaite's empirical formula were
reasonably close in the pine areas. In the hardwood areas, the water
yields were higher in the spring months, indicating that potential evam
potranspiration rates were not reached until full leaf development was
attained.

These analyses showed that dormant season recharge is proportional

to the amount of crown cover during this period. The computed snowwmelt

93

recharge was related to the maximum snowpack accumulated beneath the
crown canopy.

The effects of water-table depth were shown by the well records
from the three shallow water-table plots. Although the overall relation-
ships between cover types were not apparent due to differences in soil
conditions, water-table regimes and rooting depths, there were consistant
decreases in evaporative ground-water losses with lowering water~table
levels. Ground-water depletion due to evapotranspiration ceased when
water—table depths were lowered to 4.5 feet in the Hardwood (shallow)
area and to 5.5 feet in the Jack Pine (shallow) area. The lower limit
of evaporation effect was related to the depth of rooting.

The bookkeeping required for these computations was tedious and
time consuming. This was especially true of the shallow water—table
wells where daily accounting was used. Since the routine is a sequence
of arithmetic operations once the well record is interpreted, the entire
procedure is amenable to machine computation. A logical extension of
the present study will be the development of a computer program to reduce
daily well changes, precipitation, diurnal fluctuations and grid well
observations of water-table slope conditions to periodic water budget
values. Desirable refinements of the empirical methods used here would
be:

1. Incorporation of gravity yield functions which are

dependent on depth of the water table and the time
of drainage.

2. Numerical analysis of current grid well data to obtain
a mathematical solution of the second derivative of
the water-table elevation for use in predicting the
seepage flow recession.

Stallman (1956) has described methods for utilizing well levels

from a regularly spaced grid to obtain positive and negative accretions

94

to storage. Where the transmissibility and specific yield are known,
these methods can be applied directly to obtain periodic water budgets.
At the present time, only localized estimates are available for the
aquifers of the Udell Experimental Forest. Extention of the water
budgets, as computed in the present study, depends on the availability

of these aquifer constants. As these become available, the periodic
water yield will be computed and related to stream—flow in the bordering
channels. The effects of forest cover conditions and ground-water depths

reported here will form an important part of this broader study.

LITERATURE CITED

Childs, E. C.
1960. The nonsteady state of the water table in drained land.
Journal of Geophysical Research. 65: 780-782.

DeByle, N.V. and I. C. M. Place
1959. Rooting habits of oak and pine on Plainfield sand in central
Wisconsin. University of Wisconsin Forestry Research Note
No. 44. Madison, Wisconsin. 2 pp.

Dils, Robert E. and John L. Arend
1956. Snow accumulation under red pine of different stand densities
in Lower Michigan. U.S. Dept. of Agr., Forest Service, Lake
States Forest Experiment Station, Technical Note No. 460,
St. Paul, Minnesota. 2 pp.

Drecher, William J.

1955. Some effects of precipitation on groundwwater in Wisconsin.
Wisconsin Geological Survey Information Circular No. 1.
Madison, Wisconsin. 17 pp.

Ferris, John G.
1959. Ground water. pp. 127-191, In: Hydrology by Chester 0.
Wisler and Ernest F. Brater. John Wiley and Sons, New York.
2nd edition. 408 pp.

Harr, Milton E.
1962. Ground-water and seepage. McGraw—Hill Book Company, Inc.
New York. 315 pp.

Hart, George
1963. Snow and frost conditions in New Hampshire under hardwood
and pines and in the open. Journal of Forestry. 61: 287—289.

Hazen, Allen
1892. Experiments upon the purification of seepage and water at the
Lawrence Experiment Station. Massachusetts State Board of
Health, 23rd Annual Report, 1891. pp. 428—434.

Holstener, Jdrgensen, H.
1961. An investigation of the influences of various tree-species
and the ages of the stands on the level of the ground-water

table in forest tree stands in Bergentved. "Det forstlige
Forsdgovoensen i Danmark." 27: 237w480. Danish with English
summary.

95

96

King, F. H.
1898. Principles and conditions of the movements of ground—water.
U.S. Geological Survey, 19th Annual Report, Part 2. pp. 59-294.

Kittredge, Joseph
1948. Forest influences. McGraw-Hill Book Company, Inc. New York.
394 pp.

Lee, Charles H.
1934. The interpretation of water-levels in wells and test-holes.
Transactions, American Geophysical Union. 15: 540-554.

Lemmon, Paul E.
1956. A spherical densiometer for estimating forest overstory
density. Forest Science. 2: 314-320.

Lull, H. W. and J. H. Axely
1958. Forest soil moisture relationships in the coastal plain
sands of southern New Jersey. Forest Science. 4: 1-19.

Lull, Howard W. and Francis M. Rushmore
1960. Snow accumulation and melt under certain forest conditions
in the Adirondacks. U.S. Dept. of Agr., Forest Service,
Northeast Forest Experiment Station, Station Paper 138.
Upper Darby, Pennsylvania. 16 pp.

McGuinness, C. L.
1941. The importance of snow in relation to ground-water recharge.
Central States Snow Conference, Michigan State College.
Proceedings. Lansing, Michigan. 1: 166-172.

 

1963. The role of ground-water in the national water situation.
U.S. Dept. of Interior. Geological Survey Water Supply
Paper 1800. Washington, U.S. Government Printing Office.

Meinzer, O. E. and N. D. Stearns
1929. A study of ground-water in the Pomperaug Basin, Connecticut,
with special reference to intake and discharge. U.S.
Geological Survey Water Supply Paper 597—B. pp. 73-146.

Meinzer, Oscar Edward
1932. Outline of methods for estimating ground-water supplies.
U.S. Geological Survey Water Supply Paper 638-C. pp. 99-144.

Olmsted, F. H. and A. G. Hely
1962. Relation between ground-water and surface-water in Brandy-
wine Creek Basin, Pennsylvania. U.S. Geological Survey
Professional Paper 417-A. pp. A—l — Aw21.

Parker, G. G. and V. T. Stringfield
1950. Effects of earthquakes, trains, tides, winds and atmospheric
pressure changes on water in the geologic formations of
southern Florida. Economic Geology. 45(5): 441-460.

97

Rasmussen, W. C. and G. E. Andreasen
1959. Hydrologic budget of the Beaverdam Creek Basin, Maryland.
U.S. Geological Survey Water Supply Paper 1472. pp. 59-168.

Satterlund, Donald Robert
1960. Some interrelationships between ground-water and swamp
forests in the western upper peninsula of Michigan.
University Microfilms, Inc. Ann Arbor, Michigan. 171 pp.

Schicht, R. J. and W. C. Walton
1961. Hydrologic budgets for three small watersheds in Illinois.
Report of Investigation 40. Illinois State Water Survey,
Urbana, Illinois. 40 pp.

Smith, W. 0.
1961. Mechanism of gravity drainage and its relation to specific
yield of uniform sands. U.S. Geological Survey Professional
Paper 402-A. pp. A-l — A-12.

Society of American Foresters
1954. Forest cover types of North America. Society of American
Foresters, Washington, D.C. 67 pp.

Stallman, Robert W.
1956. Numerical analysis of regional water levels to define
aquifer hydrology. Transactions, American Geophysical
Union. 37(4): 451-460.

Striffler, W. D.
1959. Effects of forest cover on soil freezing in northern Lower
Michigan. U.S. Dept. of Agr., Forest Service, Lake States
Forest Experiment Station, Station Paper 76. 16 pp.

Striffler, William D.
1961. Intensity of soil moisture sampling is affected by depth
and vegetative cover. U.S. Dept. of Agr., Forest Service,
Lake States Forest Experiment Station, Technical Note ’
No. 603. St. Paul, Minnesota. 2 pp.

Taylor, D. W.
1948. Fundamentals of soil mechanics. John Wiley and Sons, Inc.
New York. 700 pp.

Thornthwaite, C. W. and J. R. Mather
1957. Instructions and tables for computing potential evapotrans—
piration and the water balance. Drexel Institute of
Technology, Laboratory of Climatology, Publications in
Climatology. Vol. X, No. 3. Centerton, N. J. 311 pp.

Todd, David Keith
1959. Ground-water hydrology. John Wiley and Sons, Inc. New York.
336 pp.

98

Troxell, H. C.
1936. The diurnal fluctuations in the ground-water and the flow of
the Santa Anna River and its meaning. Transactions, American
Geophysical Union. 17: 498-501.

U.S. Dept. of Commerce, Weather Bureau
1962. Climatological Data, Michigan, Annual Summary 1961. 76(13):
184-194.

U.S. Dept. of Commerce, Weather Bureau
1963. Climatological Data, Michigan, Annual Summary 1962. 77(13):
190—199.

U.S. Dept. of Commerce, Weather Bureau
1964. Climatological Data, Michigan, Annual Summary 1963. 78(13):
188-197.

Urie, Dean H.
1959. Pattern of soil moisture depletion varies between red pine
and oak stands in Michigan. U.S. Dept. of Agr., Forest
Service, Lake States Forest Experiment Station, Technical
Note No. 564. 2pp.

Weitzman, Sidney and Roger R. Bay
1958. Winter snow and water supply. The Conservation Volunteer,
Minnesota Dept. of Conservation. 21(126): 45-48.

Wenzel, Leland K.
1936. Several methods of studying fluctuations of ground-water
levels. Transactions, American Geophysical Union. 17: 400-405

White, Walter N.
1932. A method of estimating ground-water supplies based on dis-
charge by plants and evaporation from soil. U.S. Geological
Survey Water Supply Paper 659-A. 105 pp.