M . kw UH...“ . . r mafia? L‘ w! 1. .wklh. . mid... 4 ’ Z 5 9a , Glued. f? . new» . . r .3“: nay? (I: . . . .. . .2. 5%.?! r a a . unfri: Mgkhkift? . £939. 9 lit} "7 L , a O 2,5... X. «Mac... 2 fig“. 21.. 2 £ 3M“; ,. a; 3 3-, fins-r. V 3.2.2.... at , . a... 4.: I... Vr 5 . #13 ‘2... 5.0.4:»! :"vs. I . I . '0: . bu. 24.17 ..I.u 9141.-.! AZvLZi-‘I A: . ‘ .. w . V ‘i .. .I7. i5}. ul V I. .1‘ LV 3!: , . . V. .9 .3. {tv'l 1 . :3 .1. 1 a . ,... . .A..u.t. V... . ., I .. . .. M . ..\ I 1 ‘ Iv! .. e . . 1V .. 3:1 I... . a .22 :1...3:.l- J4 ‘ yifivfi‘ 2:}. 1.3;: ...:V M 7': ..._ S 2 , . 53:9... ..I.i.... s... 13151... 1.3.3.1. -M 1“ V , , V . l ul' 1 J I I. 1 Ir 3 I ' I I' ' THES‘S £~~ IMMIHHW“NH“IlHllllllH“lllllllllllfllml [V ,.) 3 1293 02080 (Yi- LIBRARY Michigan State University This is to certify that the thesise W7, ratified Geo u/ /u 641 LlflIWf/M affim CQE) Mal-flu, Sweat/f /< V/.'> “(j/Z Lt-v (SI/IS) fflf/[z Mafimfifm x194: MWRM”7%VLa OVA—f egjhgby C/WC Yii Ya has been accepted towards fulfillment /of the requirements for fi/ degree in g {5% Z ; Z Major professor 0-7639 MS U i: an Affirmative Action/Equal Opportunity Institution PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINE return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 11/00 cICIRC/DmDmpGS-pu THE INTEGRATION OF THE GEOGRAPHICAL INFORMA TION SYSTEM (018) AND THE SCIENTIFIC VISUALIZA TION SYSTEM (SVIS) FOR THE SIMULATION OF THE WATER RUNOFF IN A WATERSHED By Chia-Yii Yu A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Agricultural Engineering 1999 THl Geo mar. (GP USEC ma; GPE Wat. coll ABSTRACT THE INTEGRATION OF THE GEOGRAPHICAL INF 0W TION SYSTEM (GIS) AND THE SCIENTIFIC VISUALIZA T ION S YST EM (SVIS) FOR THE SIMULATION OF THE WATER RUNOFF IN A WATERSHED By Chia—Yii Yu The objective of the thesis is to develop a general design tool based on the Geographical Information System (GIS) for the simulation of water resource management. The Scientific Visualization System (SVIS), Global Positioning System (GPS), Differential Global Positioning System (DGPS), and Remote Sensing (RS) are used to produce three-dimensional GIS watershed maps of exceptional accuracy. The maps can be readily updated for changes in land use, landslides, and the stream system. SVIS changes a two-dimensional GIS map into a three-dimensional GIS map. GPS and DGPS establish the precise three-dimensional position coordinates of the watershed, and thus correct the less accurate geo-referenced position database of GIS. RS collects the most recent land-surface information of the watershed. The simulation of the water runoff on a vegetable farm in the Te-Chi Reservoir Watershed in Taiwan was used as an example to appraise the performance of the improved GIS design tool. The Watershed was assumed to be homogeneous. Copyright by CHIA-YII YU 1999 \I (I (1 (it d4 tilt ACKNOWLEDGEMENTS I like to express my sincere gratitude to my major professor, Professor F .W. Bakker-Arkema for his never-ending support and guidance of this study. My sincere thanks to my committee members, Dr. J. Bartholic and Dr. W. Northcott for their insight, guidance, and support in this study. I thank Dr. Alocilja for providing early assistance in my Master Degree program. I appreciate the assistance of several governmental agencies of the Republic of China (R.O.C.), i.e. the Institute of Botany of the Academia Sinica, the National Science Council, the Environmental Protection Agency, the Water Resources Bureau of the Ministry of Economic Affairs, and the Food Agency of the Taiwan Provincial Govemment for making available the water-quality, hydrological, and GIS reports and databases. Finally, a very special thank you to my parents and fiance Tsui-Ting Yang, for their never-ending love, support, encouragement, and guidance. iv LIST C LIST (1 (MP (HAP CHIP" TABLE OF CONTENTS LIST OF TABLES ----- ---- x LIST OF FIGURES -—-- -— -------- xii CHAPTER 1 INTRODUCTION ---- - 1 CHAPTER 2 OBJECTIVES --------------------- 5 CHAPTER 3 LITERATURE REVIEW --------- 6 3.1 GIS-SVIS of Watershed 6 3.2 Geographical Information System (GIS) --------------- 7 3.2.1 Geo-Referenced Data ------------- -- --- 7 3.2.1.1 Analog and Digital Maps - 7 3.2.1.2 Geo-Referenced Data Input - ---- 9 3.2.1.3 Map Projections and Coordinates Systems 10 3.2.2 GIS Computer Systems Selection and Personal Training --------------- 1 1 3.2.3 Applications of GIS ---- 11 3.3 Scientific Visualization System (SVIS) ------------------- 12 3.4 Global Positioning System (GPS) and Differential Global Positioning System (DGPS) ----------------------- 14 3.4.1 GPS Fundamentals -- - 14 3.4.1.1 GPS Satellites 14 3.4.1.2 GPS Ground-Control Stations 15 3.4.1.3 GPS Users 15 3.4.1.4 NAVSTAR GPS Performance 16 3.4.1.5 NAVSTAR GPS Satellite Signals 17 3.4.1.6 GPS Error Sources 18 3.4.2 DGPS Fundamentals 20 3.4.2.1 Carrier Phase Tracking and Code Phase Tracking for Surveying 21 3.5 Remote Sensing (RS) -- 23 3.5.1 Fundamentals of RS 23 3.5.1.1 Electromagnetic Spectrum 23 3.5.1.2 Wave Phenomena - 24 3.5.1.3 Wave Descriptions 1 24 3.5.2 Aerial Photography -------------- 25 3.5.2.] Basic Element of Aerial Photo Interpretation 26 3.5.2.1.] Types of Aerial Photography 27 3.5.2.1.2 Characteristics of Aerial Photography 27 3.5.3 Remote Sensing and GIS 28 CHAPTER 4 TE-CHI RESERVOIR WATERSHED 29 4.1 Geography 29 4.2 Terrain and Topography 30 4.3 Geology ----------- -- ------ 31 4.4 Rivers ------ 31 4.5 Climate and Hydrology 31 4.6 Land Use 32 CHAPTER 5 METHODOLOGY 33 vi 5.1 Hardware --- 5.1.1 GIS Digitizer System 5.1.2 GPS Receiver 5.2 Software 5.2.1 ArcView GIS 3.1 and Extension Programs ----- 33 34 35 35 5.2.1.1 ArcView GIS 3.1 --- 35 36 5.2.1.2 ArcView Spatial Analyst 5.2.1.3 ArcView Tracking Analyst 5.2.1.4 ArcView 3D Analyst 5.2.1.5 ArcView Image Analysis 5.2.2 Microsoft Excel 97 37 37 37 38 5.3 Methods ...... 38 39 5.3.1 Digitizing the Te-Chi Reservoir Watershed GIS Maps 40 5.3.1.1 Preparing for Digitizing ofa Watershed Map 5.3.1.2 Digitizing the Features of the Watershed Paper Map -------- 5.3.1.3 Correcting the Digitizing Errors 5.3.2 Modifying GIS Maps by GPS, DGPS. and Relative Positioning Surveying -- 5.3.3 Processing Aerial Photographs by GIS 5.3.4 Simulating Water Runoff ofA Watershed by GIS-SVIS 42 42 43 45 46 5.3.4.1 Hydrological Data and Software 5.3.4.2 Cell Runoff(mm) 5.3.4.3 Cell Sizes (m) 47 49 50 51 CHAPTER 6 RESULTS AND DISCUSSION vii 6.1 GIS Watershed Maps 51 6.1.1 Unmodified GIS Watershed Maps 51 6.1.2 Modified GIS Watershed Maps 53 6.1.3 Aerial Photographs Processed into GIS-SVIS Maps 58 6.2 GIS-SVIS Maps Illustrating Water Runoff 62 6.2.1 Slope-Type Factor ------------------ 67 6.2.1.1 Slope Sun-Shade GIS-SVIS Map - 69 6.2.1.2 Relative-Elevation GIS-SVIS Map -- 77 6.2.1.3 SinkhoIe-Area and Relative-Elevation GIS-SVIS Map --------- 78 6.2.1.4 Stream-Source Areas and Relative-Elevation GIS-SVIS Map - 8O 6.2.1.5 Slope Flow-Length GIS-SVIS Map 80 6.2.2 Land-Use Factor ---- 81 6.2.3 Slope-Type and Land-Use Factors 81 6.2.4 Calculation of Flow Rate --------------- 82 6.2.5 Flow Rate in the Drainage Area of the Chi-Chia-Wan Hydrological Station -- ---- 87 6.2.5.1 Average Annual Runoff (Type A) 87 6.2.5.2 Medium Annual Runoff (Type B) 87 6.2.5.3 Maximum Daily Runoff (Type C) 88 6.2.5.4 Flow Rate Conclusion 88 CHAPTER 7 CONCLUSIONS 90 CHAPTER 8 RECOMMENDATIONS FOR FURTHER STUDY 91 CHAPTER 9 REFERENCES ------------------ -- ---- 92 viii AWE. Amer Apps APPENDICES —-- - 100 Appendix A. Table 5.2 The Annual Runoff Data over 27-Year Period in Seven Hydrological Stations in the Te-Chi Reservoir Watershed 101 Appendix B. Table 5.4 The Maximum Daily Runoff Data over 27-Year Period in Seven Hydrological Stations in the Te-Chi Reservoir Watershed 102 Appendix C. Table 6.1 The “aat.dbf’ File ofthe Water Bodies Map ofthe Te-Chi Reservoir Watershed 103 Appendix D. Table 6.2 The “patdbf’ File ofthe Vegetable Farm Map ofthe Te-Chi Reservoir Watershed - -- 107 Appendix E. Table 6.3 The Registered Control Points and Modifications -------------- 109 Appendix F. Table 6.4 Statistical Program for the Modifications ofthe GIS Maps ---- 1 l 1 Appendix G. Table 6.10 The Area and Location of the Sinkholes in the Te-Chi Reservoir Watershed --------------------------------- l 15 1- APR TAP. Talc-l. Iabl ‘ I Tdbl lab Iat‘ I'm ur- lat 1.11 Tul T4 T4 1.: LIST OF TABLES Table 3.1 Specifications ofthe NAVSTAR and GLONASS Satellites 14 Table 3.2 Accuracy (95% probability) ofNAVSTAR PPS and SP8 16 Table 3.3 Specifications of the Signal Systems ofNAVSTAR GPS 17 Table 3.4 Wavelength and Frequency of Electromagnetic Radiation 23 Table 5.1 Specifications of the Digitized Paper Maps ofthe Te-Chi Reservoir Watershed -------- - --------- 40 Table 5.2 The Annual Runoff Data Over 27-Year Period in Seven Hydrological Stations in the Te-C hi Reservoir Watershed --------- — 101 Table 5.3 Area Proportion ofthe Over 30% Slope 48 Table 5.4 The Maximum Daily Runoff Data Over 27-Year Period in Seven Hydrological Stations in the Te-Chi Reservoir Watershed - 102 Table 6.1 The “aat.dbf" File ofthe Water Bodies Map of the Te-C hi Reservoir Watershed -- ----- 103 Table 6.2 The “patdbf’ File of the Vegetable Farm Map of the Te-Chi Reservoir Watershed -------------- --- -- -- 107 Table 6.3 The Registered Control Points and Modifications 109 Table 6.4 Statistical Program for the Modifications ofthe GIS Maps - l l 1 Table 6.5 The Smallest Area of the Various Types of Land-Use in the Te-Chi Reservoir Watershed --------------------- - 62 Table 6.6 Average Annual Runoff (mm) (Type A) in Seven Drainage Areas in the Te- Chi Reservoir Watershed -------------- 66 Table 6.7 Medium Annual Runoff (mm) (Type B) in Seven Drainage Areas in the Te- Chi Reservoir Watershed -- 66 Table 6.8 Maximum Daily Runoff (mm) (Type C) in Seven Drainage Areas in the Te- Chi Reservoir Watershed ------------------- 66 Table 6.9 Three Types of Area-weighted Runoff(mm) in the Te-C hi Reservoir Watershed ------------ 66 Table T101; Table 6.10 The Area and Location of the Sinkholes in the Te-Chi Reservoir Watershed l 15 Table 6.1 1 Predicted Flow Rate (m3/unit time) in the Drainage Area of Chi-Chia-Wan Hydrological Station - --------------------- 88 xi ‘. figure figurc figurt Fli‘dffi 5 figtzr; figs: Pgur LIST OF FIGURES Figure 4.1 Taiwan, Republic of China --------------------------------------- 29 Figure 4.2 The Te-Chi Reservoir Watershed in Taiwan, Republic of China ------------- 30 Figure 5.1 The GTCO AccuTab Surface-Lit LII Plus Digitizer System 33 Figure 5.2 The GTCO l6-button pointing device ---- 34 Figure 6.1 The Unmodified GIS Map of Water Bodies in the Te-Chi Reservoir Watershed (Line and Polygon Features) ----------------------- 52 Figure 6.2 The Unmodified GIS Map ofVegetable Farm in the Te-Chi Reservoir Watershed (Point Feature) --------------- 54 Figure 6.3 The Unmodified GIS Map of Vegetable Farm in the Te-Chi Reservoir Watershed (Polygon Feature) ------- 55 Figure 6.4 The Unmodified GIS Map of Water Bodies and Vegetable Farm in the Te-Chi Reservoir Watershed (Line and Polygon Features) 56 Figure 6.5 The GIS Map ofthe Cultivated Agricultural Areas in the Te-Chi Reservoir Watershed (Point Feature) ----------- ~ 57 Figure 6.6 The Modified GIS Map of Water Bodies and Vegetable Farms in the Te-Chi Reservoir Watershed ------------------------- 59 Figure 6.7 The Aerial Photograph ofthe T e-C hi Reservoir and Its Surroundings ------- 60 Figure 6.8 The Aerial Photograph ofthe Te-Chi Reservoir Processed by GIS-SVIS --- 61 Figure 6.9 The Annual Runoff in the Hydrological Stations of the Te-Chi Reservoir Watershed ----- -- -- 63 Figure 6.10 The Locations ofthe Hydrological Stations in the Te-Chi reservoir Watershed --- ------ --— 64 Figure 6.11 The GIS Map of the Hydrological Stations in the Te-Chi Reservoir Watershed --------------------------------- - 65 Figure 6.12 The GIS Map of Slope-Type in the Te-C hi Reservoir Watershed ----------- 68 Figure 6.13 The GIS-SVIS Map ofSlope Sun—Shade at 4:20PM in the Te-Chi Reservoir Watershed - --- --- 70 xii figure figure Figur: Fig Figure 6.14 The GIS-SVIS Map of Slope-Type in the Te-Chi Reservoir Watershed ---- 71 Figure 6.15 The GIS-SVIS Map of Relative Elevation in the Te-Chi Reservoir Watershed - 72 Figure 6.16 The GIS-SVIS Map of Stream-Source Areas and Relative Elevations in the Te-Chi Reservoir Watershed ----------------------------------- 73 Figure 6.17 The GIS-SVI S Map of F low-Length in the Te-Chi Reservoir Watershed -- 74 Figure 6.18 The GIS Map of Land-Use in the Te-Chi Reservoir Watershed ------------- 75 Figure 6.19 The GIS-SVIS Map of Land-Use in the Te-Chi Reservoir Watershed ------ 76 Figure 6.20 The GIS-SVIS Map of Area of Slope < 30% and Slope >= 30% in the Te-Chi Reservoir Watershed --------------- ---- 83 Figure 6.21 The GIS Map of Area (Slope > 30%) of Cultivations, Roads, and Landslides in the Te-Chi reservoir Watershed 84 Figure 6.22 The Location ofthe Exampled Vegetable Farm Area in the Te-Chi Reservoir Watershed — --- -—- 85 Figure 6.23 The Location of the Exampled Vegetable Farm in the Te-Chi Reservoir Watershed ------- 86 xiii andfi . I dgnw Heanl water oftxa consu dku Shflcq condi Illflm HDen. ddtah IESot “an: enun laFCI “(11¢ exec 1995 Chapter 1 INTRODUCTION GIS-SVIS of a watershed integrates the Geographical Information System (GIS) and the Scientific Visualization System (SVIS) system with aerial photographs and digital images in order to represent the characteristics of a watershed (Raper, 1989; Heamshaw and Unwin, 1994). GIS-SVIS improves on the traditional methods of watershed system assessment by its: (1) quick execution, (2) integration of different types of watershed information, (3) utilization of a variety of analysis tools. and (4) ability to consider the whole rather than part of the watershed, and (5) multi-dimensional display effect. The Watershed Geographical Information System uses the Watershed Information System (which collects the transformed Watershed Information) to simulate a watershed’s conditions. The Watershed Data is a collection of attributes of a watershed. Watershed Information is the raw watershed data used for analysis, evaluation, and decision-making (Denzer, 1993). The Watershed Information System transforms the raw watershed database into a systematical watershed database used in planning and managing of natural resources. The Watershed Geographical Information Systems comprise of: (1) a watershed geo-referenced database with information on the quantitative attributes and entities of a particular location, (2) a computer-based program for analyzing various layers of geo-referenced data and attribute-data of an entity, and for exploring the watershed relationships between entities, and (3) the list of personnel required for executing, operating, and maintaining GIS (Anonymous, 1997a; Congalton and Green, 1995; Dueker and Kjeme, 1989). with \‘irlu. whit? (Karl (Plen‘ Points “Vern: 19961 SVIS transforms a two-dimensional display of a GIS spatial database into a map with three-dimensional effect. The uniqueness of a GIS-SVIS display of a watershed is its virtual reality. SVIS is able to transform simulation data into a three-dimensional map which the human optic nerves and brain neurons can properly interpret (Denzer, 1993). The Global Positioning System (GPS) is a satellite-based radio-navigation system (Kaplan, 1996). It is an improvement over older conventional radio-navigation systems (Pierce, 1946). There are two types of GPS: (1) the Navigational Satellites for Timing and Ranging (NAVSTAR) system developed and operated by the US. Department of Defense, and (2) the Global Orbiting Navigational Satellite System (GLONASS) developed and operated by the Ministry of Defense of the Russian Federation (Lowe et al., 1997). GPS permits land, sea, and airborne users to determine: (1) the three- dimensional position of the user, i.e. the latitude, longitude, and altitude, (2) the times, i.e. the satellite vehicle (SV) time, GPS time, and Universal Coordinated Time (UCT), and (3) the velocity of the user, by calculating the user-position change between two points over time or by computing the SV Doppler frequencies. Both GPS systems operate twenty-four hours a day, in all weather, and can be used anywhere in the world (Kaplan, 1996) Differential Global Positioning System (DGPS) is a system for eliminating or drastically reducing the measurement errors caused by the effects of Selective Availability, Signal Bounce, Signal Noise, Ionosphere, Troposphere, Satellite Clock Drift, Code Measurement, Receiver Clock, Multipath Measurement, and Satellite Ephemerides (Bordin, 1996; Parkinson, 1996). A base station at a known location broadcasts the corrections of the errors (Capaccio et al., 1997). DGPS substantially imprt‘ StVet‘. gore: cenaq tfifiEI lCap; acqui remo obunr conefi pnnc' electr Pictoi VISUQ imag earn, lhe'r improves the accuracy of the GPS measurement. DGPS services are currently offered in several regions by private organizations for a subscription fee, and in some locations by governmental agencies free of charge. Consequently, as long as a DGPS service covers a certain area, a user can utilize a standard GPS receiver able to accept a particular differential input, and dramatically decrease or even eliminate the measurement errors (Capaccio et al., 1997). Remote Sensing (RS) is a reconnaissance-from-a-distance technology which can acquire information on an object or phenomenon through the analysis of data collected by remote sensors (Avery and Berlin, 1992). RS is much different from in situ sensing which obtains information by physical contact of an object (Avery and Berlin, 1992). Data collection (Lee and Marsh, 1995) and data analysis (J i and Mitchell, 1995) are two principle objectives of RS. The quality of the RS data depends on the sensitivity of the remote sensors to electromagnetic energy i.e. light, heat, and radio waves, and determines the quality of the pictorial images of an object (Lee and Marsh, 1995). Data analysis depends on traditional visual interpretation or on complex computer processing of the RS photographs and images (J i and Mitchell, 1995). RS can monitor a watershed’s environment at any site on earth. The Te-Chi Reservoir Watershed is a sub-watershed located at the headstream of the Ta-Chia-Chi river basin in the Central Mountain Range of Taiwan (Anonymous, I 996d). The multifunctional reservoir watershed is used for agriculture, tourism, flood adjustment, drinking-water supply, and hydroelectricity. The Te-Chi Reservoir Management Committee is the chief authority. The Taiwan Power Company operates the Te-Chi Reservoir Dam and supplies hydroelectric power to the Ta—Chia-Chi river basin. The total Te-Chi Reservoir Watershed area is 601.61 km2 (232.28 miz) (Anonymous, 1996d). Approximately 6.34% of the watershed area is used for agriculture (Anonymous, 1994c). The agricultural lands are mostly located along the Te-Chi Reservoir. The Reservoir has been subjected to bioenvironmental pollution for years (Anonymous, 1995c), and was therefore selected as the site for this GIS-SVIS-based water runoff study. Chapter 2 OBJECTIVES The objectives of this research are: (1) to develop GIS maps of the Te-Chi Reservoir Watershed. (2) to use the GIS maps for calculating the annual and maximum daily water-flow rates, and for developing the water-runoff maps in the Watershed. (3) to calculate the flow rates on a vegetable farm in a particular area of the Watershed. Chapter 3 LITERATURE REVIEW 3.1 GIS-SVIS of A Watershed Water resource conservation studies are important for a watershed because of the complicated nature of watershed-level environmental issues. GIS is essential to watershed simulation (Lyon and McCarthy, 1995). GIS systematically synthesizes watershed information and efficiently simulates specific geo-referenced data for analysis; it can be updated by remote sensing (Lee and Lunetta, 1995). GIS, DGPS, and RS can be integrated to produce accurate and dynamical planes with the various properties of a specific area (Lyon and McCarthy, 1995). DGPS corrects the GIS watershed maps by the proper correction of signal measurement, and RS updates the map information by image-processing techniques. Thus, with a package of GIS, DGPS, and RS, a researcher is able to analyze in depth the water issues of a watershed. SVIS, comprising a computer-aided graphical with human vision, supports the multi-dimensional display and the auto-animation of complex data sets. By integrating GIS and SVIS, the GIS dataset of a watershed can be displayed as a three-dimensional map (latitude, longitude, and altitude) or as a multi-dimensional map (latitude, longitude, altitude, and time...) (Anonymous, 1997c; Anonymous, 19961). For example, for visualization the SVIS software “Data Explorer Visualization” (Anonymous, 1999) integrates the software programs ESRI’s ARC/INFO, the ERDAS’s IMAGE, and the Oracle Database, utilizing Digital Elevation Model (DEM), Digital Line Graphs (DLG) for the various output-model formats. By using the spatial effect of SVIS, the integration of GIS and SVIS becomes an excellent tool for watershed system analysis. A GIS-SVIS system containing color differentiation and auto-animations is an excellent system- analysis and computation tool. The following sections review the principles and interactions of GIS, SVIS, GPS, DGPS, and RS in order to simulate the water runoff in a watershed. 3.2 Geographical Information System (GIS) Geographical information is of a dynamical nature because it synthesizes and updates the three information components of a region regarding: (1) space, (2) time, and (3) attributes (Chrisman, 1997). GIS is designed to input, store, update, manipulate, analyze, and illustrate all types of geo-referenced information by arranging systematically computer hardware, software, and geo-referenced data. Because of the ability of GIS of storing spatial information, analyzing geo-referenced data, and managing gee-referenced information in an integrated manner to display the area properties of a region, GIS is really a computer-based mapping system. GIS requires knowledgeable, highly trained, and experienced personnel since it are not a fully automated system (Anonymous, 1997a). 3.2.1 Geo-referenced Data 3.2.1.1 Analog and Digital Maps An expansive and adaptable definition of a map is “the physical or conceptual depiction of the characteristics of the Earth or other celestial body” (Robinson, 1976). Maps under this definition can be divided into two classes: (1) real or analog maps, and (2) virtual or digital maps. A real map is static and is a drawing or a scanned image. A virtual map is a data set stored in digital form; it is dynamic, and is more flexible than an analog map for recording, examining, interrogating, and analyzing information over time (Lyon and McCarthy, 1995). A GIS map is a digital map which represents locations on the Earth’s surface in one or several coordinate systems, and which has been projected on a flat surface. The information provided includes spatial and descriptive information (Anonymous, 1997a). The spatial information describes the spatial relationships of the various geographical features of a region by topology; it can be represented as a point feature, as a line/arc feature, or as an area/polygon feature, and as a surface feature/entities of a region. Spatial and descriptive information is stored in graphical and tabular databases (Anonymous, 1997a). The link between a graphical database and its records in a tabular database is of a one-to-one link through a numerical identifier. A point feature is represented: (1) by a discrete location with a shape too small to be shown as a line or area feature, and (2) by a point location without an area (Anonymous, 1997a). A line/arc feature is a set of ordered coordinates which represent the linear shape of a map object when: (1) it is too narrow to be displayed as an area, and (2) it is a feature without width (Anonymous, 1997a). An area feature represents a homogeneous area bounded by one or more arc features or by a set of polygons, and is measured in unit squared (Anonymous, 1997a). A surface feature is a three-dimensional representation of geographic information, and is represented by a set of continuous data in which the map features are spatially continuous, i.e., there is an infinite set of values between any two locations. There is no clear or well-defined break between possible values of a surface feature. Models build from regularly or irregularly spaced sample points on a surface can represent surfaces. Examples are the Grid model, the Lattice model, and the Triangulated Irregular Network (TIN) model (Anonymous, 1997a). An entity is a collection of objects described by the same attributes. Entities are identified during the conceptual design phase of a database and an application design, and are represented by computerized cartographic-data structures. The spatial relationships of a region are defined by its topology through mathematical procedures (Anonymous, 1997a). Connectivity (arc-node topology), area definition (polygon-arc topology), and contiguity (left-right arc topology) are three topological concepts employed in the analysis of spatial information. Various geo-referenced data models are available in the literature to represent geographical information, e.g. the Coverage model, Image model, and Drawing model (Anonymous, 1997a). 3.2.1.2 Geo-referenced Data Input There are many methods to input geo-referenced data, e.g., by map digitizing (Anonymous, 1996c), by map or aerial photography scanning (Anonymous, 1998b), by map tracking with DGPS, and by RS imaginary and aerial photography (Anonymous, 1998c). Each process may encounter the following problems: (1) overlaying of two or more maps in the same region due to different map scales (Anonymous, 1996c), and (2) mistakes in the ground registration of the spatial database (Anonymous, 1997a). The specific features and the degree of accuracy of two maps at different scales are difficult to match up (Stevens, 1946). If the spatial database is not properly registered, a serious problem can occur during the latter stages of the analysis and assessment (Anonymous, 1996c). In addition, map digitizing can cause the wrong scale and symbols to be produced for displaying the desired details. 3.2.1.3 Map Projections and Coordinate Systems Spatial data registration (multipurpose cadaster) requires a ground-based coordinate system so that the original data can be transformed and a fixed latitude/longitude relationship can be established (Anonymous, 1997a). Because of the curved surface of the Earth, a map projection is necessary to produce a flat map for a coordinate system of a particular spatial data registration (Synder, 1987). A map projection is a mathematical technique for calculating the parallels and meridians on a map. The projection applies the mathematics of an ellipsoid, of a sphere (Latitude-Longitude), or of a flat coordinate system to develop (Chrisman, 1997): (1) a cylindrical projection, (2) a conical projection, and (3) an azimuthal (planar) projection. Several important projections such as the Mercator projection (Anonymous, 1996c), the US Quadrangle projection (Anonymous, 19966), the Lambert Equal Area projection (Anonymous, 19966), the Lambert Conformal projection (Anonymous, 1996c), the Robinson projection (Anonymous, 1996 b), and the Gnomic projection (Anonymous, 19966) can be derived from the three basic projections. These map projections can be represented in a number of coordinate reference systems in specifying the ground registration of the spatial database. The coordinate reference systems include: (1) the Spherical coordinate system, (2) the two-dimensional Cartesian coordinate system, (3) the three-dimensional Cartesian coordinate system (Chrisman, 1997), (4) the State-Plane coordinate system (Anonymous, 1996c), (5) the Universal Transverse Mercator (UTM) coordinate system (Congalton and Green, 1995), and (6) the US Public Land Survey 10 coordinate system (Anonymous, 1996c). The correct selection of the map projection and the coordinate system is important in the registration of the geo-referenced database of a region because the GIS maps of different properties of a region cannot be overlaid if different map projections and coordinate systems are used (Muehrcke, 1986; Snyder, 1987; Thompson, 1988; Congalton and Green, 1995). 3.2.2 GIS Computer Systems Selection and Personnel Training Not only should the characteristics of a geo-referenced data be well understood, the GIS computer system should be chosen carefully. Proper estimation of the hardware/software computing needs of the GIS-related research project is essential. In selecting the correct software, the following should be considered (Gupyill, 1988; Parker 1989): (l) the data input and editing should be simple, (2) the software should contain a cartographic analysis tool and should have modeling capability, (3) the software should run on different computer systems and operating systems, and (4) the software should be able to run under different hierarchical and relational database-management systems. In selecting the hardware, the following should be considered (Gupyill, 1988; Parker 1989): (1) the size of the hard disk, (2) the clock speed and RAM of the computer, (3) the number of users, (4) the compatibility with input/output devices. State-of-the-art GIS equipment and technology cannot be operated and properly maintained without well-trained personnel. Consequently, the staff occupies an essential place in a GIS system (Anonymous. 1997a). 3.2.3 Applications of GIS The potential applications of GIS are broad. As long as a problem is associated with space, GIS is an excellent tool. In this research, GIS is applied to the simulation of 11 the water runoff in a watershed. The GIS maps can perform issue-settling tasks and undertake question-interrogating jobs. 3.3 Scientific Visualization (SVIS) SVIS is important in environmental research. By representing numerical data in a visual format, SVIS can provide environmentalists and engineers with a better understanding of the results of their research. The human sense of sight intuitively reflects reality (Denzer, 1993). SVIS assists GIS to: (1) create an exploratory spatial database by employing the Digital Terrain Model (DTM), the Digital Elevation Model (DEM), or other topographical database model (Anonymous, 1996f), and (2) use the exploratory spatial database to display a three-dimensional map (Anonymous, 1997c). SVIS: (1) transforms symbolic data into geometric data and enables researchers to intuitively observe the results of simulation (Raper, 1989), (2) offers a method for seeing the unseen to enrich the process of scientific discovery and foster unexpected insights (Robinson, 1976), (3) allows the generation of images from complex multi-dimensional data sets (Anonymous, 1998b), and (4) allows the use of computer graphics (Domik, 1994), image processing (Anonymous, 1998b), computer vision (Heamshaw, 1994), and computer-aided design (CAD). The goal of SVIS is to provide new scientific insight through visual means. An estimated fifty-percent of the brain’s neurons are associated with vision. SVIS in scientific computation aims at putting neurological machinery to work. Three- dimensional displays stimulate more neurons and therefore a larger portion of the brain in the problem-solving process. With two-dimensional contour maps, the mind must first 12 build a conceptual model of the relief before an analysis can be made. Considering the cartographic complexity of certain terrain, this is an arduous task for even the most dexterous mind. Three-dimensional displays simulate spatial reality, thus allowing the viewer to quickly recognize changes in elevation. Some numerical models of watershed runoff movement require a large database (Domik, 1994). The most efficient way for researchers and analysts to study this information is to visually represent it. SVIS is used in many disciplines to interpret large, complex database sets and gain insight into the trends, patterns, dependencies, and missing data within a database (Worboys, 1995). SVIS applied to GIS depends on psychological cues to create a natural three- dimensional display on a two-dimensional monitor. SVIS models do not result in photographs but in renditions (Keller and Keller, 1992). The process of generating a three-dimensional scene is termed rendering. To render a realistic scene, individuals rely on visual perspective cues and subtle changes in color and texture (Keller and Keller, 1992). Depth in a sense can result from feature obstruction and overlap, or from the addition of atmospheric attenuation such as fog or haze. Usually, the light intensity and clouds are measures of the relative distance within a scene. The presence of trees, or a seasonal characteristic such as snow, artificially enhances the sense of reality. A physiological cue, such as accommodation, convergence or the retinal disparity, can balance the three-dimensional image of a two-dimensional image (Domik, 1994). SVIS is commonly used to model a terrain surface such as a watershed surface (Anonymous, 1994a). Most terrain algorithms are based on Fractal Geometry which in turn is based on the concept of Self-Similarity. Self-Similarity accounts for the change in 13 distance from a spatial feature; i.e., what appears at one scale is represented at the same or another scale (Domik, 1994). The Fractal Dimension of a surface can describe a landscape form. A fractal dimension is measured as a real number ranging between two and three, where two is for a perfectly smooth surface and three is for an infinitely variable surface. 3.4 Global Positioning System (GPS) and Differential Global Positioning System (DGPS) 3.4.1 GPS Fundamentals NAVSTAR is based on a constellation of satellites that continuously transmit coded signals in two carrier frequencies L1 and L2 (Parkinson and James, 1996; Kromer and Landis, 1992). A GPS receiver should receive the navigational signals from at least four satellites in order to identify its three-dimensional position (latitude, longitude, and altitude) and its velocity in real time (Kaplan, 1996). A GPS system consists of: (1) the GPS satellites, (2) the GPS ground-control stations, and (3) the GPS users. 3.4.1.1 GPS Satellites Table 3.1 Specifications of the NAVSTAR and GLONASS Satellites GPS NAVSTAR GLONASS No. of Space VeHicFes 24 24" Launch Base Cape Canaveral, USA Baikonur Cosmodrome, Kazakstan No. of Orbital Planes 6 3 No. of Satellites Per Plane 4 8 Orbital Altitude (km) 20.200 19,130 Inter-orbital Plane Angle (Degree) 60 120 Orbital Inclination (Degree) 55 64.8 Period of Revolution 11hr, 58min, OOSec l 1hr, 15min, 445ec * Two space vehicles were decommissioned in 1996. Data source: Lowe et a1, 1997 and Kaplan, 1996 The GPS satellites send radio-navigation signals from space. Both the US. NAVSTAR and the Russian GLONASS GPS satellites orbit the Earth, and form a 14 nominal Operational Constellation. Each satellite transmits a signal at a specific frequency (Lowe et al., 1997). Some satellites have the same frequency but since they are located in antipodal positions in an orbit plane, or in different orbit planes, they do not appear at the same time in a user’s view (Blanchard, 1996). Table 3.1 describes the specifications of the US. and the Russian GPS satellites. Additional information on the satellites is given in Section 3.4.4. 3.4.1.2 GPS Ground-Control Stations GPS ground control is provided by the tracking stations. A station measures the satellites’ signals to compute their precise orbital posiyion (ephemeris) and to correct the clock (Parkinson, 1996). The Master Control station transmits the satellites’ ephemeris and clocks offsets to the satellites which in turn incorporate the correction data into the radio signals to be sent to the GPS receivers (Kromer and Landis, 1992). The NAVSTAR GPS satellites are controlled by the Master Control station located at Schriever Air Force Base, CO, U.S.A and four tracking stations located around the world (Kromer and Landis, 1992). The GLONASS GPS satellites are controlled by the Ground-based Control Complex (GCC) located at Moscow, Russia and a series of Command Tracking Centers (CTC) located at different locations in the Russia (Kuranov, 1995; Langley, 1995). The author employs the NAVSTAR system in this thesis because a GLONASS GPS receiver is not commercially available in the U.S.A. 3.4.1.3 GPS Users A GPS system relies upon the precise distance measurement between the GPS user and the satellites (Anonymous, 1996a). Users determine their position on the earth 15 by accurately measuring the distance from four to twelve satellites. The satellites act as reference points and transmit their positions and time signals to the GPS user. GPS technology consists of the GPS satellites, receivers, processors, and antennas, and can be used in navigating, positioning, timing, tracking, and mapping. A GPS receiver must receive at least four satellite signals to accurately compute the four dimensions of latitude, longitude, altitude, and time (Blanchard, 1996); the velocity of a user is calculated from his/her positions over an elapsed time period (Anonymous, 1996a) 3.4.1.4 NAVSTAR GPS Performance The NAVSTAR GPS performance largely depends on the US. government actions and on the effects of atmospheric noise and bias. The 1994 Federal Radionavigation Plan describes the type of NAVSTAR GPS services: (1) Precise Positioning Service (PPS), and (2) Standard Positioning Service (SPS). PPS cryptographic equipment is only available to designated U.S. governmental agencies, to the US. and Allied military, and to the civilian users specifically approved by the US. government. SPS is available to the users from all over the world without charge (Anonymous, 1995a). Table 3.2 Accuracy (95% probability) of NAVSTAR PPS and SPS ervnces Horizontal Accuracy (m) 22 100 Vertical Accuracy (m) 28 156 Time Accuracy (nanosecond) 100 340 Source: US. Federal Radionavigation Plan, 1994 The US. Department of Defense degrades SPS accuracy in order to control the Selective Availability (SA) (See Section 3.4.1.6) (Parkinson, 1996) by adding a time- 16 varying bias (See Table 3.2). The SPS figures are 95% accurate; i.e. the equivalent of two-distance root-mean-square (2 drms) or twice the radial error standard deviation. 3.4.1.5 NAVSTAR GPS Satellite Signals Table 3.3 represents the specifications of the signal systems of NAVSTAR. The NAVSTAR system broadcasts two microwave carrier-phase signals L1 and L2 (Parkinson and James, 1996; Kromer and Landis, 1992). The L1 signal contains the Navigation Message and the SPS code; the L2 signal provides the PPS code which enables the measurement of a physical error such as the ionospheric delay. In order to obtain specific information, the signals are modulated by three binary codes (Parkinson and James, 1996; Kromer and Landis, 1992): (1) a C/A (Coarse Acquisition) code, (2) a P (Precise) code, and (3) a Navigation Message. Table 3.3 Specifications of the Signal Systems of NAVSTAR GPS Signal Systems \ GPS Types NAVSTAR Carrier Phase Frequency: L1 (MHz) 1575.42 Carrier Phase Frequency: L2 (MHz) 1227.60 Signaling CDMA' Type of PRN Code GOLD Number of Code Elements (C/A Code, bit/millisecond) 1023 Number of Code Elements (P Code, bit/millisecond) 2.35"‘10l4 C/A Code Chipping Rate (Mbit/second) 1.023 P-Code Chipping Rate (Mbit/second) 10.23 Navigation Message (Chipping Rate, bit/sec) 50 Navigation Message (Modulation) BPSK NRZ Navigation Message (Total Length, second) 750 Navigation Message (Subframe Length, second) 6 1. k: frequency channel number (k=0,1.2...) with different channel spacing for L1 and L2. 2. FDMA: Frequency Division Multiple Access; CDMA: Code Division Multiple Access. Data source: Axelrad and Brown, 1996; Parkinson and James, 1996; Filatchenkov, 1996; Kuranov, 1995; Raby, 1994. The C/A code, a constantly changing Pseudo Random Noise (PRN) with a short chipping rate, is regarded as a Spoof code to modulate the L1 carrier phase of SPS in order to acquire the P code (Parkinson and James, 1996). Each satellite is given a specific unencrypted C/A PRN code for identification, thus allowing any user of the system to decode and use the transmitted data. The US. Department of Defense de-synchronizes the satellite clock and thereby introduces an intentional error in the C /A code. The P code, a constantly changing PRN at a long chipping rate, is an Anti- Spoofmg code for modulating both the L1 and L2 carrier phase signals of PPS (Parkinson and James, 1996; Krorner and Landis, 1992). Because of its higher modulation bandwidth, the signal is significantly more precise that the C/A code. It is combined with an encrypted Y code, and thus a PPS user requires either a classified Anti-Spoofing Module for each receiver channel or a cryptographic decoding key. The Navigation Message modulates the L1 carrier-phase signal mixed with C/A code signal, and contains data on a satellite’s orbit, the clock correction, the ephemeris, the almanac, and other system parameters. 3.4.1.6 GPS Error Sources Between space and a GPS receiver on the ground, the transmitted radio- navigational microwaves from the GPS satellites are affected by noises (Parkinson, 1996). Other sources of error are bias and human mistakes. Specifically, the sources of the noise errors are: (1) the noise in the PRN code of the GPS satellites (Parkinson, 1996), and (2) the noise in the GPS receiver (Leick, 1995). The error in the PRN code is caused intentionally for security reasons. The noise error source in the GPS receiver is caused by the noise inherent in any electronic instrument measurement. The noise errors are: (1) selective availability, (2) signal bounce, (3) signal noise, (4) code measurement mistakes, and (5) receiver clock mismatch (Capaccio et al., 18 1996 & 1997; Langley, 1997; Brodin, 1996; Misra et al., 1993 & 1996; Parkinson, 1996; Rossbach et al., 1996; Raby, 1994; Hartman et al., 1991; Doebelin, 1990). The sources of the bias errors are: (1) selective availability, (2) satellite clock drift, (3) satellite ephemeride error, (4) ionospheric delay, (5) tropospheric delay, and (6) multipath reflection (Capaccio etal., 1996 & 1997; Langley, 1997; Brodin, 1996; Misra et al., 1993 & 1996; Parkinson, 1996; Rossbach et al., 1996; Raby, 1994; Hartman et al., 1991). According to the “1994 Federal Radionavigation Plan”, GPS policy requires some services to be degraded intentionally. The SA bias is an example of bias in the SPS information controlled by US. Department of Defense; it is introduced in the PRN code and is a time-varying bias. Each satellite signal has a different SA bias in the C/A code signal. This bias results in the NAVSTAR GPS SPS performance to be reduced from 30m to 100m. The errors due to the tropospheric and ionospheric delays are caused by natural factors. The satellite ephemeride error, multipath reflection, and satellite clock drift are the result of instrumental factors. Natural factors are due to Instrumental factors and include: (1) the ephemeris error because a GPS receiver does not update the Navigation Message on time, thus causing the Almanac to be off, (2) the multipath error caused by a signal being reflected off a coarse surface near the GPS receiver and being received by the receiver, and (3) the error in the reading of the satellite clock (Capaccio et al., 1996 & 1997; Langley, 1997; Brodin, 1996; Misra et al., 1993 & 1996; Parkinson, 1996; Rossbach et al., 1996; Raby, 1994; Hartman et al., 1991). The human error in obtaining the correct GPS information can be substantial and may be caused by incorrect computer Operations at the control station or wrong selection of the geodetic datum at the receiver (Anonymous, 1996a). 3.4.2 DGPS Fundamentals The fundamental purpose of DGPS is to correct the bias errors at a location with the corrected bias errors at a known position, such as a GPS receiver’s location, or a base station. By using two identical receivers simultaneously, one at a reference point with known location coordinates and the other at unknown location coordinates, the unknown location coordinates can be established because the differential positioning removes the error sources common to both receivers. Applying this positioning correction is restricted to a limited range. Both receivers should use the same GPS satellites during the measurement, and have identical Geometric Dilution of Precision (GDOP) indices in order to be identically affected by the bias errors (Capaccio et a1, 1997). The components of GDOP are (Anonymous, 1996a): (1) the Position Dilution of Precision (PDOP) or Spherical Dilution of Precision (SDOP) for displaying the measurement condition of the three-dimensional position, (2) the Horizontal Dilution of Precision (HDOP) for displaying the measurement condition of the latitude and longitude, (3) the Vertical Dilution of Precision (VDOP) for displaying the measurement condition of the altitude, and (4) the Time Dilution of Precision (TDOP) for displaying the measurement condition of the time. Vector differences between a GPS receiver and the NAVSTAR GPS satellites magnify the GPS measurement errors. In a position fix, the volume of the envelope 20 formed by the unit-vectors pointing from a GPS receiver to the NAVSTAR GPS satellites, is inversely proportional to the GDOP (Anonymous, 1996a). The GDOP is computed from the geometric relationships between the position of a GPS receiver and the positions of the satellites (Hofmann-Wellenhof, 1992). For planning purposes, the GDOP is often computed from Almanac information and from the estimated position of the GPS satellites. It should be kept in mind that the estimated GDOP: (I) does not take into account obstacles which block the line-of—sight from the receiver to the satellites, (2) cannot be displayed on the screen of the receiver, and (3) is usually computed with using parameters calculated from the navigation-solution processes (Capaccio, 1997; Leick, 1995). Generally, the measurement errors from the satellite signals are multipled by the appropriate GDOP term to estimate the position or time error (Parkinson, 1996). The various GDOP terms are computed using the navigation covariance matrix. While each of the GDOP terms can be individually computed, they are covariantly linked and thus are not independent of each other (Capaccio, 1997; Leick, 1995). This procedure is explained in section 5.3.3. 3.4.2.1 Carrier Phase Tracking vs. Code Phase Tracking for Surveying The generator of a GPS receiver continuously produces a pre-determined PRN code (Leick, 1995). The code repeats the same 1023-chip PRN code sequence every millisecond. The signal generated by the GPS receiver can match the satellite signal either partially or in firll. When the NAVSTAR satellite and receiver codes match, the signal is detected. The bandwidth (cycles) of the PRN code is so wide that even if the receiver generates the same code, the two are often not in sync. This results in an out-of- phase match, and may cause a 300 meters measurement error (Brodin, 1996; Armstrong, 21 1992). If a carrier phase is used instead of a code phase, the Carrier Phase Tracking system eliminates this problem. The Carrier Phase Tracking technique uses a special device to track the L1 and/or the L2 carrier phase signals (Brodin, 1996). L1 has a bandwidth of 19 centimeters, and is smaller than the C/A code. Tracking and measuring the carrier phase signals improves the accuracy of a GPS receiver’s measurements to +/- 1.0 mm (Brodin, 1996). Carrier Phase Tracking does not provide improvement for the timing function. When modulating a carrier signal with a time-tagged binary code, the carrier signal does not carry a time-tag that distinguishes one cycle from another. Carrier-phase information contains information on the phase cycles and on the fractions of cycles over time. At least two identical GPS receivers are required for tracking the carrier phase signals simultaneously (Brodin, 1996). The difference in the ionospheric delay at two GPS receivers must be small enough to insure that the carrier phase cycles are properly accounted for. This usually requires that two GPS receivers are located within a limited distance of each other to prevent this problem. The accuracy of the Carrier Phase Tracking of a measurement also depends on the location of the user (Capaccio et al., 1996 & 1997; Langley, 1997; Brodin, 1996; Misra etal., 1993 & 1996; Parkinson, 1996; Rossbach et al., 1996; Raby, 1994; Hartman et al., 1991). With a carrier-tracking receiver, and using the Relative Positions surveying method, several positions can be measured within a limited range from one reference point. The surveying method is able to fix the positions of the points in relation to each other (Parkinson, 1996; Leick, 1995) and in relation to the location of the Control Points, 22 with an accuracy on the order of millimeters (Anonymous H, 1984; Clarke, 1963). See Section 5.3.3 for more detailed information. 3.5 Remote Sensing (RS) 3.5.1 Fundamentals of RS 3.5.1.1 Electromagnetic Spectrum Table 3.4 Wavelength and Frequency of Electromagnetic Radiation Name Wavelength Frequency < 0.1 nm Gamma rays X—rays 0.1nm-10nm Ultraviolet (UV) light 10 nm - 0.4 pm 750 ~ 3,000 THz Visible light 0.4 pm - 0.7 pm 430 ~ 750 THz Infrared (IR) Near 1R Waves 0.7 pm ~ 1.3 pm 230 ~ 430 THz Waves Short Waves 1.3 pm ~ 3 pm 100 ~ 230 THz Intermediate 3 pm ~ 8 pm 38 ~ 100 THz Waves Thermal IR 8 pm ~ 14 um 22 ~ 38 THz Waves Far IR Waves 14 um ~ 1 mm 0.3 ~ 22 THz 0.1m~100km 0.3 THz~30kHz Radio Waves Data source: Serway, 1992 The Electromagnetic Radiation (EMR) consists of energy propagated through space between the electric and magnetic fields. The EMR comprises the entire 23 electromagnetic spectrum; i.e. from cosmic rays, gamma rays, X—rays, ultraviolet light, visible light, infrared radiation, to radio waves (Serway, 1992) (See Table 3.4). 3.5.1.2 Wave Phenomena When electromagnetic waves are radiated through space and encounter an object as small as a molecule of air, the radiation will either be reflected from the object, absorbed by the object, or transmitted through the object (Halliday et al., 1993). Thus, the total amount of radiation that strikes an object, i.e. the incident radiation, is equal to the reflected radiation plus the absorbed radiation plus the transmitted radiation (Ohanian, 1985) In remote sensing, the reflected radiation is the critical part. It is the radiation that causes the human eye to see colors, infrared film to record vegetation, and radar images of the earth to be created and to become visual (Asrar, 1989). 3.5.1.3 Wave Descriptions The frequency or wavelength of the EMR identifies the type of electromagnetic waves (see Table 3.4) (Serway, 1992). The velocity of electromagnetic waves is equal to the speed of light, i.e. 3 x 108 meters/sec (Ohanian, 1985). The electric and the magnetic fields are important concepts used to mathematically describe the physical aspects of electromagnetic waves. The electric field vibrates in a direction transverse (i.e. perpendicular) to the direction of travel of an electromagnetic wave. The magnetic field vibrates in a direction transverse to the direction of an electromagnetic wave and transverse to the electric field. Polarization is defined as the orientation of the electrical field; it is usually described in terms of horizontal polarization and vertical polarization. 24 3.5.2 Aerial Photography Due to the orientation of the optical axis of a RS camera with respect to the surface of the Earth at the time of film exposure, aerial photography can obtain: (1) a vertical airphoto taken with the camera’s optical axis oriented in a vertical or nearly vertical angle to the local ground surface (90" +/- 3"), and (2) an oblique airphoto taken with the camera’s optical axis tilted 20° or more from the vertical (Avery and Berline, 1992). An oblique photograph can be further classified as: (1) a high-oblique airphoto showing the each surface, the horizon, and a portion of sky, and (2) a low-oblique airphoto showing only the earth surface. Aerial photography has two benefits: (1) it offers cartographers and planners detailed measurements, and (2) it provides observers information on land use and environmental conditions (Ulliman, 1995). Although both GIS maps and aerial photos present an overlook of the earth, aerial photographs are not maps. A map is an orthogonal representation of the earth surface; it is directionally and geometrically accurate (at least within the limitations imposed by projecting a three-dimensional object onto two-dimensions). Aerial photographs display a high degree of radial distortion, and thus the topography is distorted (Ulliman, 1995). Measurements made from a photograph are not accurate until corrections are made of the distortion. Nevertheless, aerial photography is a powerful tool for studying the earth environment (Anonymous, 1998b). Because GIS software can correct for radial distortion, aerial photographs are an excellent data source for many projects, especially for those which require spatial data of 25 a location at periodic intervals (Anonymous, 1998b). Typical applications include land use surveys and habitat analyses. 3.5.2.1 Basic Elements of Air Photo Interpretation Aerial photographs are different from regular photos in three important aspects: (1) objects are portrayed from an overhead and unfamiliar position, (2) infrared wavelengths are seen, and (3) photos can be taken at different scales (Wolf, 1983). The following Interpretation Factors can assist interpreters in identifying objects in an aerial photograph (Wolf, 1983; Avery and Berline, 1992; Ulliman, 1995; Yu, 1997): (1) Tone (hue or color): tone relates to the relative brightness or color of the elements in a photograph. It is a basic interpretive element because without tonal differences, the other elements cannot be distinguished. (2) Size: the size of an object must be considered in the context of the scale of the photograph. The scale is essential for determining if an object is a small pond or a large lake. (3) Shape: shape relates to the general outline of an object. A regular geometric shape is usually an indicator of a man-made design. (4) Texture: the “smooth” or “rough” texture of the features of an image is an indication of a uniform and non-uniform wave-frequency reflection, and thus of a change in the tone in a photograph. Because texture is produced by a set of features which are too small to identify individually, texture recognition depends on the sense of sight in response to the extent of the roughness of a feature. Grass, cement, and water generally appear as "smooth", while a forest canopy appears as "rough". 26 (5) Pattern (spatial arrangement): the pattern formed by objects in a photo can be recognized by its various spatial arrangements. A random pattern formed by an unmanaged area of trees and an artificial pattern formed by an evenly spaced row of trees in an orchard are two different spatial arrangements. (6) Shadow: shadow aids interpreters in determining the height of an object. However, shadow may also obscure an object. (7) Site: site refers to a particular topographical or geographical location on an aerial photograph. It is important identifying vegetation type and landforms. (8) Association: some objects are found “in association” in combination with other objecs. The association of an object can provide useful insights about other objects. (9) Stereo: stereo shows the relative elevation of an object. (10) Resolution: resolution refers to the degree of detail shown in an aerial photograph. 3.5.2.1.] Types of Aerial Photography There are several types of aerial photography due to: (1) the type of film, (2) the lens resolution (line pairs/mm), (3) the filter type, and (4) the peripheral instrumentation (Avery and Berline, 1992). The various types are: (1) black and white photography, (2) infrared black-and-white infrared photography, (3) color infrared photography, (4) normal color photography, (5) panchromatic photography, and (6) special aerial photography, e.g., ultraviolet and additive-color. 3.5.2.1.2 Characteristics of Aerial Photography 27 People shoot an object or phenomenon in the various electromagnetic spectrums in order to record instant information. Aerial photography can (Avery and Berline, 1992; Anonymous, 1998b): (1) record an image of any earth-surface area, (2) show any feature of an object or phenomenon, and (3) display an object or phenomenon at any spatial resolution and geometric fidelity. Aerial photographs can provide multispectral images of an area and overlapping areas, while GIS can analyze various levels of information provided by aerial photography (Anonymous, 1998b). 3.5.3 Remote Sensing and GIS Remotely sensed images have some features which are ideal for a GIS data source because: (1) its regional view, (2) its repetitive looks at the same area, (3) its ability to scan over a broader portion of the spectrum than the human eye, (4) its focus on a specific bandwidth in an image, (5) its ability to look at a number of bandwidths simultaneously, (6) its ability to record signals electronically and to provide geo-referenced digital data, and (7) its ability to be effective under night conditions and in bad weather. 28 Chapter 4 TE-CHI RESERVOIR WATERSHED 4.1 Geography Te-Chi Reservoir Watershed is a sub-watershed located at the headstream of the Ta-Chia River basin in the Central Mountain Range of Taiwan. The area is about 601.61 square kilometers (Anonymous, 1996d). The watershed has a heart type in shape. Figure 4.1 shows the country of Taiwan, Republic of China. Figure 4.2 shows the location of the Te-Chi Reservoir Watershed. ‘ "gMatsu foo ; 7V _, 1 ijatru) . ; . ' ' ‘ a," _. I[ nuvvnrv] ""‘"’—"‘~ 2‘ N i" ° 3° “1"“ ' I East China Sea ”.1 O 10 Kllometers ‘ "{K) ‘ \l (2“! J?! L" [Ill Ufl Chum . I, 00...?) ' T.“'hy.\‘.Chllun‘ i “I. '1", "5 Wuch 'lu Hsu 3 l . ‘ .EJ “{TAIW‘N) 7..., In? (QITupEI ya a, ‘ (‘HINA I - ;L' . Inn New ‘ , Hslnellp. "'0 f~_..’ 1, T I i), “/‘JSuaa (1‘34”) 3“" ,3“ " ““"TAI WAN Jihad/3:,“ " ,I‘ "_ ', [unmen Dao SIM” . Tunphi trtomoleltma " . ,‘u ' (Quemoy) T' Ilehung...‘ “:10- 22(3 '2':3;:m) (“PA") '1. I ‘_ . , (TMW‘NI Chunk”. To a( 7#__A2‘,,N ' I .1, 'I. ' " "* "J" ' Nant' on, 'T‘ 6?)? “unlit-n _‘ L I”? P . 7 TWIN“. “ - on 11» .|’_ 16-. M. L113 1 . , g ‘ _ u '- 3 - 1;}, r/'_I_ . _ _ fl _ _ £F°"'m.,f..h.u.nt': jg, fmufu 3,. “Q. ,1, - - . . Street: of Cancer- C:r;:’ “" P' enghu Ch’tinlao ,7. kg 1' 13.” fl . Fr. (Pestadorex) " e“ 1' (J. 4m I ,.~./’ I ebb \ .T'Iihll M Ch'ouglwng .r-/ ‘11" o l ‘ ‘y p ' “ Phi" i . : Kaolulun 1' "3 L0 Tao PP "c South china I " .Tuni on; Sea } Sea chh'n’u Yu ‘vj Tuwu I Penguin" ' Lon Yri 7- -- , 4~—————r --— MM "End. .fi,‘ L —- 22* N- urs IIB‘E r2 01“" Pl 122 E Copyright © 1994, 1995, 1996, 1997 The Learning Company, Inc. AllRights Reserved. Figure 4.1 Taiwan, Republic of China. Data source: Compton’s Interactive Encyclopedia Deluxe, 1998. 29 Figure 4.2 The Te-Chi Reservoir Watershed in Taiwan, Republic of China 4.2 Terrain and Topography The Te-Chi Reservoir Watershed contains three types of terrain: (1) the Ridge Mountain Range, (2) the Snow Mountain Range, and (3) the Central River Valley. The watershed was formed in the glacial period (Lin, 1974). The Ridge Mountain Range area is located in the east/southeast area of the Watershed. Most of the mountains have a height of over 3,000 meters. Small creeks zigzag through this area, and stream into an over-a—hundred-meter—deep gorge. The Snow Mountain Range area is located in the northwest area of the Watershed. The mountains are between 3,600m and 3,800m in height. Snow Mountain has an elevation of 3,884meters and is the highest peak in the Watershed (He, 1982). 30 ETA"! The Central River Valley of the Watershed consists of the upstream valley of the Ta—Chia River, a gorge-type valley, a horn-like valley, and the down stream multi- streamlet area. The whole valley is at an elevation of over 2,000m. The average slope of the riverbeds in the Watershed has a gradient of 1/60, with a range of 1/90 to 1/36. 4.3 Geology Taiwan is located on the Pacific Convergent Plate Boundary. The western part of the Central Mountain Range of Taiwan lies on the Eurasia Continental Plate, and the eastern part on the Pacific Oceanic Plate. The island of Taiwan was formed due to the orogeny caused by the collision of the two plates during the so-called Continental Collision (He, 1982). Metamorphic and sedimentary rocks dominate the Watershed; the rocks are relatively soft, fractured, and weathered. Extensive erosion occurs because of the intense rainfall and the resulting floods. Frequent earthquakes occur which further undermine the stability of the hill slopes. 4.4 Rivers The stream type in the Te-Chi Reservoir Watershed is short, steep, and ephemeral. The Ta-Chia river is the main stream The Watershed contains nine tributaries: the Ho- Huan creek, the Bi—Lu creek, the Er-Wu creek, the Nan-Hu creek, the Si-Chi-Lang creek, the Xue-Shan creek, the Chi-Chia—Wan creek, the You-Sheng creek, and the Yi-Ka—Wan creek. The Yi-Ka—Wan creek is the main upstream tributary; it receives water from the eastern mountains of the Watershed (He, 1982). 4.5 Climate and Hydrology 31 The climate in the Te-Chi Reservoir Watershed is subtropical. The monsoon prevails from October to March. Orographic rain, torrential rain, typhoons, and thunderstorms cause abundant rainfall from May to September (Anonymous, 1996c). The average annual precipitation in the total Watershed is over 2,500mm, but is over 3,000mm in the Ridge Mountain Range area. The rainfall from May to September accounts for about 76% of the total average precipitation. In 1996, a typhoon caused a 15:77“? total rainfall of 900mm in 9 hours, 1157mm in 12 hours, 1575mm in 18 hours, and 1749 mm in 24 hours. Comparing the 1749m in 24 hours to the 2,500mm in an average year, points to the necessity of considering typhoons in the hydrological modeling of the Watershed (Anonymous, 1996c). 4.6 Land Use The land use in the Te-Chi Reservoir Watershed is: (1) natural forests (coniferous forest, broadleaf forest, and mixed forest), 397km2, 66.0%, (2) forest plantations, 108km2, 18.0%, (3) fruit orchards, 34.5km2, 5.7%, (4) vegetable culture, 34ka, 0.6%, and (5) tea plantations, 0.2km2, 0.03%. Agricultural use accounts for 6.34% of the total Watershed area (Anonymous, 1996d). 32 Chapter 5 METHODOLOGY 5.1 Hardware 5.1.1 GIS Digitizer System The digitizer system consists of a GTCO AccuTab Surface-Lit LII Plus Digitizer with Mounting Base as shown in Figure 5.1. It includes: (1) a surface-lit tablet (model No: 2M-3648AL-l6) with 91 .44cm x 121 .92cm active area, 0.00635mm as (150 lpmm) resolution, and +/- 0.0508mm accuracy, (2) a controller with standard firmware, RS-232 cable, and 9 to 25 pin adapter, (3) a 16-button cursor with 4-meter cable and glass reticule with ultra-fine etched crosshair as shown Figure 5.2. (4) a 3.5-inch disk with ADI, Windows, mouse and WINTAB drivers, and a self-diagnostic program for drive electronics and microprocessor, (5) a mounting base---Dakota Workstation with Monitor (model No: DKT-XJ 1000A), and (6) a 120V power supply (Anonymous, 1996b). Figure 5.1 The GTCO AccuTab Surface—Lit LII Plus Digitizer System 33 C beta: of Pointing Devices Figure 5.2 The GTCO l6-button pointing device (The most left one) The altitude range of the digitizer system is between 0 and 3077meters. The baud rate is between 1,200 and 34,400. The cursor switches use an elastomeric keypad. The system has to be operated at a temperature between 5° C and 46° C and at a humidity range between 10% and 90% (Anonymous, 1996b). 5.1.2 GPS Receiver A DeLorme Tripmate GPS receiver has been employed in this study. A computer with an RS-232 interface with a NMEA-0183 (National Marine Electronics Association) protocol is used to transfer the GPS receiver data. The receiver has 12 channels and a GPS L1 C/A code generator. It has a cold start-up of about 3 minutes at 25° C and a warm start-up of about 1 minute. The unit can operate in the —10° C to 60° C temperature range and weighs (with batteries) is 0.27kg. (Anonymous, 1996a). Using Street Atlas USA 4.0 software, the DeLorme Tripmate can display any location in the US. With AAA Map'n'Go 2.0 software, the DeLorme Tripmate can locate any place in the United States, Canada, Mexico and the Caribbean (Anonymous, 1996a). The DeLorme Tripmate GPS receiver can track 12 satellites simultaneously. Its accuracy depends on several factors: (1) the atmosphere, (2) the ionosphere, and (3) the 34 _ s W position of the receiver sensor. Buildings, natural structures, and heavy foliage can obstruct the DeLorme Tripmate signals, and decrease the accuracy of the user position by preventing the satellite signals from reaching the receiver sensor (Anonymous, 1996a). The DeLorme Tripmate GPS receiver is a SPS device. SA affects the accuracy of SPS (see section 3.4.1.6). With SA, SPS is to within 100m horizontally and 156m vertically. 5.2 Software 5.2.1 ArcView GIS 3.1 and Its Extension Programs 5.2.1.1 ArcView GIS 3.1 ArcView GIS 3.1 made by Environmental System Research Institute, Inc. (ESRI) has the ability to: (1) display data, (2) query data, (3) create data, and (4) use other types of data, e.g. CAD data. ArcView GIS 3.1 displays data by creating a map in a variety of spatial data formats; e. g. the ARC/INFO spatial data formats. It can display tabular data about ground cover, land formation, and water quality to a map. The software is able to represent data on a map by symbolizing and charting the data, by labeling the map with text and graphics, and by choosing map projections. ArcView GIS 3.1 can also design and print various map layouts (Anonymous, 1996c). The data—query function of ArcView GIS 3.1 is useful in obtaining information, and is an essential part of the software. Five sources of information are available to query (1 ) the attributes of the features, (2) the features with particular attributes, (3) the features near other features, (4) the features that fall inside a polygon, and (5) the features of 35 special interest to the user (Anonymous, 1996c). All such data can be aggregated and summarized into relevant statistics, making the data easier to interpret. ArcView GIS 3.1 can create new data by: (1) developing additional spatial data, (2) editing existing spatial data, and (3) digitizing a map. The software creates new spatial data either by developing a new point theme, line theme, or polygon theme, or by using a digitizing tablet to digitize a map into a point feature, a line feature, or a polygon feature. Editing spatial data can be done within certain themes. ArcView GIS 3.1 data is compatible with several other data types, e. g. image-type data, Computer-Aided Design (CAD) data, Spatial Database Engine (SDE) data (Anonymous, 1996c). Image-type data includes scanned-map data, aerial-photograph data, and satellite-imagery data. CAD data are a set of non-GIS graphical data for engineering or architecture design, and can be employed as GIS data in ArcView GIS 3.1. With the spatial Database Themes extension of ArcView GIS 3.1, SDE data can be added to the database of a map in order to explore, query, and analyze the map data in ArcView (Anonymous, 1996c). In this thesis, four additional software programs are used with basic ArcView GIS 3.1: (1) ArcView Spatial Analyst, (2) ArcView Tracking Analyst, (3) ArcView 3D Analyst, and (4) ArcView Image Analysis. 5.2.1.2 ArcView Spatial Analyst The software ArcView Spatial Analyst can create, query, map, and analyze cell- based raster data and perform integrated vector-raster analyses. Its specific function is to provide solutions to problems that require distance modeling or other continuous-surface modeling. The software can generate surface representations from various data sources, 36 and develop new information from the overlaying of multiple theme types. The major functions of ArcView Spatial Analyst are (Anonymous, 19960: (1) to convert feature themes (point, line, and polygon) to a grid theme, (2) to create raster buffers based on distance from feature or grid themes, (3) to create density maps from themes containing point features, (4) to create continuous surfaces from scattered point features, (5) to create contour, slope, and aspect maps and add hill-shades, (6) to perform Boolean queries and algebraic calculations on multiple grid themes, (7) to perform neighborhood and zone analyses, (8) to select a special interest grid on a feature, and (9) to produce a grid display. 5.2.1.3 ArcView Tracking Analyst GPS technology offers an inexpensive, quick, and convenient way to collect data. The ArcView Tracking Analyst software can instantaneously display real-time GPS stream-data for temporal-data and spatial-data. The data which the Tracking Analyst software can import includes (Anonymous, 1998c): (1) real-time GPS tracking stream- data, (2) real-time logged-in data. 5.2.1.4 ArcView 3D Analyst The ArcView 3D Analyst software can produce: (1) Triangulated Irregular Networks (TIN), (2) three-dimensional vector geometry, and (3) interactive perspective views. Specifically, the software is able: (1) to develop and modify surface models, (2) to develop 3D shapefile themes, (3) to edit TIN files, and (4) to planimetrically display surfaces (Anonymous, 1997c). 5.2.1.5 ArcView Image Analysis 37 The ArcView Image Analysis software processes and thereby enhances the display of an image, in particular: (1) satellite imagery, (2) aerial photography, and (3) remotely-sensed data such as infrared thermal imagery. The image processing can include: (1) image-to-map rectification, (2) spectral and color enhancement, (3) automatic feature boundary detection and extraction, (4) detection of the change in same-location images shot at different times. and (5) establishment of shapefiles by feature extraction (Anonymous, 1998 e). 5.2.2 Microsoft Excel 97 The dBASE file (dBASE III or dBASE IV) of Microsoft Excel 97 can be directly loaded into ArcView GIS 3.1. The dBASE files save only the text and data values contained in the cells of an active worksheet. The cell formatting, page-layout setting, graphics, and other Excel features are lost. Only the information stored in a limited number of columns is saved: (1) in dBASE III, 128 columns, and (2) in dBASE IV, 256 columns (Anonymous, 1996g). In dBASE III and IV, only the range of cells called “Database” can be manipulated; it is necessary to redefine the range to include new cells. The values in the first row of data in Database determine the data type in each column. If the first cell in a row contains a blank value, all the cells in the row are read as text fields (Anonymous, 1996g). 5.3 Methods Simulating the water runoff in a watershed requires the modeling of multiple “layers” of spatial and temporal data. In this thesis GIS, SVIS, DGPS, GPS, and RS are integrated with forest hydrology to simulate the water-runoff. The main steps are: (1) 38 digitize the watershed maps by GIS (Anonymous, 1996c), (2) modify the GIS maps by GPS and DGPS (Parkinson, 1996), (3) process the aerial photographs by GIS (Anonymous b, 1998), and (4) simulate the water runoff by GIS-SVIS (Arnell, 1995). Digitizing the GIS watershed maps requires information on the watershed environment and properties. The maps should represent the topographical characteristics of the stream/water-body system, the topography, the soil type, the land use/cover, the geological properties, and the hydrological record. The digitizing of the Te-Chi Reservoir Watershed map is described in detail in Section 5.3.1. To process the aerial photographs of Te-Chi Reservoir Watershed by GIS requires the integration of GIS into a desktop-mapping system. This provides a realistic backdrop for the geo-referenced database and updates the terrain environment and recent natural and man-made changes (Anonymous, 1998b). The details are given in Section 5.3.2. To correct a GIS map by GPS an 39 121°26’29.82”. The physical properties of the Watershed maps were obtained from the Water Resource Bureau, Ministry of Economic Affairs, Taiwan, Republic of China. The combined GIS maps digitized by ARC/INFO contain information on the topography, land use, streams and water bodies, geology, and the road features of the Watershed. The specifications of the digitized paper maps are listed in Table 5.1. Table 5.1 Specifications of the Digitized Paper Maps of the Te-Chi Reservoir Watershed Information Types Information Map Year 1985 Map Unit Meter Map Projection Transverse Mercator’ Central Meridian 1 2 1 °00'00" * Transverse Mercator projection is similar to the Mercator projection except that the cylinder is longitudinal along a meridian instead of the equator. The zero elevation is at sea level in Keelung, Taiwan, Republic of China. The cylindrical projection of the Transverse Mercator projection is longitudinal along a central meridian line and not along the equator. The result is a conformal projection. The central meridian is centered on the particular region which is to be highlighted. The centering on a specific region minimizes the distortion of the properties in the region. The north-to-south projection of the meridian is best for north-south landmasses (Anonymous, 1996c). 5.3.1.1 Preparing for Digitizing of A Watershed Map The GTCO AccuTab Surface-Lit LII Plus digitizer (2M-3648AL-16) with Mounting Base (DKT-XJ 1000A) and ArcView GIS 3.1 software are used to produce the digitized GIS maps. Before digitizing the maps in ArcView GIS 3.1, several steps are required: (1) setup the digitizer system, (2) prepare the paper map, (3) prepare the digitizing, and (4) register the coordinates of several points on the paper map. 40 The digitizer setup steps include: (1) check the compatibility of the digitizer system and the software, (2) install the up-to-date digitizer driver, (3) setup the driver control panel, and (4) configure the button of the digitizer puck-buttons (Anonymous, 1996e) The latest version of the WinTab digitizer driver is used to operate the digitizer in the Microsoft Windows environment. It should be noted that ESRI successfully tested the digitizer system under ArcView GIS 3.1 (Anonymous, 1996 a; Anonymous, 1996b). Configuring the digitizer puck buttons on the WinTab control panel under ArcView GIS 3.1 is an essential step. The GTCO digitizer has 16 puck-buttons which collectively act as the digitizer-pointing device. In order to configure the button properly, ArcView GIS 3.1 requires the digitizing tablet to be able to be toggled in two modes: ( 1) in absolute (digitizing) mode, and (2) in relative (mouse) mode. In absolute mode, the location of the tablet is digitized to a specific location on the screen. The movement of the pointing device on the tablet surface causes the screen cursor to move to the same position as on the monitor. The ArcView GIS 3.1 user interface cannot select the menu choice, buttons, and tools when the screen cursor is locked in the drawing area of the View. In relative mode, the tablet-pointing device functions like a mouse. However, there is no correlation between the position of the screen cursor and the surface of the digitizing tablet. Preparing the paper map entails minimizing the map distortions. Two steps are needed to prepare a paper map: (1) select an reliable, up-to-date, and unfolded paper map, and (2) pre-choose the exact coordinates of at least four control points. 41 The preparation for digitizing requires loading of the Digitizer Extension programs, and proper selecting of the same map units and the same map projection (See Table 5.1) (Anonymous, 1996c). Registering the paper map specifies the global location of any position on the map so that the desired features in geographical space are digitized (Anonymous, 1996 a; Anonymous, 1996 e). There are two conditions for registering a paper map: (1) digitize at least four-control points on the map when it is first registered, and (2) use the TIC files (tic.dbf) of ARC/INFO software to register the control points on the map. In this thesis, thirty-eight comer points (i.e. the points intersected by the straight lines of latitude and longitude on the map) and forty-eight mountain-peak points are saved in a TIC ARC/INFO file in order to register the map (see Table 6.3 in Appendix E). 5.3.1.2 Digitizing the Features of the Watershed Paper Map ArcView GIS 3.1 allows the user to digitize a feature on a watershed paper map in three different ways: (1) by digitizing a feature, i.e. a point, a line, or a polygon feature into a new theme, (2) by editing an existing theme and make it active, and (3) by adding specific information to the View. Each feature can be digitized into two modes: (1) by the point mode, and (2) by the stream mode (Anonymous, 1996c). In the ArcView GIS 3.1 interface, a certain shape feature can be developed and saved into a shape file by combining the “Editing and Drawing” tools with a certain shape theme, such as a point, a line, or a polygon theme. Each shape file contains only one feature. The “Shape Properties” setting can be used to view or edit a shape property of any selected feature. 5.3.1.3 Correcting the Digitizing Errors 42 The digitized spatial data needs to be confirmed and rectified. By using the programs in the ARC/INFO software, the digitizing errors can be identified and corrected, and the topology can be reconstructed (Anonymous, 1997a). The details of the correction are contained in the user guide of ARC/INFO. 5.3.2 Modifying GIS Maps by GPS, DGPS, and Relative Positioning Surveying The digitized GIS map can be modified: (1) by the Total Stations and Relative Positioning Surveying option, or (2) by the GPS, DGPS, and Relative Positioning Surveying option. The second method is used to modify the digitized GIS maps, because the GPS receiver is cheaper, slighter, and more convenient to use for a steep and mountainous watershed (See Section 3.4) than the Total Station option. Employing GPS and DGPS to modify the GIS maps is a challenge because the DGPS service regions do not include Taiwan. However, the GPS do receivers still receive the signals of the GPS-satellites. Two identical DeLorme Tripmate GPS receivers, DGPS technology, and Relative Positioning surveying technology are used to modify the GIS maps. The objectives are: (1) to cancel the errors common to both GPS receivers, (2) to find the bias of each GPS receiver, and (3) to find the relative positions of the measured points (Capaccio, 1997; Leick, 1995). Section 3.4.2 elaborates the details. Microsoft Excel 97 is used to develop the statistical program (See Table 6.4 in Appendix F) and ArcView GIS 3.1 is to represent the modified points on the GIS map. The statistical program uses the difference of n-variables in a normal distribution to calculate the bias of the GPS receivers and the differential values of the input coordinates of the digitized maps. The numbers of sampled points (N) should be at least 43 thirty points. The degree of free (df) is equal to N-2. In this thesis, there are two variables X and Y, and thus the df is N-2. The two-dimensional sexagesimal-notation coordinates are latitude (X) and longitude (Y) and are measured by the GPS receivers A (XA, YA) and B (XB, YB). They are input into the statistical program which automatically converts the sexagesimal- notation coordinates into decimal notation and calculates the mean (U m MB), UY 75% Area (m2) 54993274 4681 1393 181139023 257071232 Percentge of the Total Area 9.14% 7.78% 30.1 1% 42.73% According to Table 5.3, the area of the over 30% slope accounts for about 90% of the total watershed area. The steep slopes increase the water velocity and results in changes of flow direction and flow length (Hudson, 1981; Kirkby and Morgan, 1980). The land use/cover mainly affects the runoff movement and the flow accumulation although at times a typhoon and the monsoon from the Pacific Ocean also strongly affect it (Anonymous, 1970 ~ 1996c). Note: typhoon and monsoon affect are not considered in this thesis. Due to the long rainy season and ample precipitation in the Watershed, the soil frequently approaches the water saturation status (Anonymous, 1970 ~ 1996c). The valleys have thicker topsoil than the peaks in the high mountain areas. Most of the t0psoil in the valleys readily reaches the saturation status and thus rarely affects the runoff movement in the Watershed (Hudson, 1981; Kirkby and Morgan, 1980). The soil factor rarely affects the runoff movement in this watershed. Most of land in the Watershed is in the forest land. Considerable evapotranspiration occurs in the forests resulting in the high humidity. The high hmnidity results in the higher pressure of the water vapor in the air because the air is saturated with 48 water vapor (Merva, 1995). The evaporation of the runoff is little affected in this watershed. 5.3.4.2 Cell Runoff (mm) The cell runoff (mm) is obtained by using the area-weighted method, which measures the stream-flow as a volume per unit time (cms-day) (Arnell, 1995). The steps to find the cell runoff (mm) are: (1) find the cell flow (mm) of the average annual runoff (cms-day): (a) determine the runoff generated by the drainage measured at the hydrological station (i.e. by subtracting the volume of runoff measured at the upstream station). See Table 5.2 in Appendix A. (b) select the average of the annual runoff (ems-day) of each station and use the method of the area-weighted cell runoff to find the (cell runoff (mm). The cell runoff (mm) is of Type A. (2) find the cell flow (mm) of the median annual runoff (cms-day): (a) find the frequency histogram of the average annual runoff (cms-day) for each hydrological station from Table 5.2 (See Appendix A). (b) select the median number of the maximum frequency range of the annual runoff (cms-day) of each station. (c) find the cell runoff (mm) by the method of the area-weighted cell runoff by using the median number of the maximum frequency range of the annual runoff (cms-day). The cell runoff (mm) is of Type B. (3) find the cell flow (mm) of the maximum daily runoff (cms-day): (a) select the maximum daily runoff (ems-day) of each hydrological station from 49 Table 5.4 (See Appendix B). (b) find the cell runoff (mm) by using the method of the area-weighted cell runoff (mm). The cell runoff (mm) is of Type C. (4) compare the cell runoff (mm) of type A, type B, and type C. (5) use ArcView GIS 3.1 and its extension software to produce the grid cell map, the flow directions map, the flow accumulation map, and the flow length map. 5.3.4.3 Cell Size (m) A cell size should be at least one quarter of the minimum mapping unit in order to simulate the runoff movement of a certain landform with the smallest area. The minimum mapping unit can be found in the attribute table of the land-use in the Watershed (See Section 6.2). 50 Chapter 6 RESULTS AND DISCUSSION 6.1 GIS Watershed Maps 6.1.1 Unmodified GIS Watershed Maps The GIS digitizing of the Te-Chi Reservoir Watershed created five ArcView shapefiles which contain the spatial data of the vector or raster data type. The shapefiles are non-topological files in which the geometric locations with their attribute information and geographic features are stored. The information in the files defines the geometry and the attributes of the geo-referenced features. The file extensions of the five shapefiles are: (1) shp, (2) shx, (3) dbf, (4) sbn and sbx, and (5) ain and aih (Anonymous, 1996 a). The “shp” files store the geometric features of the Watershed area. The “shx” files contain the index of the area. The “dbf” files are the dBASE files which store the attribute information of the features (the data can be displayed as a feature table). The “sbn” and “sbx” files store the spatial indices of the features and are created when the “theme on theme selection”, “spatial join”, or “create theme” is chosen and is executed in the Shape field. The “aim” and “aih” files store the attribute indices of the active fields in a theme- attribute table and are created when the “Link on the tables” is executed (Anonymous, 19966) The unmodified GIS map of the water bodies in the Te-Chi Reservoir Watershed is digitized in line feature as shown as Figure 6.1. The flow directions of the streams are from north to south except that the streams in the south are from south to north. The map shows: (1) the network of the streams, (2) the stream lengths of sub-tributaries, and (3) the flooded areas of the Te-Chi Reservoir. Its attribute database is contained in the 51 NH 230505. N a a Z 8:... . 23 . V... 2:. .8" .\ /\ .8“ . «3. \K «8. . SN. . on". . 2b \/\ o3 . 8. \./x. 84 . B , .. .8535: .E .523. 233535.33 .3585... 2.3.... :8: .355. 20..» D €3.85... a....8...8..£h 5... a. u..- a 82¢ 20.35..» a .3233. .86.. 323mm". combo... d 2...: 32.3325 .3233. Eco... 9: 5 8.8m .32... 3 gas. 20 85.82:: 2: E 2:9“. 52 “aat.dbf” file, i.e. the Arc Attribute Table (aat) the “dbf” file format. The “aat” table of the Watershed is presented in Table 6.1 (See Appendix C). The unmodified GIS map of the vegetable fields in the Watershed is shown in point feature in Figure 6.2 and in polygon feature in Figure 6.3. Their attribute database is contained in “pat.dbf”, i.e., the Point / Polygon Attribute Table (pat) in the “dbf” file format. The map shows the various vegetable farms. The database contains information on both point and polygon features (See Table 6.2 in Appendix D). Figure 6.2 indicates the center and size of a particular area, and Figure 6.3 shows its shape. Combining Figures 6.1 and 6.3 results in Figure 6.4. It shows that the vegetable fields are distributed along the main tributaries of the Ta-Chia-Chi River in the Central River Valley area (See Section 4.2). Figure 6.5 shows the unmodified GIS map of the cultivated agricultural areas in the Watershed, i.e. the tea plantations, fruit orchards, vegetable fields, and farms which cover the regions adjacent to the Te-Chi Reservoir flood area. 6.1.2 Modified GIS Watershed Maps The twenty-four measured points which are registered on the digitized GIS maps are shown in Table 6.3 (See Appendix E). Their locations are quantified at least thirty times, and were statistically analyzed to establish the biases of the GPS receivers. The biases were used to find the relative positions of a group of points. The author developed a unique method to represent the corrected location-coordinates with statistics in Table 6.4 (See Appendix F) (See Section 5.3.2). The method is explained in the next paragraph. The modified GIS map of water bodies and vegetable fields in the Watershed is shown in 53 230526. N a N 2 reason D 833 . N513. N33; . own: 2&3 . whoa... 2.3m . mNmVN mNnvN - «2.3 33.. . name name . EN 33:025.: :32: 32:3 625 used—Ea“. 0.339) A238“. EB... 8:225 .858”. .5...» 2.. a. menu“. «33%; .o 9.2 20 8585:: 2:. «a 2:2“. 54 E 230525. N o N Z 85." . £33 a 2.82 - n33 mm.“ 23a . 28» H.” 3.8» . 2.3.“ D on?" - 8a. 33. . 38 \L . 88. E ”u «M: .2355 :82: 22:3 62$ 93.25.“. 038000) ¢ 3 l N . 1 .233... cog-on: ages“; .358”. 20.2 2.. 5 mean. 233025 .0 32 29 35.85:: 9: 3 95?. 55 nun 830825. N o N Z é . N53: N53. . an: 0:.- . as... as... . 0N3 0N3 . gr 83. . 88 88. :u =55!!! .5. to... .123 93.93.: .3} 88. . 28» .\ 28 . :3 ,, SI . :32 «8.. 8:\/\ 82 . 8.2 z. . .8\/\ 8.. E x .353...- .I .533. 2...... €8.88... 21!! v8... .358: 29.» .3"... 2:15....» :1: I v... .25 30.1.8.8 .- 35.... 30.... 39.33.... gonzo; d 2...: 62.2355 32:33. 20.: 05 c. 2E: c35ouw> .28 350m .353 he 3.2 26 35008:... 2:. v.0 2:2”. 56 H 23050:! N o N 633.8. 2E: a €88.33 3» I .3333 .2220 :2“. _...¢. 235.355: 9.125.“. .333; D .3285... 93.8535 5.: a. a... n“ 32¢ 29.20..» a €2.83. 20.... $233... cog—on. can 05.: 3:92.25 .3233. 20$... 2: E «32 _u..3_:utm< 335.30 9: no gas. may 0:... m6 2:9“. 57 Figure 6.6. The difference between Figures 6.4 and 6.6 cannot be distinguished visually but is obvious from the data in Table 6.3 (See Appendix E). Assigning units and applying conversion factors is done in the Representative Fraction (RF). The scale of the digitized maps is 1:25,000, and therefore 1cm on the map equals 250m in the Watershed. Also on the map, 2’30” latitude (X) equals 16.9cm and 2’30” longitude (Y) equals 18.45cm. Thus, each 1” latitude (X) equals 28.17m within the watershed and each 1” longitude (Y) equals 30.75m. According to the differential distances in Table 6.3 (See Appendix E), the maps do not represent the actual locations very well unless corrections are made in the coordinates values. The RMS error is the distance between the starting location and the actual location of a position (Anonymous, 1996c). It is calculated for at least four sets of two control points defined in the Image Analysis theme and in the corresponding feature theme. In ArcView GIS 3.1, an acceptable RMS error is lower than 1. The RMS error represents the accuracy of the digitized map but not the accuracy of the actual location. Thus, map corrections are necessary. 6.1.3 Aerial Photographs Processed into GIS-SVIS Maps Figure 6.7 shows an aerial photograph of the Te-Chi Reservoir and the surrounding area; the figure is made up of forty individually-scanned aerial photographs which have been merged and saved as a TIFF file. ArcView GIS 3.1 along with the Image Analysis software was used to process the aerial photographs in Figure 6.7 into a GIS-SVIS image theme (See Figure 6.8). With the application of the Seed Tools program in the Image Analysis software (Anonymous, 1998b) and the Interpretation Factors, four types of the landform can be 58 MU £32.35. N o N Z Sank . «=3. 2...: . 2... ..... 2... . 2.8. 2.8. . 8.3 . , an... . 8.2 8... - a... a... . In .328... .5... 2...... .82. 3.... .38...) 8.2. . 2... , 1.. 3... 3.... 8.. . «n..\/\ a... . .3. \< .8. - 2.. \< 2.. . 8. \< 8. . k \< .858... .E .52.... 5.8383... .853... .3. .8: .35.... 20... HI .3235. 33.8.5... 5... a. u... E .25. 20.35.... _. .35.... 20...— Ammsaaom couzoa new as... .3553; .3233. 50...... m5 5 2:5... snfiaoo> can 3:25 .325 .0 gas. £0 35022 2:. 9o 2:2". 59 Figure 6.7 The Aerial Photograph of the Te-Chi Reservoir Watershed and Its Surroundings Smoothnew_dele1.img N :Layer_3 :Layer_2 :Layer_1 60 .53.... 9.2.5. 29.20.... .- 35.! 20.... «5.36 3 .388... .228...» .3581 .5... .53 £8.82... it... 2:. .6 2.6.“. 61 distinguished in Figure 6.8: (1) the blue region of the Te-Chi Reservoir and the Ta-Chia- Chi River, (2) the curved grayish black lines of the terraces along the Reservoir (Paine, 1981), (3) the curved white lines of the tributaries flowing into the Reservoir and the River, and (4) the regions of the various vegetation types (Ulliman, 1995). The gray- toned irregular patterns represent the coniferous forest lands, and the grayish white regular areas represent the broadleaf forest lands (Avery and Berlin, 1992). Comparing the conditions of the cultivated agricultural areas between Figure 6.5 (Year 1985) and Figure 6.8 (Year 1987), no change is seen. 6.2 GIS-SVIS Maps Illustrating Water Runoff In a watershed with steep slopes, the choice of the cell size should be such that the runoff movement can be simulated for the smallest area of a specific type of land use and land cover, i.e. a cell size with at least a quarter of the minimum mapping unit can simulate the runoff movement in a certain landform with the smallest area. According to Table 6.5, a fruit orchard accounts for the smallest area, i.e. 137.38m2 (11.7m x 11.7m). Therefore, the grid size is set equal to 5.86m for producing the slope grid map and the land-use grid map. Table 6.5 The Smallest Area of the Various Types of Land-Use in the Te-Chi Reservoir Watershed Type of Land Use Farm Forest Landslide Orchard Tea Plantation Vegetablmeld Area(mz) 1601 I601 287 137 1364 271 The cell size of 5.86m-by-5.86m is small compared to other grid maps using, e.g. the USGS DEM map, which uses a cell-size of 2km-by 2km. A small cell-size is more precise in simulating water runoff in a steep, high-elevation watershed than a large cell- size. 62 88 82 82 . £2 82 as. 29 32 I ll 1 It. .l 1'11 1'. l r l .17.: #0 _ z ' _ ! . 6’. b. _ (V «v .. . ) )J’ coon < _ x, . 1 M _ . 000. M _ _ H a 88 . Sigh . z . 82,655+ H 88 :98ch 85.53: + 022.93; 93.50.35 Ill . 008? 322...? z _ F 2Iém2+ _ K _ _, \ . 88F . a r _ 8o: . 88. _. 88? 852255 .6230”. 20.3 2. do 2255 .3392: «5 5 :83 .255 9:. we 9:3 (Rep-sum) young lenuuv 63 3 z: . 2:. /._.. SS. 582 83.32 «873.22 $5.232 «5:52 84. $2 :32: .593... 8:53.355 530:2. 5:953: o 31.52 a 32.95» c 3.55 . 53:6: 0 9.3-20.3» 0 53 2.8.20 . 225» 3.3.22: I1! 23050:! N a N no.2 09855 5:16: 3.4. $9.65 33.62012 539.335 .3233. 29$ 65 E 2235 3030.29»: 05 *0 26:30.. 9:. 2.6 952“. 64 83. . 2:. .. 23. Ban .1 Ema - «8. \< «8. - 8% \< an". . 2.0 \< 2.0 - 8.. \< 8v- 8 \< :82: .505... 82535.23. c285... 5.5.53: 3.75: 0.22.95» 3.55 531.0... 93.20.30 .55 620.20 :35» 39.06.23: 83050:! 0.0 o «6 II ill -111 .1/1, 531.017.. u. p/Omz-m::m 311502 ,.| -1 -.1. / communicmcwémzz / .1 1 1 93-20-30 5362.020 nos—0.0003 052000”. 20.0... 05 E 622.30 2666.22... 65 06 922 0.0 2:. :0 659.. 65 The runoff data over a twenty-seven-year period in seven hydrological stations in the Te-Chi Reservoir Watershed is tabulated in Table 5.2 (See Appendix A) and is illustrated in Figure 6.9. The data indicates the average annual runoff (ems-day) and its frequency of occurrence. Figures 6.10 and 6.11 show the locations of the hydrological stations. Table 6.6 Average Annual Runoff (mm) (Type A) in Seven Drainage Areas in the Te- Chi Reservoir Watershed Hydrological Station mm Am 1111.2) Aungc Amunl Rimmamdm Arm Manna by m Suion (hill) Rumle 1.1mm Arm tans-thy) M(m) Chih-Lo 77 2147 77 214 7 2396 Chi-Chia-Wan 11 I 1948 11 1 1948 1520 Saran-1mg I 56 3208 46 I 26 1 2379 Nan-H11 126 2240 126 2240 1540 Ho-Huan 129 2465 129 2465 1657 Hum-Shah Junction 258 5019 4 314 7456 Sm-Mao 4 l 7 9078 3 850 26796 Table 6.7 Medium Annual Runoff (mm) (Type B) in Seven Drainage Areas in the Te- Chi Reservoir Watershed fiddog'rd 9mm Wham?) mmimdm m1m1 mmwmmm mammmm memo Gib-lo 77 1211411) 77 2113 2457 0%]an 111 11114711) 111 14:0 1124 321101-133 156 ram-mu) 46 1135 3841 Nn-Hi 11) 31m~31m 123 . 3045 1194 1~b-Hm 129 151D~ZID 129 1677 11(1) HmSmhuin 258 611114111) 4 1547 3672) 9.11M» 417 9.11M!!!) 3 410 13559 Table 6.8 Maximum Daily Runoff (mm) (Type C) in Seven Drainage Areas in the Te- Chi Reservoir Watershed Hydrological Station Damage Am- (Lml) Mxinmm 1x111. Runoff(umn-dufl Am. 1.1.6.6166 1.. (1....ng 31.11.... (111.21 lean m mam-.1 Am. (ems-dry) Rummnun) Chili-Lo 77 191 77 191 213 Chi-Chia-Wan 11 l 181 111 181 141 Szu-Chi-Lang 156 267 46 85 161 Nan-Hu 126 214 126 214 I47 Ho-Huan 129 I49 129 14‘) IOO Huan-Shan Juncuon 258 640 4 276 6552 Sun-Mao 417 904 3 48 1521 Table 6.9 Three Types of Area-weighted Runoff (mm) in the Te-Chi Reservoir Watershed Cell Runoff Type Type A Type B Type C Cell Runoff (mm) 1961 2156 200 Type A: Average Annual Runoff Type B: Medium Annual Runoff Type C: Maximum Daily Runoff 66 The elevation of the Te-Chi Reservoir Watershed is between 2,000m and 3,884m (Lin, 1974). The orographic rains and the local thunderstorms occur frequently (Anonymous, 1996d), and therefore the use of microclimatological data is essential for hydrological simulations. Table 6.6 shows the average annual runoff (Type A), Table 6.7 the median annual runoff (Type B), and Table 6.8 the maximum daily runoff (Type C) of the seven stations. Figure 6.11 is a reference used to mark the boundaries of a particular hydrological station. The area-weighted runoff method along with the information from Tables 6.6, 6.7, and 6.8 is used to calculate the three types of average runoff (mm) in the Te-Chi Reservoir Watershed. The results are shown Table 6.9. Type B runoff is larger than the runoff of Type A. Therefore, type A runoff can be ignored and type B runoff is used to simulate the water runoff movement in the Watershed. Type C runoff (200.08mm) is a reflection of the maximum daily rainfall in the watershed. Therefore, when torrential rain is simulated, the simulation uses the runoff of type C. The slope-type and land-use are two topographical factors affecting the water- runoff in the Watershed. For the design of the watershed-runoff system, an in-depth analysis of the water-runoff in the watershed has to be made. Thus, the GIS-SVIS maps of the slope-type and of land-use of the water-runoff in the watershed have to be available. 6.2.1 Slope-Type Factor 67 “in 0 28222.: N o a 2 $2. A U $2 1 $2 WU $8 1 $8 a $3 1 $8 H“ $8 1 $2 I $0.. 1 $0 UH $0 1 $. fl. .8250 .225. $6 .HU 6262.60. .25 86.0 6622262. 22:33. 201; 2.. 5 25.2.6.0 .6 an... 0.0 2: 2.0 «so... 68 The two-dimensional GIS map of SIOpe-type in Figure 6.12 was used with ArcView GIS 3.1 to produce the three-dimensional GIS-SVIS map of slope sun-shade in Figure 6.13. The GIS-SVIS map of slope sun-shade (Figure 6.13) was combined with the GIS map of slope-type (Figure 6.12) to make the three-dimensional GIS-SVIS map of slope-type shown in Figure 6.14. By employing the GIS-SVIS maps of slope-type (Figure 6.14), water runoff was depicted by four GIS-SVIS maps: (1) the three-dimensional GIS- SVIS map of flow directions in Figure 6.15, (2) the three-dimensional GIS-SVIS map of Sinkholes in Figure 6.16, (3) the three-dimensional GIS-SVIS map of flow accumulation in Figure 6.17, and (4) the three-dimensional GIS-SVIS maps of upstream flow-length in Figure 6.18 and of downstream flow-length in Figure 6.19. 6.2.1.1 Slope Sun-Shade GIS-SVIS Map The results of the computation of the sun-shade in the Watershed were used to illustrate the hypothetical illumination of a surface on a graphical display. For analysis, the display can be used to determine the length of time and intensity of the sun at a particular location. The brightness of the graphical display is an indication of the type of weather. The elevation data of 271 mountain-peaks collected from the contour maps of the Watershed were added into the tabular elevation-data (48 mountain-peaks) of the GIS map of slope-type (Figure 6.12). These 319 mountain-peak elevation data were used to achieve the different effects by tentatively adjusting the azimuth and altitude parameters. An azimuth at 315° (the position of the sun in the northwest of the Watershed) and an altitude at 25° (the sun shadow computed at 4:20 PM) were used to produce the GIS- SVIS sun-shade map. 69 LL) 2*m 3 0 h V V N N IIiIBDDDE 3m .53... are. 82.9.... H 83.505. a o N €292; 2.. .o .8352 a... c. "cam 9.. .o 33.8.... 3:93.25 .3233. 20-6... 05 E s: as a ouazwcam 2.2m 3 3.2 £5.96 2: 9... 2.5.“. 7O H m 23.52.: a o a 2 $2 A D $2 1 $3 ! $8 1 $3 .HH $9. 1 $8 y $8 1 $3 I $3 1 $... my $... 1 $. fl. 38:5 .225. $.. D .58.» 6.5. 8%» 2.2m unseen; .3233. 29.: 2.. a. 83.2.2». .0 nus. m_>m.m.o 2: 3... 2.6.". 7'1 .1133 w 9.8.52.8 N a u .3 . a: D «2 . 5 y 3... . 8. D 3“ . 2“ .U .586 "0:0— 329 Gaza->2” Clad—OK 52.3225 .3333. .206... 2.. :. :o=a>a_m 03.23. .0 an: m.>m1m_0 2:. 3.6 959.... 72 .113]. .1 28.30:! N o N 332 . 3am I 33» . 88 my .53.» 6.5 $.00 .0 .8952. . 32¢ oohaoWEuohw 3:222.) .3233. 20.0... 2: c. . ocoagofi 2523. ecu 32¢ 350983.50 no as: m.>m.m_6 2:. 2.6 8.6.“. 73 H £30555. N a N 38 . 82 E. 32 . 2.... 28 . 2.3 I can . 88 y 0N0». 23 m. nFoN 1 .32 I hour 1 o I .58.» 6.6 .52... 59.3.3... 02.2325 32:36”. 20$... «5 :. £23.32“. 3 as: 25.90 2: =6 2:9“. 74 m ”run 9.0.0805. N o N Z .88. «2... 8.3.2.3 0.5 02.03.0000 .3233. a «802.0 2.... :0_.a...o>o.....2... 0.002 uoaonxm 2...“. 2.3.3.; 05.320 8:35.... 3.. I 22.20 .3... i 5.3.5.... .83... D .83.. .85.... I .83.... .8535 7. .020". 2.0.2.50 I 30.02.00. 02...... 00:05... UUUIHUULI 02.20.25 .3333. 20.0... a... 0. 32.9.... .o an... m... 2: 2... 2.5... 75 Z .33. no.3 3:35.. 3.0 3.82350 .3230: a 239.3 g 2.... :o:3:.>o..a.2_n_ Fm 0392 932.5 I 2.... .338; ' ecu-3:0 ' gases... .8. . 22.9.0 «.9... 5.55... .83.. B .83.. .85.... I «no.3... 03.305 .33“. 2.22.50 9....» .Eood "2.5. on: E...— x.--1m. _ 28.52.! N o « .3533; .3230: 20.: 2: c. 25.2... 3 an... 25.20 2F a; 95a... 76 The azimuth angle, altitude angle, and representative-fraction size were used to display the virtual reality of a GIS-SVIS map. By changing the angles of the azimuth and altitude, the three—dimensional visualization of a watershed could be displayed. Also, by changing the scale of a GIS-SVIS map, the far-and-near distances to an object were visualized, i.e. the smaller the scale, the larger an object (See Section 3.3). In Figure 6.13, the fine black lines with a shade index above 245 are streams. The remaining areas with a shade index between 245 and 246 are the high-mountain areas an elevation between 2,000m and 3,884m. 6.2.1.2 Relative-Elevation GIS-SVIS Map The flow direction in a cell was calculated from its relative elevation viz a viz eight surrounding cells, using the relationship: drop = changed elevation (Z) / distance x 100. Distance is equal to the distance between two cell centers, i.e. since the cell size is set to 5.86m, the distance between two orthogonal cells is 5.86m and between two diagonal cells is 8.29m. If the gradients to the adjacent cells are the same, then the water runoff to the neighboring cells will be the same. If the gradients to the adjacent cells are dissimilar, the water will flow to the neighboring cell at the lowest elevation (Greenlee, 1987) The GIS-SVIS relative-elevation map in Figure 6.15 was used to calculate the flow-directions of the water in a central cell to its eight-neighboring cells. Each cell is assigned a Z value, which ranges from 1 (the lowest elevation) to 255 (the highest elevation). If a central cell is lower than its eight-neighboring cells, then it is given a value lower than its eight-neighboring cells, and the flow direction is thus towards this cell. If multiple-neighboring cells of a central cell have the same low value, these cells 77 are still assigned the lowest value, i.e. then a central cell has the same Z value in multiple directions, and thus the occurrence of a one- or multiple-cell sinkhole is eliminated (See Section 6.2.1.3) (Anonymous, 1996f). The water flow in the Te-Chi Reservoir Watershed depends mainly on the topography of: (1) the Ridge Mountain Range (area A), (2) the Central River Valley (area B), and (3) the Snow Mountain Range (area C) (See Section 4.2). Figure 6.15 shows the topography of the three areas. The total area of the pink regions (drop value between 219 and 255) in Figure 6.15 represents the largest area in the Watershed (it accounts for about 70% of the total area). The pink regions are distributed over the entire watershed. The area of the yellowish brown regions (drop value between 183 and 218) is the second largest area in the Watershed (it accounts for about 14% of the total area). These regions located at the second highest elevation in the watershed receive the water runoff from the pink regions and move their runoff toward the green-, blue-, purple-, or red- colored regions. The red-colored regions (drop value between 1 and 37) constitute the third largest area in the Watershed (these account for about 12% of the total area). Thus, the water runoff in the areas A and C flows toward area B, i.e. to the southeast of the Watershed. By comparing Figure 6.14 with Figure 6.15, some regions represent an abnormal phenomenon because it applies that water in some areas flows from areas with a lower slope toward areas with a higher slope. The Sinkholes cause this “phenomenon” (see next section). 6.2.1.3 Sinkhole-Areas and Relative-Elevation GIS-SVIS Map 78 A sinkhole occurs when all neighboring cells are higher than the central cell or multiple central cells (Anonymous, 1996f). Water is contained in a sinkhole, and does not flow anywhere (except may be into the ground). There are 228 Sinkholes in the Watershed. Each sinkhole is assigned a unique number from 1 to 228. Table 6.10 (See Appendix G) shows the size of each sinkhole. The largest sinkhole, No. 26, covers 18,346 hectares, and is located in areas A, B, and C. In Figure 6.15, it shows the flow-direction of the water in some lower elevations toward the higher elevations. This phenomenon is caused because a sinkhole-area in a lower-slope location stores the water from the streams, the runoff, and the groundwater, and this water can flow out when the water pressure of the surface water bodies is lower than the water pressure of the sinkhole-areas during a drought. During a rainy period, the water pressure in a sinkhole-area is lower than the water pressure in a surface water- body, and thus either the water is restricted in the sinkhole-area or the surface water can flow into a sinkhole-area when the water pressure of the surface water body is much higher. Therefore, during a dry period, a sinkhole should be taken into account in calculating the runoff. During the rainy season, the sinkhole-area fills up first. To develop an accurate representation of flow-direction and flow accumulation, a data set free of sinkhole-areas is required. The naturally-occurring sinkhole-areas in a data set with cell sizes of over 10 meters are rare (Mark, 1988) except for the glacier- and heart-type topographies in a watershed. Otherwise, the sinkhole-areas are considered the data errors. The sinkhole-areas in the Te-Chi Reservoir Watershed were caused by the glaciation (Lin, 1974), and thus the sinkhole-areas occurring in the Watershed dataset are COI'I'CCt. 79 6.2.1.4 Stream-Source Areas and Relative Elevation GIS-SVIS Map The accumulated water-runoff in a down-slope cell was calculated by weighing the relative elevation of the cells surrounding it (except for the Sinkholes, see Section 6.2.1.3). The accumulated water is dependent on the number of surrounding cells and their tendency to pass on accumulated water (Anonymous, 1996f). A stream area is an area of concentrated water-runoff accumulation. Mountain ridges do not accumulate the water; i.e., the cells in a high topographical area have zero flow accumulation. The GIS-SVIS map in Figure 6.16 shows the areas of the flow-accumulation in the Watershed; the map also shows the relative elevation of the stream-source areas. The mountain ridges (light-gray-colored areas) account for the largest area of the Watershed. The remaining colored areas in the Watershed are the stream-source areas, i.e. the water sources of the stream. 6.2.1.5 Slope Flow-Length GIS-SVIS Map Figure 6.17 is the GIS-SVIS map of the flow-length distribution in the Watershed. The distance of the flow path of each cell is calculated in order to find the length of the longest flow path within the Watershed. It was used to calculate the time required for water to flow from the most remote point in the area to the outlet of the watershed, i.e. the time of concentration (in minutes). Most of the mountain ridges in the watershed run in the north-south direction and thus the flow paths run in the east-west direction. The flow-lengths are much longer in regions A and C (which are over 2,000m in elevation) than in the valleys in area B. The 80 longest flow-lengths are in the northwest of the Watershed(they are colored red in Figure 6.17). 6.2.2 Land-Use Factor The GIS-SVIS map of slope sun-shade (Figure 6.13) is combined with the GIS map of land-use (Figure 6.18) to produce the GIS-SVIS land-use map shown in Figure 6.19. There are fourteen types of land-use distributed in different topographies in the Watershed. Each land-use type results in a particular surface roughness, thus increasing or decreasing the velocity of the water runoff on the various slopes. In Figure 6.19, the natural forests of coniferous trees, broadleaf trees, and mixed trees account for the largest area in the watershed. In the agricultural regions of area B, the water-runoff-driven erosion has caused nutrient and sediment loading and eutrophic conditions in the main reservoir (Anonymous, 1996d). Water erosion in the watershed is affected by the interaction of torrential rainfall, water runoff, steep slopes, land-use conditions, and underground geological features. The kinetic energy resulting from the velocity of the water runoff becomes the power source driving other erosion factors in causing the considerable soil losses, decreased water storage, degraded water quality, and nutrient-related algae blooms which release dangerous toxins into the reservoir and surrounding streams. It is essential to control the flow-velocity of the water runoff in a certain locations of the Watershed. 6.2.3 Slope-Type and Land-Use Factors The mountain slopes affect the flow direction, accumulation, flow length, and the velocity of the water runoff in the Watershed. The land-use also influences the velocity of the water runoff. For the design of the hillside soil-conservation construction and 81 practices, the velocity of the water runoff is an essential input because it is the major driving force causing the soil erosion. To cultivate an agricultural crop on a hillside, the relationships between the physical properties of the water-runoff and the types of land-use should be considered in order to reduce the water-quality deterioration and the soil erosion. Figure 6.20 depicts the slopes of the terrain which are smaller than 30% and greater than 30%. When a slope is smaller than 30%, the velocity of the water runoff in an agricultural area should be reduced to less than 1.5 m/sec. When a slope is greater than 30% no agriculture should be practiced, one should adhere to the criteria for slope-land cultivation (Anonymous, 1990), and one should practice soil conservation, e. g. bench terracing and hillside ditching. In Figure 6.8 (Section 6.1.3), some bench terraces are shown to be along the main reservoir. Figure 6.21 shows that some agriculture is still practiced on slopes greater than 30%, in particular fruit orchards, vegetable fields, and tea plantation. Figure 6.21 also shows where landslides have occurred in the agriculture areas. Figures 6.22 depicts the location of a typical 21 .8ha vegetable field located in a mountainous area (slopes over 30%) beside a large stream-source. The vegetable field does not conform to the criteria for slope-land cultivation. Figure 6.23 is a detailed map of the vegetable field and its surroundings. It is located in the drainage area of the Chi-Chia-Wan hydrological station. The water from the vegetable field and the surrounding area collects in the southwest comer of the vegetable field. This lowest point receives the water from 33,325 cells (1 14ha). The calculation of the flow rate in the lowest-elevation cell per unit time is described below. 6.2.4 Calculation of Flow Rate 82 “I'm m 0.32.5.5. N o N $8 "A 32m MU $8 v 8.2m Hum §8 "A 86.9 new 3.5» v 329 he moo? on... 8:92.25 .3333. 30.8 2: 5 $8 2 2.2m 25 $8 v 32m 3 932 co gas. m_>m-m_o 2: c3 2:9". 83 88: U 3.8 A 32m. 22.20 :2... 33 :U 3.8 A 3.23 «.8: 833m 28 I 3.8 A 8.29 2.» 5.83.23 888 U 3.8 A 3.29 2.2.86 28 U 3.8 A 829 8:85... so» 838 U 3.8 A 3.29 8.3.83 «88 U 3.8 A 82m. .83. 8:. NM 3.8 A 329 so: 0358:, nosfiofig 302261 20.2. 2: 5 3232.3 can .331 60:33.30 .0 3.8 A 2.29 892 .o 85. m5 2: r3 2:9“. 84 23220:! N o N Z 583 . roman Sumo . ammo M... . ammo . p o D .586 "2.5 i=8 .0 39.52. 392 853.....35 .82.. 226. on; E E3“. 0353; cine-nu 05 B cos-00.. 2:. — Ea“. oBEomo> uoEmem 05 :o 5:80.. och 252225 .3230: 20.0... «5 E «2.4 ES“. a33cmo> uoEmem m5 :0 20:30.. m.:. «Nd 2:9". 85 ..m._,_mo,m.8..8 5.9... .235 9: “38.382 :8 2.: i 5W8 Snag 0.9;3 new? .Emu w_nmummm> umEmem 9.: :0 Samoa.— 9:. .11... .1. . 5. _. ...1... 359.225 .3333. 30.2. 05 m. ES“. o38¢mm> uoEEuxm a... no :0333 2: «Ne 2:9“. 86 The accumulated water in a cell (m3) can be calculated by using the following of relationship: [area (m2) of the cell] x [number of cells located above the cell] x [annual cell runoff (mm)]. The area (m2) of a cell is 34.3396m2 [5.86m x 5.86m]. In Figure 6.16, by using the “Identify” tool, the number of cells located above any cell is found. The annual runoff (m) can be either the average annual runoff (Type A), the medium annual runoff (Type B), or the maximum daily runoff (Type C), for the total watershed (see Table 6.9, Section 5.3.4.3, and Section 6.2) or for a particular drainage area in the Watershed (see Tables 6.6, 6.7, and 6.8). The calculations are shown below and the results of the accumulated water in a cell are shown in Table 6.11. 6.2.5 Flow Rate in the Drainage Area of the Chi-Chia-Wan Hydrological Station 6.2.5.1 Average Annual Runoff (Type A) The average annual cell runoff for the total watershed is 1,96lmm (see Table 6.9) and thus the flow rate is: (34.34m2) x 33,325 cells x 1,961mm = 2,244,264m3 of water per year The average annual cell runoff for the drainage area of the Chi-Chia—Wan hydrological station is 1,520mm (see Table 6.6) and thus the flow rate is: (34.34m2) x 33,325 cells x 1,520mm = 1,739,438m3 of water per year 6.2.5.2 Medium Annual Runoff (Type B) The medium annual cell runoff for the total watershed is 2,156mm (see Table 6.9) and the flow rate is: (34.34m2) x 33,325 cells x 2,156mm = 2,466,809m3 of water per year The medium annual cell runoff for the drainage area of the Chi-Chia-Wan hydrological station is 1,124mm (see Table 6.7) and thus the flow rate is: 87 (34.34m2) x 33,325 cells x 1,124mm = 1,286,269m3 of water per year 6.2.5.3 Maximum Daily Annual Runoff (Type C) The maximum daily annual cell runoff for the total watershed is 200mm (see Table 6.9) and thus the flow rate is: (34.341112) x 33,325 cells x 200mm = 228,965m3 of water per day The value should be used in the design of erosion-prevention systems. The maximum daily annual cell runoff for the drainage area of the Chi-Chia-Wan hydrological station is 141mm (see Table 6.8) and thus the flow rate is: (34.3396m2) x 33,325 cells x 141mm = 161,356m3 of water per day Table 6.11 Predicted Flow Rate (m3/unit time) in the Drainage Area of Chi-Chia-Wan Hydrological Station Cell Runoff Type Accumulated Water (m3lunit time) Total Watershed of Type A 2,224,4264 m3/year Chi-Chia-Wan Drainage Area (Type A) 1,739,438 m3lyear Total Watershed of Type 8 2,422,809 m3/year Chi-Chia-Wan Drainage Area (Type 8) 1,286,269 m3/year Total Watershed of Type C 228,965 malday Chi-Chia-Wan Drainage Area (Type C) 161.356 m3/day 6.2.5.4 Flow Rate Conclusions 1. The annual and maximum water flow rates (m3/unit time) in the Te-Chi Reservoir Watershed are larger than in the Chi-Chia-Wan drainage area. 2. The predicted flow rates in Table 6.11 did not take into account some extraneous influences (infiltration, evapotranspiration, human/animal use) but can be considered as reasonably accurate estimates of the water runoff in a particular region (i.e. a vegetable farm) of the Watershed. 3. The water infiltration and evapotranspiration can decrease the predicted flow rates in Table 6.11 by 10% ~ 30%. 88 4. In considering soil-erosion prevention and stream-source protection for a particular farm in the Watershed, the maximum daily runoff in its drainage area should be used (e.g. 161,356m3/day for the vegetable farm in the Chi-Chia-Wan drainage area). 89 Chapter 7 CONCLUSIONS (1) GIS maps were developed of the vegetable fields and the water bodies in the Te- Chi Reservoir Watershed in Taiwan. (2) The annual and maximum daily water flow rates in the Watershed were calculated, and the water-runoff GIS-SVIS maps were developed. (3) The water flow rates on a vegetable farm in the Chi-Chia-Wan drainage area in the Watershed were calculated assuming that the drainage area is homogenous. 90 Chapter 8 RECOMMENDATIONS FOR FURTHER STUDY Listed below are some recommendations for future study: . The newly-developed design tool can be applied to the production of additional GIS-SVIS maps, e.g. road maps, street maps, in the Te-Chi Reservoir Watershed, or in other watersheds. . The design tool can be applied in Precision Agriculture. . The design tool can be used to simulate and/or monitor the water-quality pollution from non-point sources and point sources in the Te-Chi Reservoir Watershed. 91 Chapter 9 REFERENCES Anonymous. 1999. Deep Computing Institute, IBM Corporation. http://www.research.ibm.com/dci /dx releu se.html. May 24, 1999. Anonymous. 1998a. Compton ’s Interactive Encyclopdia Deluxe (in CDplorm). TLC Properties Inc., Cambridge, MA. Anonymous. 1998b. Using Arc View Image Analysis. ERDAS, Inc., Atlanta, GE. Anonymous. 1998c. Using Arc View Tracking Analyst. TASC, Inc., Reading, MA. Anonymous. 1997a. Understanding GIS-T he ARC/INFO Method. 4th ed. Environmental Systems Research Institute, Inc., Redlands, CA. Anonymous. 1997b. Radio Technical Commission for Aviation Services Incorporated, Aeronautical Spectrum Planning For 19972010, DO-237, SC-l85. Anonymous. 1997c. Using Arc View 30 Analyst. Environmental Systems Research Institute, Inc., Redlands, CA. Anonymous. l996a. Delorme Tripmate User’s Guide. DeLorme Inc., F reeport, ME. Anonymous. 1996b. GTCO AccuTab Surface-Lit LII Plus Digitizer User ’s Manual. Anonymous. 1970 ~ 1996c. Hydrological Year Book of Taiwan, Republic of China. Water Resources Bureau, Ministry of Economic Affairs, Taiwan, Republic of China. Anonymous. l996d. The Planning for the Investigation of the Non-point Sources Pollution and the Control Tactics for T e-C hi Reservoir watershed. Environmental Protection Bureau, Taichung Hsien, Taiwan, Republic of China. Anonymous. l996e. Using Arc View GIS. Environmental Systems Research Institute, Inc., Redlands, CA. Anonymous. l996f. Using the Arc View Spatial Analyst. Environmental Systems Research Institute, Inc., Redlands, CA. Anonymous. 1996g. Microsoft Excel 1997 User ’3 Guide. Microsofi Corporation. Anonymous. 1995a. Global Positioning System Standard Positioning Service Specification. 2nd ed. United States Coast Guard Navigation Center. Anonymous. 1995b. Radio Technical Commission for Aviation Services Incorporated. Minimum Aviation System Performance Standards DGNSS Instrument Approach System: Special Category 1 (SCAT I). DO-217. SC-159. April 19, 1995. 92 Anonymous. 1995c. The Third Period of Integral Policy Making, Water Quality Monitoring, and Managing Modeling for Te-Chi Reservoir Watershed II. Final Report. Te-Chi Reservoir Committee, Taiwan, Republic of China. p100. Anonymous. 1994a. Content Standards for Digital Geospatial Metadata. Federal Geographic Data Committee (FGDC), U. S. Geological Survey, Washington, DC. Anonymous. 1994b. Radio Technical Commission for Maritime Services. RTCM Recommended Standards for Differential NAVSTAR GPS Service - Version 2.1. RTCM Special Committee No. 104. January 3, 1994. Anonymous. 1994c. The Setup of Information System and Landuse/cover Change Analysis of T e-Chi Reservoir Catchment. Final Report, ERL Report 06-3-84-0020. Te-Chi Reservoir Committee, Taiwan, Republic of China. pl 85. Anonymous. 1993. Global Positioning System monographs. The Institute of Navigation. Washington, DC. Anonymous. 1991. ICD-GPS-ZOO: GPS Interface Control Document. GPS Joint Program Office. ARIN C Research. Anonymous. 1990. Soil Conservation Handbook. Soil and Water Conservation Service, Taiwan, Republic of China. Anonymous H. 1984. Standards and Specifications for Geodetic Control Networks. Federal Geodetic Control Committee. National Geodetic Information Branch. Maryland, USA. Anonymous. 1981. Standard Specifications and Recommended Practices for Horizontal and Vertical Control Surveys. 3rd Edition: Special Publication 1. National Mapping Council of Australia. Canberra: National Mapping Council of Australia. Anonymous. 1978. Specifications and Recommendations for Control Surveys and Survey Markers. Surveys and Mapping Branch, Ottawa, Canada: nergy, Mines and Resources Canada. Athlstrong, A.P., P.A. Collier, and R]. Leahy. 1992. Leveling Using the Global Positioning System. Urban Water Research Association of Australia, Research Report No. 47, Department of Surveying and Land Information, The University of Melbourne. Amell, NW. 1995. Grid Mapping of River Discharge. Journal of Hydrology. 167 (1995) 39-56. ASl‘ar, G. (ed). 1989. Theory and Applications of Optical Remote Sensing. John Wiley and Sons, Inc. New York, NY. 93 Avery, T.E., and G.L. Berlin. 1992. Fundamentals of Remote Sensing and Airphoto Interpretation. 5th edition. Macmillan Publishing Company, New York, NY. Axelrad, P. & Brown, R. G. 1996. GPS: Theory and Applications Volume 1: GPS Navigation Algorithms. Astronautics and Aeronautics. 163: 409-434. Blanchard, W. November 1996. The Global Nav. The Situation Up To Now. NA V96: Global Navigation Systems Where are we now. International Conference of the Royal Institute of Navigation, London. pp. 1 .1-1 . 1 8. Brodin, G. 1996. GNSS Code and Carrier Tracking in the Presence of Multipath. Proceedings of the 9th International Technical Meeting of the Satellite Division of the Institute of Navigation GPS-96, Kansas City, September 17-20 1996. p1389- 1398. Capaccio, S., D.M.A.Walsh, D.Lowe, P.Daly. GLONASS as an Accompaniment to the GPS Satellite System. Surveying in the Offshore Oil & Gas Industry Technology Seminar, The Hydrographic Society in Scotland, (Aberdeen, UK), April 3, 1997. Capaccio, S., and D.M.A.Walsh. 1996. Results of GPS and GLONASS real-time differential positioning over a 3 70km baseline. Civil Aviation Authority Institute of Satellite Navigation, (Leeds, UK), Jan., 1996. Calbaccio, S., D. Lowe, D.M.A.Walsh, P.Daly, G.Richards, A.Wolfe, and H.Mistry. February 1997. GPS/GLONASS LAAS and RAAS En-route Positioning and CA TI Precision Approach with FMS Integration: Demonstration and Evaluation (Project Boscombe). Institute of Satellite Navigation. Capaccio, S., D.M.A.Walsh, D.Lowe, P.Daly. GLONASS as an Accompaniment to the GPS Satellite System. Surveying in the Offshore Oil & Gas Industry Technology Seminar, The Hydrographic Society in Scotland, (Aberdeen, UK), April 3, 1997. Capaccio, s., and D.M.A.Walsh. 1996. Results ofGPS and GLONASS real-time differential positioning over a 3 70km baseline. Civil Aviation Authority Institute of Satellite Navigation, (Leeds, UK), Jan., 1996. Capaccio, S., D. Lowe, D.M.A.Walsh, P.Daly, G.Richards, A.Wolfe, and H.Mistry. February 1997. GPS/GLONASS LAAS and RAAS En-route Positioning and CA T 1 Precision Approach with FMS Integration: Demonstration and Evaluation (Project Boscombe). Institute of Satellite Navigation. Capaccio, s., D.M.A. Walsh, D. Lowe, and P. Daly. 1997. Real-time differential GPS/GLONASS trials in Europe using all-in-view 20-channel receivers. Journal of Navigation, p193-208. 94 Chrisman, NR. 1997. Exploring Geographic Information Systems. John Wiley & Sons, Inc., New York, NY. Clarke, D. 1963. Plane and Geodetic Surveying; Volume II. Constable and Company Ltd. London, UK. Congalton, R.G. and K. Green. 1995. Wetland and Environmental Application of GIS: The ABCs of GIS: An Introduction to Geographic Information System. CRC Press, Inc., Boca Raton, FL. Denzer, R. 1993. Graphics for Environmental Decision-Making. IEEE Computer Graphics & Applications, 13 (3): 58-64. Dueker, K.J. and D. Kjeme. 1989. Multipurpose Cadastre: Terms and Definitions. ASPRS and ACSM, Falls Church, VA. Doebelin, E. 1990. Measurement Systems Application and Design. 4th ed. McGraw-Hill Company. New York, New York. IDomik, G.O.1994. Visualization Education. Computers & Graphics, 18 (3): p277-280. Greenlee, D. D. 1987. Roster and Vector Processing for Scanned Linework, Photogrammetric Engineering and Remote Sensing. October 1987. 53 (10): 1383- 1387. Gllptill, S.C. (ed). 1988. A Process for Evaluating Geographic Information Systems. US Geological Survey Open File Report 88-105. Reston, VA. I‘I'E3al‘nshaw, H. M. and Unwin, D. J. (eds.). 1994. Visualization and GIS. London: Wiley, Inc., London, UK. Ji, W. and LC. Mitchell. 1995. Analytical Model-Based Decision Support GIS for Wetland Resource Management. CRC Press, Inc., Boca Raton, FL. Evans, T. B. May 1986. Microwave Landing System. IEEE Aerospace and Electronic Systems Magazine. pp. 6-9. Federal Aviation Authority. October 21, 1996. FAA National Airspace System (NAS) Architecture, version 2.0, ASDl , Washington. Filatchenkov, S. 1996. NA V96: Global Navigation Systems: Where are we now. The Current Status and Future Plans for GLONASS. International Conference of the Royal Institute of Navigation, London. November 5-6. 1996. pp.4.1-4.9. I‘12).lliday,D., R. Resnick, and J. Walker. 1993. Fundamentals ofPhysics. 4‘'1 Edition. John Wiley & Son, Inc. New York, NY. 95 Hartman, R.G., M.A. Brenner, and N.K. Kant. 1991. GPS/GLONASS Flight Test, Lab Test and Coverage Analysis Tests. Proceedings of the Institute of Navigation GPS-91. Albuquerque, NEW MEXICO. September 11-13, 1991. pp.333-344. Harvey, BR. 1986. Transformation of 3D Coordinates. The Australian Surveyor 33(2): 105 - 125. He, Chun-Sun. 1982. Tectonic Evolution of Taiwan Explanatory T ext of the Tectonic Map of Taiwan. Central Geological Investigation Research Institute, Ministry of Economic Affairs, Taiwan, Republic of China. Hoffmann—Wellenhof, B., H. Lichtenegger, and J. Collins. 1994. GPS: Theory and Practice. 3rd ed. New York: Springer-Verlag. Hudson, N. 1981. Soil Conservation. London: B.T. Bastford Limited. Kaplan, E.D. (ed). 1996. Understanding GPS: Principles and Applications. Artech House Publishers. Boston, MA. Kayton, M. 1989. Introduction to Aircraft Navigation, Navigation: Land, Sea, Air & Space. IEEE Press. pp.229-244. Kelly, R. J. and E.F.C. LaBerge. 1990. MLS: A Total System Approach. IEEE Aerospace and Electronic Systems Magazine. May, 1990. 5: 27-39. Keller, P.R., Keller, M.M. 1992. Visual Cues: Practical Data Visualization. IEEE Computer Society Press, Inc., Los Alamitos, CA. Kirkby, M.J. and R.P.C. Morgan. 1980. Soil Erosion. New York, NY: John Wilry & Sons, Inc. KOVar, K. and HP Nachtnebel. 1996. Application of Geographic Information Systems in Hydrology and Water Resources Management. Oxfordshire, UK: International Association of Hydrological Sciences. Institute of Hydrology Press. I(I‘Otrler, D. and J. Landis. 1992. GPS Interface Control Document. ICD-GPS-200, IRN2OOB-PROOJ, Rev.B-PR. US. Air Force. July], 1992. I(“Tallow V. Global Navigation Satellite Systems Panel (GNSSP). Montreal, CANADA: GLONASS - Interface Control Document. November. 14-24. Langley, R. B. 1997. GLONASS: Review and Update. GPS World. July, 1997. pp.46-51. Leick, A. 1995. GPS Satellite Surveying. 2nd ed. John Wiley & Sons, Inc. New York, New York. 96 Lee, C. and S. Marsh. 1995. The Use of Archival Landsat MMS and Ancillary Data in a GIS Environment to Map Historical Change in an Urban Riparian Habitat. CRC Press, Inc., Boca Raton, FL. Lee, K.H. and RS. Lunetta. 1995. Wetland and Environmental Application of GIS: Wetlands Detection Methods. CRC Press Inc., Boca Raton, FL. Lin, Chao-Chi. 1974. Taiwanese Topography. Central Geological Investigation Research Institute, Ministry of Economic Affairs, Taiwan, Republic of China. Lyon, J .G. and J. McCarthy (ed). 1995. Wetland and Environmental Application of GIS: Introduction to Wetland and Environmental Application of GIS. CRC Press Inc., Boca Raton. Lyon, J .G. and J. McCarthy (ed). 1995. Wetland and Environmental Application of GIS: Water Resource Engineering Application of Geographic Information System. CRC Press Inc., Boca Raton, FL. Lyon, J .G. and J. McCarthy (ed). 1995. Wetland and Environmental Applications of GIS: Introduction to Other Issues and Applications. CRC Press, Inc., Boca Raton, FL. Mark, D. M. 1988. Network Models in Geomorphology, Modeling in Geomorphologic Systems. John Wiley and Sons, Inc. New York, NY. Merva, GE. 1995. Physical Principles of the Plant Biosystem. St. Joseph, MI: American Society of Agricultural Engineers. Misra, 1)., R]. Abbot, EM. Gaposchkin. 1996. Integrated Use ofGPS and GLONASS: Transformation Between W G584 and PZ-90. Proceedings of the 9th International Technical Meeting of the Satellite Division of the Institute of Navigation GPS-96. Kansas City. September 17-20. 1996. pp.307-3l4. Misra, P. et a1. 1993. SGS85 - WGS84 Transformation: Interim Results. MIT ATC Project Memorandum 42PM-SATNAV-0004. Feb 1993. IVlllehrcke, PC. 1986. Map Use: Reading, Analysis, Interpretation. JP Publication, Madison, WI. Ohanian, H.C. 1985. Physics. 15‘ ed. WW. Norton & Company, Inc. New York, NY. Paine, DE. 1981. Aerial Photography and Image Interpretation for Resource Management. John Wiley and Sons, Inc. New York, NY. 1Darker, H.D. (ed). 1989. The GIS Sourcebook. GIS World, Inc., Fort Collins. CO. 97 Parkinson, B. W. 1996. GPS: Theory and Applications Volume I: Introduction and Heritage of NA VST AR, the Global Positioning System. Astronautics and Aeronautics. 163: 3-28. Parkinson, B. W. 1996. GPS: Theory and Applications Volume I--- GPS Error Analysis. Journal of Astronautics and Aeronautics, 163: 469-483. Parkinson, B.W. and J .J . Spilker. (ed). 1996. Global Positioning System: Theory and Practice. Volumes 1. American Institute of Aeronautics and Astronautics, Inc., Washington, DC. Parkinson, B.W. and J .J . Spilker. (ed). 1996. Global Positioning System: Theory and Practice. Volumes II. American Institute of Aeronautics and Astronautics, Inc., Washington, DC. Parkinson, B. W. and J .8. James (ed). 1996. Global Positioning System: Theory and Practice. Volumes I and II. American Institute of Aeronautics and Astronautics, Inc. Washington, DC. Pierce, J .A. 1946. An Introduction to LORAN. Proceedings of the IRE and Waves and Electronics. 34: 216-234. Raby, P. 1994. Integrating Signal Information fiom GPS and GLONASS. Department of Electronic and Electrical Engineering, The University of Leeds. August 1994. Raper, J. (ed). 1989. Three Dimensional Applications in Geographic Information Systems. Taylor & Francis, Inc., Philadelphia, PA. Robinson, AH. and BB. Petchenik. 1976. The Nature of Maps. University of Chicago Press, Chicago, IL. Rossbach, U., H. Habrich, and N. Zarraoa. Transformation Parameters Between PZ-90 and WGS-84. Proceedings of the 9th International Technical Meeting of the Satellite Division of the Institute of Navigation GPS-96. Kansas City. September 17-20, 1996, pp.279-285. Rowson, S, et al. 1994. Performance of Cat 111 B Automatic landings Using C/A code Tracking Differential GPS. Journal of the Institute of Navigation. 1994, pp127- 144. Serway, RA. 1992. Physics for Scientists & Engineers with Modern Physics. 3rd ed. Saunders College Publishing. Orlando, FL. Schwab, G.O, D.D. F angmeier, W.J. Elliot, and R.K. Frevert. 1993. Soil and Water Conservation Engineering. 4"1 Edition. New York, NY: John Wiley & Sons, Inc. Stevens, SS. 1946. On the Theory of Scales of Measurement. Science 103: 677-680. 98 Synder, J .P. 1987. Map Projections-A Working Manual. USGS Professional Paper 1395. US. Government Printing Office. Washington, DC. Thompson, M.M. 1988. Maps for America. US. Government Printing Office. Washington, DC. Van Graas, F, et a1. 1995. FAA/Ohio University/UPS Autoland Flight Test Results ", Navigating the 90 's: Technology, Applications and Policy. Proceedings of the National Technical Meeting, Institute of Navigation. Anaheim. January 18-20, 1995. pp.145-l66. Ulliman, J. J. 1995. Aerial Photo Interpretation of Renewable Natural Resources: Course Notes and Workbook. Department of Forest Resources, University of Idaho. 83844. Moscow, Idaho. Walsh, D.M.A. 1995. A feasability study of Category III Precision Approach and Landing using GPS, GLONASS and Pseudolites. CAA Institute of Satellite Navigation. Leeds, UK. May, 1995. Wells, D. (ed). 1989. Guide to GPS positioning. Fredericton. Canadian GPS Associates. NB, Canada. Wolf, PR. 1983. Elements of Photogrammetiy with Air Photo Interpretation and Remote Sensing. 2nd ed. McGraw-Hill Book Company. New York, NY. Worboys, M.F. 1995. GIS: A Computing Perspective. Taylor & Francis, Inc., Bristol, PA. Yu, Chia-Yii. 1997. Remote Sensing of the Environment: Course Note. Department of Agricultural Engineering, Michigan State University, East Lansing, MI. 99 APPENDICES 100 8.8 88: 888 8.8:. 8.82 8.82 8.8.8 8.5: 82 e885 unnuuuuuu unnuununu 88388 uuuuuunnu 88328 ununuuuuu 88382 88 uunuuuuuu unnunnunu 88338 unuuuuuuu 8838: unnunuunu 988.88 :8 unuuuuuuu unuunuuun 8888 annual” 88388 nuuuuuuuu 88388 88 nunuununu nunununuu 88388 uuuuuuuuu 82388 ununuunun 82388 88 nunnnunuu uuunnnunu 88388 ununuuuun 88388 uuuuununu 883.: m 88 unuuuunuu nuuuunuuu 82.28 ununnnuun 82388 unnuuuuuu 88388 88 88382 88382 883 88 88388 88388 uuunuunun 8838: 88 88388 88388 88388 88388 82388 unnuuuunn 88382 88 88388 88_ .88 88382. 88388 88388 unuuunuuu 888.88 88 88_ .88 82 .88 88388 888.88 88388 unnuununu 88388 88 8838: 88382 828.88 888.88 88388 unuunnnuu 882.82 88 88328 883 38 888.88 888.32 888.28 nunuuuunn 88388 88 88388 823:: 88388 883: 8 28: .88 uuuuuuuuu 88388 88 Hull” 888.88 88388 88382 _ 82 _ 8:. unuunnnuu 888.88 82 nuunnuunu 88382 888.83. 888.88 88388 unuuuunnu 88388 88 88388 88388 88388 883882 88388 ununuuuuu 88388 88 88388 88388 88338 88382.: 88338 uuunuuunn 82.388 88 882 .82 88382 88388 883 _ 88 88 _ .88 uuuuuuunu 88388 88 8838: 888.8: 88382. 82388 88388 88382 88382 88 88328 88382 88_ .82 888.28 88388 38: .88 8838: 88 882 .88 82388 88388 88388. 828.88 ununununu unuunuunu 88 8838: 888.22 88388 88388 888.8: 88_ .82 88382 88 88_ .88 88388 88388 823282 888.88 88388 838.88 88 888.82 8838... 88388 88388 8838: 88382 88388 88 88388 883 :8 88328 8838: 88_ .88 88388 88388 88 88388 8238: 88388 288.88 888.88 88388 888.82 38 888.82 88382 88383. 88388 88388 88338 82388 88 3.5.3 5:. 8.5.8.3 8.85:. 53.3335 82.83% 83.30.38. 5:58: 3:82 .30> 828:8 455393»: 3.35:2... 3.5.8.... 828683 hotomom 308,—. 8: E 22285 Rama—8?»: 52% 5 vote: 80.335 .5>o 83 b023— 85: 2:. Nm 2an < XHQZm—n—n?‘ 101 :u my; -- H HOW—u _ APPENDIX B Table 5.4 The Maximum Daily Runofi‘ Data Over 27-Year Period in Seven Hydrological Stations in the Te-Chi Reservoir Watershed Chl—Chm- Wan -§ § 3 5 : t: E: t a a I 102 APPENDIX C Table 6.1 The "aat.dbf" file of the Water Bodies Map of the Te-Chi Reservoir Watershed FNODE_ TNODE_ LPOLY_ RP()LY_ LENGTH STREAM_ STREAM_ID 0 0 0 0 1233323000 1 19 0 0 0 0 2790.733000 2 27 0 0 0 0 1273323000 3 29 0 0 0 0 346274200 4 45 0 0 0 0 302 342700 5 52 0 0 0 0 302.342700 6 53 0 0 0 0 936.563500 7 54 0 0 0 0 1911.960000 3 61 0 0 0 0 1193.323000 9 63 0 0 0 0 2037646000 10 69 0 0 0 0 1071 960000 1 1 71 0 0 0 0 459 411300 12 79 0 0 0 0 915 979300 13 34 0 0 0 0 1325 636000 14 36 0 0 0 0 1403523000 15 90 0 0 0 0 1325636000 16 91 0 0 0 0 216.563500 17 93 0 0 0 0 1629 117000 13 94 0 0 0 0 2327 940000 19 104 0 0 0 0 320.000000 20 105 0 0 0 0 969 705600 21 107 0 0 0 0 2305 097000 22 103 0 0 0 0 1971.372000 23 122 0 0 0 0 2161.666000 24 124 0 0 0 0 2453234000 25 126 0 0 0 0 433.137100 26 127 0 0 0 0 329705600 27 131 0 0 0 0 2931 077000 23 132 0 0 0 0 200000000 29 133 0 0 0 0 2950 733000 30 134 0 0 0 0 1019411000 31 133 0 0 0 0 193137100 32 142 0 0 0 0 1235.930000 33 143 0 0 0 0 2074303000 34 144 0 0 0 0 629.116900 35 147 0 0 0 0 2397646000 36 153 0 0 0 0 1525 636000 37 166 0 0 0 0 193137100 33 171 0 0 0 0 319411300 39 174 0 0 0 0 1302254000 40 175 0 0 0 0 359411300 41 134 0 0 0 0 113137100 42 133 0 0 0 0 1066274000 43 139 0 0 0 0 942254000 44 215 0 0 0 0 1331960000 45 1 0 0 0 0 2710.733000 46 2 0 0 0 0 1035.636000 47 3 0 0 0 0 1132543000 43 4 0 0 0 0 2221077000 49 5 0 0 0 0 1775 391000 50 6 0 0 0 0 1673 323000 51 7 0 0 0 0 1391.372000 52 3 0 0 0 0 3709 606000 53 9 0 0 0 0 2164.509000 54 10 0 0 0 0 1235930000 55 1 1 0 0 0 0 2135 097000 56 12 0 0 0 0 346274200 57 13 0 0 0 0 302342700 53 14 103 CC AA 9" CCOCCC COOCOOOOOOOOCOCCCOOOCOCOOOCOCOOOOC C COCCCCCCCOCOCOOOOOOOOCCC CCOCCOCOOOCCOCOOCCOOCCCCC‘OOOOOCCOCO C COOOOOOOCC F5 \— 104 4417 057000 1382.254000 1358 823000 191 1.960000 1 102 254000 7396411300 1425 097000 1365 686000 80000000 3370 195000 869 1 16900 699 41 1300 273 137100 602 842700 2041 666000 1085685000 2787 940000 1695 391000 296 568500 1358823000 1874803000 402842700 998822500 442842700 5098 722000 2154 803000 1391 960000 2775.392000 6756979800 2025 097000 426 274200 5655 878000 2907 940000 369 705600 675 979800 1731 371000 249 705600 1422 254000 706 274200 1398823000 715 979800 1012548000 1938234000 402.842700 136.568500 273137100 13491 17000 2374215000 513137100 489 705600 2147 940000 2867 940000 193 137100 1985097000 612 548300 2517.646000 1212 548000 2547940000 2368 529000 59 60 61 62 63 64 65 66 67 68 69 7O 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 9O 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 1 10 1 1 1 112 113 l 14 115 1 16 1 l7 CCCCCCOOOC OOOOOOCC CCCOOCCO A A v ‘v OCCC CCC‘OOOCCCOOOO OCCCC CCOOOCCCOCOCOOCCCCCCCCCOCCCCCCCCCCC A \n' OCCCC COOCC‘OCC OOOOOCOOOOOCOCC‘COOOCOCO A \v COOOOCCCO 105 13421254000 2550.783000 1422254000 369 705600 466.274200 5722 152000 892548300 336 568500 2505097000 659 41 1300 1318 823000 369 705600 4057 057000 2781 .077000 2328529000 57941 1300 2081 666000 369 705600 313 137100 3993 626000 482842700 4109606000 826274200 902.254000 2431 960000 369 705600 131 1.960000 5682 151000 2581 077000 1492549000 2783 920000 3167352000 336568500 1575 391000 3477 646000 1099 4 1 1000 1578 234000 1514 803000 4476468000 702 254000 506274200 482842700 1391.960000 3773 038000 682 842700 1075 .980000 2027.940000 1591 960000 692 548300 56 568540 369 705600 932 548300 562842.700 129941 1000 96568540 176 568500 1231960000 2347 940000 1365 686000 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 I45 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 243 244 245 246 247 248 249 N N N N 'Ju 'Ju 'JI 'Jv b) N -- C S'l‘R13AM_1D = Stream-ll) 106 0 0 0 0 626 274200 0 0 0 0 4729016000 0 0 0 0 955.979800 0 0 0 0 3160 489000 0 0 0 0 2378.234000 0 0 0 0 1069117000 0 0 0 0 120 000000 0 0 0 0 2281 666000 0 0 0 0 273 137100 0 0 0 0 1485 686000 0 0 0 0 689 705600 0 0 0 0 862 254000 0 0 0 0 892 548300 0 0 0 0 3393 626000 0 0 0 0 1601 666000 0 0 0 0 160 000000 0 0 0 0 3207 352000 0 0 0 0 972 548300 0 0 0 0 909.1 16900 0 0 0 0 553137100 0 O 0 0 732548300 0 0 0 0 1598.823000 0 0 0 0 1069.1 17000 0 0 0 0 256 568500 0 0 0 0 1278823000 0 0 0 0 682842700 0 0 0 0 932 548300 0 0 0 0 772 548300 0 0 0 0 1 135 391000 0 0 0 0 2844 509000 0 0 0 0 1405 686000 0 0 0 0 l 191.960000 0 0 0 0 1975392000 0 0 0 0 2481 666000 0 0 0 0 120 000000 0 0 0 0 1035 980000 0 0 0 0 369 705600 0 0 0 0 10195.770t'100 0 0 0 (1 5086 173000 FNODE_ = Internal node number for the begmmng of an arc (from-node) '1'N01)1{_ = lntemal node number for the end 01‘ an arc (to-node) l.P()1.Y_ = Internal number for the left polygon RPOLY_ = internal number for the r1ght polygon LENGTH = Length ol‘eaeh are measured in coverage units STREAM_ = Internal are number 259 260 261 262 263 264 265 266 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 APPENDIX D Table 6.2 The "pat.dbt" File of the Vegetable Farms Map of the Te-Chi Reservoir Watershed AREA PER11~1ETER V1ZGETABL_ VEGETABL_1 LANDUSE 8450228000000 99084590000 1 0 218419600000 4272 860000 2 148 V 140387100000 4405 426000 3 149 V 45748260000 1860 044000 4 153 V 21941 220000 1300698000 5 156 V 3267 586000 313 867500 6 0 1434 078000 156 931000 7 164 V 191938 600000 6947 825000 8 184 V 2992 098000 265 325300 9 188 V 15202 320000 720 244000 10 194 V 6693 203000 416 664100 1 1 195 V 10947 520000 517 653000 12 214 V 5606 078000 529 838100 13 O 1883 910000 173 225900 14 0 125944 500000 1708 999000 15 236 V 75221.720000 1552847000 16 241 V 10552 500000 488 069000 17 262 V 9652672000 536 769800 18 263 V 3917730000 324 581100 19 264 V 4464 117000 253 600500 20 275 V 3210 680000 282 950200 21 302 V 14519 820000 838 644700 22 304 V 5626 805000 428 397000 23 312 V 15168 610000 799 490100 24 314 V 15876 130000 669 432900 25 317 V 8857 430000 479 950100 26 326 V 31064 940000 1665 745000 27 334 V 39703670000 2107 025000 28 345 V 13986 750000 536 834000 29 353 V 5798 848000 345 574100 30 354 V 7027 563000 443014300 31 411 V 24525 730000 724 898700 32 438 V 16005360000 831 753500 33 443 V 43331 460000 1044 744000 34 449 V 43056 790000 1536946000 35 500 V 91423470000 3302 516000 36 501 V 15926810000 714 944400 37 526 V 2382 055000 241 309000 38 539 V 7912 422000 535 533000 39 556 V 51675950000 1485 429000 40 577 V 9174 590000 550 438700 41 599 V 13738550000 616 008600 42 619 V 15478 710000 664878200 43 644 V 14725780000 1253 900000 44 676 V 10727 830000 969 223300 45 680 V 4952457000 481 682300 46 810 V 6285 234000 541.026900 47 825 V 272289 900000 3633 878000 48 920 V 143989 600000 2699 589000 49 927 V 4582 969000 292 213900 50 929 V 1 1064 800000 470 469200 51 979 V 46968 000000 1326 083000 52 983 V 10513 080000 494 719100 53 1016 V 5621 020000 310 408200 54 1017 V 99315 940000 2203641000 55 1034 V 13869 540000 543 890300 56 1037 V 107 APPENDIX D Table 6.2 The "pat.db1“ File ofthe Vegetable Farms Map ofthe Te-Chi Reservoir Watershed 90333860000 1536 558000 57 1042 V 13654.100000 525 920100 58 1047 V 6382.988000 355964300 59 1050 V 39876820000 2020 862000 60 1062 V 2448105000 205.553700 61 0 8519871000 369342100 62 1065 V 46884 680000 1 152.359000 63 1075 V 21150 510000 705 015200 64 1076 V 4625 586000 281 015700 65 1079 V 36106410000 1683 497000 66 1099 V 271 035200 65770700 67 0 6001 016000 314 600900 68 l 153 V 11961 050000 483 464200 69 1926 V 5524.781000 326 440200 70 1934 V 9070 816000 675 668900 71 1950 V 96563470000 1738 859000 72 1959 V 48086 140000 1340978000 73 1986 V 32276.780000 931 959200 74 2000 V 3566344000 302 925300 75 2045 V 59078.020000 1494687000 76 2048 V 1088.961000 129836200 77 2067 V 5513660000 323885200 78 2069 V 141578 700000 3108 471000 79 2078 V 12403340000 618 908600 80 2097 V 5048 609000 321 336400 81 2107 V 2561 602000 297 405900 82 0 4525 387000 341 265300 83 2113 V 370 035200 80 694260 84 2143 V 12110700000 481 349300 85 2153 V 3607 371000 305 305000 86 2194 V 9901 199000 514 192700 87 2199 V 5118 652000 297 398800 88 2205 V 32968 050000 1 127 395000 89 2609 V 15627980000 690 046700 90 2617 V 230521 900000 3367 702000 91 2620 V 2339.871000 192 154100 92 0 31598.280000 1345450000 93 2633 V 2103922000 179 481600 94 0 144672100000 4061283000 95 2659 V 48339030000 1 138 529000 96 2665 V 81900 440000 2322.125000 97 2683 V 67621 200000 2067 832000 98 2687 V 9730 125000 636 985200 99 2704 V AREA = Area ot'eaeh polygon measured in coverage units PERIMETER = VEGETABL_= V1~j(j15'1‘/\13L_1 = LANDUSE = Length ol‘cach polygon boundary measured in coverage units Internal polygon number Vegetable-1D Mark 01‘ a type ot‘land use (vegetable farm) 108 APPENDIX F Table 6.4 Statistical Programs for Modification of the GIS Map (Measured Points: 2? / ZQ) ID GPS Receiver A 2P / ZQ Latitude (XA) Longitude (YA) Sexagesumal Notation Decamal Notation Sexagesimal Notation Deamal Notation Item Degree Minute Second Degree Degree Minute Second Degree 1 120 59 58 120 9994444 24 15 0 24 25 2 121 O 1 121 0002778 24 14 58 24 24944444 3 121 O l 121 0002778 24 14 59 2424972222 4 120 O 2 1200005556 24 14 57 24 24916667 5 121 0 0 121 24 I4 58 24 24944444 6 121 0 I 121 0002778 24 I4 58 2424944444 7 121 0 0 121 24 15 0 24 25 8 121 0 0 121 24 15 0 24 25 9 120 59 59 120 9997222 24 15 1 24 25027778 10 121 0 1 121 0002778 24 15 2 24 25055556 11 120 59 59 1209997222 24 15 0 24.25 12 120 59 59 120 9997222 24 15 2 2425055556 13 121 0 0 121 24 15 1 2425027778 14 I21 0 1 121 0002778 24 15 1 2425027778 15 120 59 59 120 9997222 24 14 59 2424972222 16 120 59 59 120 9997222 24 14 59 24 24972222 17 120 59 59 120 9997222 24 15 0 24 25 18 120 59 59 120 9997222 24 14 59 2424972222 19 120 59 57 120 9991667 24 15 0 24.25 20 120 59 58 120 9994444 24 15 I 2425027778 21 121 0 2 121 0005556 24 15 1 2425027778 22 121 0 3 121.0008333 24 15 1 24 25027778 23 121 0 1 121 0002778 24 15 2 2425055556 24 121 0 1 121 0002778 24 I4 59 2424972222 25 121 0 0 121 24 15 0 24 25 26 120 S9 58 120 9994444 24 I4 59 2424972222 27 I21 0 0 I21 24 I5 0 24 25 28 120 59 58 120 9994444 24 15 O 24 25 29 121 0 3 1210008333 24 15 1 24 25027778 30 121 0 1 121 0002778 24 15 1 2425027778 31 121 0 1 121 0002778 24 15 2 24 25055556 32 120 59 59 120 9997222 24 15 1 24 25027778 33 120 59 58 120 9994444 24 15 2 24 25055556 34 121 0 0 121 24 14 59 2424972222 35 121 0 0 121 24 14 59 24 24972222 36 121 0 0 121 24 15 O 24 25 37 121 0 1 121 0002778 24 15 1 24 25027778 38 121 0 2 121 0005556 24 15 I 24 25027778 39 121 0 1 121 0002778 24 15 2 2425055556 40 121 0 2 121 0005556 24 14 59 2424972222 41 121 0 3 1210008333 24 15 0 24.25 42 120 59 59 120 9997222 24 14 58 2424944444 43 120 59 58 120 9994444 24 14 53 24 24944444 44 121 0 l 121 0002778 24 14 59 24 24972222 45 121 0 1 121 0002778 24 14 59 24 24972222 46 120 59 59 120 9997222 24 15 0 24 25 47 120 59 58 120 9994444 24 IS 1 2425027778 48 121 0 2 121 0005556 24 I4 59 24 24972222 49 121 0 1 121 0002778 24 15 0 24 25 50 121 0 1 121 0002778 24 15 l 24 25027778 109 GPS Receiver 8 Latitude (X8) Longitude (YB) Sexagesimal Notation Decimal Notation Sexagesmal Notation Decimal Notation Degree Minute Second Degree Degree Minute Second Degree 121 20 1 121.3336111 24 22 30 24375 121 20 1 121 3336111 24 22 31 24 37527778 121 20 2 1213338889 24 22 31 2437527778 121 20 1 1213336111 24 22 31 2437527778 121 20 3 1213341667 24 22 30 24375 121 20 1 1213336111 24 22 30 24 375 121 20 2 1213338889 24 22 30 24.375 I21 20 1 1213336111 24 22 3O 24 375 121 19 59 1213330556 24 22 29 2437472222 121 19 58 121 3327778 24 22 29 24 37472222 121 20 2 1213338889 24 22 29 2437472222 121 19 58 1213327778 24 22 30 24375 121 20 1 1213336111 24 22 30 24.375 121 20 1 1213336111 24 22 30 24 375 121 20 1 121 3336111 24 22 29 2437472222 121 I9 58 1213327778 24 22 29 2437472222 121 19 59 121.3330556 24 22 28 24.37444444 121 I9 58 1213327778 24 22 29 2437472222 121 19 59 1213330556 24 22 30 24.375 121 19 57 1213325 24 22 31 2437527778 121 19 57 121 3325 24 22 31 2437527778 121 20 0 1213333333 24 22 29 2437472222 121 20 0 121 3333333 24 22 29 2437472222 121 19 58 1213327778 24 22 30 24375 121 19 57 121 3325 24 22 31 24.37527778 121 20 1 1213336111 24 22 30 24.375 121 20 1 1213336111 24 22 30 24 375 121 19 2 1213172222 24 22 29 2437472222 121 20 3 121 3341667 24 22 29 2437472222 121 19 58 1213327778 24 22 30 24.375 121 20 59 1213497222 24 22 29 2437472222 121 20 2 121 3338889 24 22 29 24 37472222 121 I9 58 1213327778 24 22 30 24 375 121 19 59 1213330556 24 22 30 24.375 121 20 1 121.3336111 24 22 30 24.375 121 19 59 1213330556 24 22 29 2437472222 121 20 0 1213333333 24 22 29 24 37472222 121 20 0 1213333333 24 22 30 24375 121 20 0 121 3333333 24 22 30 24.375 121 20 0 1213333333 24 22 30 24375 121 20 1 121 3336111 24 22 31 2437527778 121 19 59 121 3330556 24 22 32 2437555556 121 I9 58 1213327778 24 22 33 24 37583333 121 19 58 121 3327778 24 22 32 24 37555556 121 20 1 121 3336111 24 22 33 2437583333 121 20 2 1213338889 24 22 32 2437555556 121 19 59 1213330556 24 22 30 24,375 121 I9 59 1213330556 24 22 29 24.37472222 I21 20 1 121 3336111 24 22 28 24 37444444 121 20 2 1213338889 24 22 27 24.37416667 110 Item OMQOMbWN X(XA - XB) ~0334166667 0333333333 -0.333611111 -l.333055556 -0 334166667 -0 333333333 -0333888889 -0333611111 -O.333333333 -0 3325 -0334166667 -0 333055556 -0.333611111 -0 333333333 -0.333888889 0333055556 -O.333333333 -0 333055556 0333888889 .0 333055556 -0.331944444 -0 3325 -0 333055556 -0.3325 -0 3325 0334166667 -0 333611111 -O.317777778 -0333333333 03325 -0 349444444 -0 334166667 -0 333333333 -0.333055556 -0 333611111 -0 333055556 -0 333055556 -0.332777778 -0333055556 -0 332777778 -0 332777778 -0 333333333 ~0333333333 -0,3325 -0 333333333 ~0334166667 -0.333611|11 -0.3325 -0 333333333 -0 333611111 -17 66416667 X(XA-XB)"2 0111667361 0111111111 0111296373 1777037114 0111667361 0111111111 011148179 0111296373 0111111111 011055625 0111667361 0110926003 0111296373 0111111111 011148179 0110926003 0111111111 0110926003 011148179 0110926003 0110187114 0.11055625 0110926003 011055625 011055625 0.111667361 0111296373 0100982716 0111111111 011055625 012211142 0111667361 0111111111 0110926003 0111296373 0110926003 0110926003 0110741049 0110926003 0110741049 0110741049 0111111111 0111111111 011055625 0111111111 0111667361 0111296373 011055625 0111111111 0111296373 7220515664 Y(YA - YB) -0 125 -01 125833333 -0 125555556 -0 1261111 11 -0125555556 -0 125555556 -0.125 -0 125 -0 124444444 -0 124166667 -0, 124722222 -0124444444 -0 124722222 -0. 124722222 -0125 -0.125 -0 124444444 -0.125 -0 125 -0 125 -0.125 -0 124444444 -0 124166667 -0 125277778 -0 125277778 -0 125277778 -0 125 -0 124722222 0112444444444 -0 124722222 -0 124166667 -0 124444444 -0 124444444 -0 125277778 -0 125277778 -0 124722222 -0 124444444 -0 124722222 -0 124444444 -0 125277778 -0 125277778 -0 126111 111 -0 126388889 -0 125833333 -0 126111111 -0.125555556 -0. 124722222 0 125 -0 124444444 -0 123888889 -6 249166667 Y(YA - YB)A2 0.015625 0015834028 0 015764198 0 015904012 0 015764198 0 015764198 0 015625 0 015625 0 01548642 0 015417361 0 015555633 001548642 0015555633 0 015555633 0015625 0 015625 0 01548642 0.015625 0 015625 0 015625 0 015625 0 01548642 0 015417361 0 015694522 0015694522 0 015694522 0015625 0015555633 0 01548642 0 015555633 0015417361 001548642 0 01548642 0 015694522 0015694522 0 015555633 001548642 0015555633 0 01548642 0 015694522 0 015694522 0 015904012 0 015974151 0 015334028 0 015904012 0 015764198 0 015555633 0 015625 0 01548642 0 015348457 0781057485 111 (X(XA - XB)—uX(XA - XB))"2 0 000365447 0000398003 01000386996 0 959953607 0 000365447 0000398003 0 000376144 0 000386996 0 000398002 0 000431947 0000365447 0000409163 0 000386996 0000398003 0000376144 0 000409163 0000398002 0000409163 0000376144 0 000409163 0000455348 0000431947 0 000409163 0 00043 1947 0 000431947 0000365447 0 000386996 0001260644 0000398003 01000431947 1 47371E-05 0 000365447 0000398003 0000409163 0 000386996 0000409163 0000409163 0 000420478 0 000409163 0000420478 0 000420478 0 000398002 0 000398003 0 00043 1947 0 000398003 0.000365447 0 000386996 0000431947 0 000398003 0 000386996 01980059983 (Y(YA - YB)-uY(YA - vewz 2 77778E-10 722513-07 33274388-07 1.27188E-O6 3.27438E-07 3,27438E-07 2.77778E-10 277778540 2 90401E-07 6669441507 6 8179E-08 2 90401E—O7 6.8179E-08 6.8179E-08 2.777785-10 2777785-10 290401E-07 2.7777813-10 2.777785-10 2.777788-10 2.77778E-10 2.9040113-07 6.66944E-07 8 669755-08 8 669751308 8 66975E-08 2.777785-10 68179E~08 2 90401E-07 681795-08 6.66944E-07 2.90401E-07 2.90401E-07 8 66975E-08 8 669755-08 6 817913-08 2 90401E-07 6.8179E-08 2,90401E-07 8 66975E-08 8.66975E-08 1.27188E-06 1 975595-06 722513-07 1.271885-06 3 274385-07 68179E-08 277778E-10 290401E-07 1 197815-06 1.5804E-05 N= 50 df= 48 ux(XA-XB) = ~0.353283333 uy(YA-YB) = .0 124933333 StDX(XA-X8) = 0 000384207 StDY(YA-YB) = 9.99067E-l4 ux(XA-XB) +/- 2 StDX(XA-XB) = uy(YA-YB) +/- 2 StDY(YA-YB) = Use Receiver A and Receiver 8 to measure point P and pount 0, respectively: The coordinates of point P measured by Receiver A = The coordinates of pornt Q measured by Receiver B = The landmarks‘ coordinates of point P and point Q: pomt P = ( 121 point Q =- ( 121 24 25 2429166667 V’aF—(Sigma x(XA-XB))"2= Var=(Sigma y(YA-Y8))"2= -0 353283333 -0 124983333 00196012 3.1608E-07 +/- +/- 121 0002778 1209994444 Bias between the landmarks coordinates of point P and the measured coordinates of point P from Receiver k - > Biu = ( The predicted coordinates of point 0 measured by Receiver 8 Component X= IZI 3530056 +I- Component Y: 24 41637222 +l- -_._.._:==> The differential value between the predicted coordinates of pount O and the landmark coordinates by Receiver 8 = at 95% confidence at 95% confidence -0 000277778 Component X = 0 353005556 +/- 0 000768414 Component Y = 0124705556 +/- I 998l3E-13 Convert to Sexagesimal notation: +/- Degree Minute Component X = - 0 0 Component Y = - 0 0 112 -0 000277778 0000768414 199813E-13 Second 0.488016 0 8799984 ) t 95% confidence t 95% confidence +/. +/- 0000768414 1998131543 24 2502 7778 24 29222222 Degree 0 0 at 95% confidence at 95% confidence Minute 0 0 Second 5 01937E-07 7 65418E-10 Stnkholc 1D 1 cad°ubw~ ..._.._......_ me-un—c 16 61 62 63 64 65 67 68 69 70 7 1 72 73 74 75 76 77 78 Area (hectare) 112736 91118 25271 9456 11129511 121111 74894 6676 56694 (1796 22561 1172 8272411964 16159 59118 99l72.76~111 126369728 21599 60114 1211117 466 4611253 651111 1110969692 “10924 (111-14 11175045 179 19916 968 1111110 1676 4165.1 91411 2118111 9244 144157 (A1111 57911211712 55595 11124 117411611 246 23556 9656 11114590183 271611 (1612 152742 54011 14511 2911 111169117 765 72628 254 1112709 7436 417121 1212 1 111219411 41 I 14419 .1472 4171-1 1108 574123 7724 67216 9168 15452 112 (11459 581111 111.190 546 52951 (1632 12112116576 5191 1172 1117211 2411 75141 (11124 6621191 61114 21119118492 222245 11912 279111119411 1111191791411 4112460112 12111141 (1524 111122 9916 17511111 752 1151776 21 7311644 796 4921111 6468 111494 6916 275369 252-1 5111211 1124 54909111114 756467114114 29512 (156 77161 (1812 17116211 4296 111011.15 4052 111699 6024 72917 1104 11125 6508 1111719 9116 295157 91114 211911 94.12 91111211 212 68679 2 41276 1992 41962 9912 l 111230.116 Locauon (am) A a: >>>>>>>nnww>>>>>>>fifi>>>wawmmm>>nn>>>aawan>nnfi>>>>>>>>h>>>>>>>>>>>>>>>>>>>>>>> APPENDIX G Sinkhole ID 79 m1 111 112 83 84 85 86 87 88 89 5X1 91 92 93 94 95 96 97 98 99 Ill) 101 102 103 104 105 Ill) 1117 1118 I119 110 111 112 113 114 115 116 117 118 11‘) 120 121 122 123 124 125 126 127 128 129 I30 131 132 133 134 135 136 137 1311 139 140 141 142 143 145 146 147 148 149 150 151 152 153 154 155 156 113 Area1hcctarc) 121071 1264 826119 7568 511273 1714 111111156 512 17111115 1596 325.51 941111 12622 62 25445 6416 177114 11111111 241181199 8611 11049211224 541117 11111111 121111125 612 114556 91156 151175 01141 2961610 499 4116477 8452 411417 71192 111562 244 205556 11456 11611115 271 1116979 122 68404 4832 3191114 1171111 5112115 622 211124 1.124 5329114 9116 (16412 7116‘ 679511116114 19421 116118 835411 24611 7114.11 6464 109171 626 16.517 1114 1127210964 55252 416-1 597511 9114 196416 1424 571196 .5656 12217 7011 255,114 926 241091-1624 27609 111114 711291 1612 991169 746 15266 7692 611505 244 30491 56411 (1261169 198 111577 7216 21911116an11 2111-19 2264 140964 11511 14197911 749 61364 11652 1111111111 5511 7.5272 4612 2114111111 21211 241922 4112 791115 419(- 654115 6172 15644 511411 421111 (17114 579446 41114 181152 44114 549124 5416 (19915 4256 111112691152 111191691111 651197 6924 491195 411111 21111215 1144 (1111126 7476 19112 111114 1471795256 17272 1111111 519111 11112 14371 9196 [when (area) A nan)>>>>nn>n>>>>nnn>>1:1wn>mnnn>>12106000)nn>nnn>w>mnnnnnnn>n>nnnn>>n>wn>>>>>n>nw Sinkholc ID 157 1511 159 1611 161 162 161 164 I65 166 167 1611 169 1711 171 I72 17‘. I71 I75 176 177 I711 179 11111 1111 1112 1111 184 1115 186 1117 11121 1119 1911 191 I92 I91 194 I95 I96 197 1911 199 2611 2(11 202 2111 2114 2115 2116 2117 21111 2119 2111 211 212 21.1 211 215 216 217 2111 219 220 221 222 22,1 224 225 221. 227 ‘51s -- Table 6.10 The Area and Location of the Sinkholes in the Te-Chi Reservoir Watershed All: (hectare) 55492 7936 2571114 907 2991197916 20054 3264 2113095 6624 (116601 11576 (1091 15 8248 44916 1968 2116180996 613992048 111464411292 1222411 976 1911997 9112 82-149 .1796 11435011611 32107 526 41791 2912 68747 8792 291106 7728 427117 1416 1261697211 1.179111 664 6119.1 1672 13715 84 220460232 343396 1630100812 20717031!“ “[125 6208 14113 5624 2111433 (1688 61111429 03211 1011719 1716 4136119 1612 1146416 111104 5110602 (18118 1193911 7892 87188 2444 96156 9176 115608 (1228 411711115 . 7688 12114976564 1576117 4412 105697 2888 29111756576 111947 096 161224 422 1698 3 , 7492 53192 11404 6716112576 1491123 67411 1099176256 44092 (1464 190144 41128 41171 1804 757531576 40864124 28604 11868 37121 1076 1711909116 16757 72-18 22.1379 (1911 536727948 324852616 40967 1428 1721111 2068 110148 (1264 101192992 21 91128 9952 941121 8248 36199976 43576 9524 Localion (area) C nmnnnnnnnfinnnnnfinnnnfifinnnnnand:caracecommwwwmwwmwwmamwmwmmmamfinn>>>>>>cnw>>>>