UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A evaluating soil water drainage of a humid mountain forest site in southwestern British Columbia by… Cheng, Jie-Dar 1972

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1972_A6_7 C44.pdf [ 3.13MB ]
Metadata
JSON: 831-1.0101732.json
JSON-LD: 831-1.0101732-ld.json
RDF/XML (Pretty): 831-1.0101732-rdf.xml
RDF/JSON: 831-1.0101732-rdf.json
Turtle: 831-1.0101732-turtle.txt
N-Triples: 831-1.0101732-rdf-ntriples.txt
Original Record: 831-1.0101732-source.json
Full Text
831-1.0101732-fulltext.txt
Citation
831-1.0101732.ris

Full Text

EVALUATING SOIL WATER DRAINAGE OF A HUMID MOUNTAIN FOREST SITE IN SOUTHWESTERN BRITISH COLUMBIA BY TWO FIELD TECHNIQUES by JTE-DAR CHENG B.S. , Taiwan Provincial Chung Hsing University, 1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of Forestry We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August, 1972. In p resen t ing t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Columbia, I agree tha t the L i b r a r y sha l l make i t f r e e l y a v a i l a b l e f o r reference and s tudy. I f u r t h e r agree t h a t permiss ion f o r ex tens ive copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . I t i s understood tha t copying o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be al lowed w i thou t my w r i t t e n pe rmiss ion . Department o f ' d>/^3S£/^L r The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada Date /4u9t*?t- // CD-ABSTRACT This study was based on the premise that watershed management on mountainous forested land generally, and i n the Coast Mountains of British Columbia more specifically, w i l l be benefitted by. further knowledge of s o i l water drainage from the root zone of forested soils and by such development of methods of measuring s o i l water drainage as w i l l make increased sampling more feasible. The study was concerned with (1) the development and application of two methods for evaluating s o i l water drainage; a tension lysimeter system and a method based on Darcy's equation, (2) the exploration of the relationship between r a i n f a l l , s o i l water drainage and streamflow. It was carried out at a forest site of Jamieson Creek Experimental Water-shed i n the upper Seymour River basin, southwestern B.C. near Vancouver. The tension lysimeter system incorporated a simple capability for manually regulating suction on the lysimeter plate i n close conformity with the tension i n surrounding s o i l and also a lysimeter plate that ensured satisfactory hydraulic contact between i t and the s o i l . The application of Darcy's equation for calculating s o i l water drainage was based on f i e l d determinations of both s o i l hydraulic conductivity and hydraulic gradient at the study site. Soil water drainage rates measured by the tension lysimeter system and those calculated by Darcy's equation showed good agreement, although the former were consistently and slightly higher. It was concluded that each method can provide reasonable estimates of s o i l (ii.) water drainage and may be particularly useful i n developing countries where a cheap labor source permits extensive and frequent s o i l water drainage sampling within a watershed. Soil water drainage amounts obtained by these two methods during each of two drying periods were in good agreement with those estimated from a water balance equation. Possible sources of error associated with s o i l water drainage measurements by tension lysimeter and by the method of Darcy's equation are discussed and possible improvements suggested. This study also indicated that, i n the humid coastal region of southwestern British Columbia, s o i l water drainage i s a major component of water balance for the root zone of forested s o i l and deserves further study. The time trends of s o i l water drainage were found to be similar to that of streamflow from the small watershed containing the study site . This suggests that the geologic, edaphic, topographical, clima-tological conditions favor a large and direct contribution of root zone s o i l water drainage to streamflow. ( i i i ) TABLE OF CONTENTS Page ABSTRACT i LIST OF APPENDICES i v LIST OF FIGURES v LIST OF TABLES v i ACKNOWLEDGEMENTS y i i A. INTRODUCTION 1 B. PHYSICAL BASIS 6 C. INSTRUMENTATION AND MEASUREMENTS 9 (a) Tension lysimeter system 9 (b) The method of Darcy's equation 17 1. Hydraulic gradient determination 17 2. Hydraulic conductivity determination 20 (c) Supplementary instrumentation and measurements 21 D. RESULTS 22 (a) Soil hydrologic characteristics 22 (b) Soil water drainage measured by tension lysimeter ... 25 (c) Soil water drainage calculated by Darcy's equation .. 26 (d) Comparison between two methods of drainage evaluation 30 (e) Soil water drainage evaluated by water balance equation 35 (f) Relationships between r a i n f a l l , s o i l water drainage and streamflow 36 E. DISCUSSION 43 F. SUMMARY AND CONCLUSION 50 G. LITERATURE CITED 54 APPENDICES 57 LIST OF APPENDICES Appendix Page I The determination of the s o i l hydraulic conductivity characteristic 57 II The recorded hydraulic head values for the period of August 22 to September 26, 1971 60 III The calibration curve for the neutron s o i l moisture probe 66 (v) LIST OF FIGURES Figure Page 1 The layout of f i e l d instrumentation at the study site 10 2 Diagram of a tension lysimeter plate 12 3 Outflow unit .• 15 4 Details of tensiometer cup, brass f i t t i n g and hydraulic line unit 19 5 Water retention curves for the 0-40 cm; 40-80 cm and 80-100 cm s o i l layers of Strachan gravelly sandy loam 23 6 Hydraulic conductivity characteristic of the 60-80 cm layer of Strachan gravelly sandy loam 24 7 Mean daily hydraulic head profiles for August 22 to September 1 29 8 Comparison of mean daily drainage rates at the 80— cm depth of s o i l from measured values by tension lysimeter, and values calculated using Darcy's equation, for August 22 to September 26, 1971 34 9 Drainage rates from values calculated using Darcy's equation and from values measured by tension l y s i -meter for the 80— cm depth of s o i l , and r a i n f a l l intensity 38 10 Streamflow discharge from Jamieson Creek Experimental Watershed, drainage rate calculated using Darcy's equation for the 80— cm depth of s o i l at the study s i t e , and r a i n f a l l intensity 40 11 The daily streamflow discharge from Jamieson Creek Experimental Watershed and mean daily drainage rate for the 80-cm depth of s o i l at the study site ... 41 (vi) LIST OF TABLES Table Page 1 Values of bulk density for three s o i l layers 22 2 Mean drainage rates at the 80 cm depth of s o i l as values measured by tension lysimeter for various periods when f i e l d service was not available 27 3 Mean daily drainage rates at the 80 cm depth of s o i l from values measured by tension lysimeter, and from values calculated by Darcy's equation, for August 22 to September 26, 1971 31 4 Drainage from the 80 cm depth of s o i l during two drying periods as measured by tension lysimeter as calculated by Darcy's equation and as derived by a water balance equation 37 ( v i i ) ACM0WLSX3F1IENTS Sincere thanks are due to Dr. B.C. Goodell and Dr. R.P. Willington, Professor and Assistant Professor respectively, Faculty of Forestry (Forest Hydrology), University of British Columbia, for their guidance, criticisms, suggestions and encouragement i n a l l stages of this study. Special thanks are extended to Dr. P.G. Haddock and Dr. T.M. Ballard, Professor and Assistant Professor respectively, Faculty of Forestry, University of British Columbia. They served on my research committee and gave valuable help during the writing phase of this study. The suggestions, help, and encouragement from Dr. J. deVries and Mr. T.L. Chow, Assistant Professor and graduate student respectively, Department of Soil Science, University of British Columbia, are highly appreciated. My gratitude i s also extended to Dr. T.A. Black, Assistant Professor, Department of Soil Science, University of British Columbia, for his suggestions and criticisms. To Mr. C. O'Loughlin and Mr. A. Plamondon, fellow graduate students, go my thanks for their help during the f i e l d and writing stages. Appreciation i s given to Mrs. Mona R. Home for her assistance i n drawing the diagrams of this thesis. The cooperation of the staffs of the Greater Vancouver Water Board during the f i e l d phase of this study i s appreciated. ( v i i i ) This study was supported by research grants held by Dr. Goodell (N.R.C,. 67-5864) and Dr. Willington (U.B.C. Committee on Research Grant 29-9514). EVALUATING SOIL WATER DRAINAGE OF A HUMID MOUNTAIN FOREST SITE IN SOUTHWESTERN BRITISH COLUMBIA BY TWO FIELD TECHNIQUES A. INTRODUCTION A major goal of forest hydrology research i s to provide guide-lines for management that w i l l ensure that hydrological aspects of the forest environment are adequately recognized. The root zone of the s o i l i s , hydrologically, the most important s o i l layer of a forested watershed. The hydrologic properties of this zone and the related processes taking place within i t control or influence both the amounts and rates of water movement to stream channels. More specifically, the phenomena of surface runoff, i n f i l t r a t i o n , evapo-transpiration, s o i l water storage and drainage are a l l strongly governed by the s o i l characteristics of the root zone. To a large extent, watershed management i s the management of this root zone. Therefore, evaluation of the components of i t s water balance i s funda-mental to understanding watershed hydrology and the determination of potential hydrological impacts of forest management. The water balance of a root zone of s o i l can be simply expressed as follows: P = E T + R + A S + D where P i s the precipitation ET i s evapotranspiration R i s the surface runoff AS i s the change i n s o i l water content, and D i s the deep drainage. Although this equation i s simple and readily understandable, some of the components are d i f f i c u l t to measure. Precipitation, s o i l water 2. content, and surface runoff can be readily measured, but drainage and evapotranspiration are both d i f f i c u l t either to measure or to calcu-late. Drainage is. often estimated (Rose and Stern, 1965) or considered insignificant. I f each of the other variables i s evaluated, then an estimate of evapotranspiration i s obtained. This estimation of evapo-transpiration, however, can be c r i t i c i z e d , since i t reflects a l l errors i n the measurement or estimation of other variables. The error may be particularly large i f the drainage term i s considered to be negligible when, in fact, i t i s not. I t has been reported that neglect of the drainage term resulted i n errors i n calculated values of evapotranspira-tion of up to +28% and -29% (Rouse, 1970). A knowledge of s o i l water drainage i s also important i n under-standing the leaching of soluble nutrients, the resultant effects on stream water quality, and the effects of forest operations on these parameters. Many techniques have been devised to evaluate s o i l water drainage under f i e l d conditions. These include direct measurement by lysimetry and indirect methods based on determinations of s o i l water content or on measurement of hydraulic gradient and hydraulic conductivity of s o i l . High cost i s usually associated with these techniques and tends to r e s t r i c t their uses to only a very few points. The extrapolation of such results, even to small watersheds, must be approached with caution because of the spatial va r i a b i l i t y i n s o i l characteristics. Early studies of s o i l water drainage by confined type lysimeters showed that the s o i l - a i r interface between s o i l and the lysimeter 3. retards the normal s o i l water flow and that the lysimeter sidewalls also r e s t r i c t natural root development (Neal et a l . , 1937; Richards et a l . , 1939; Walliham, 1940; and Colman, 1946). Cole (1958) developed an unconfined type of tension lysimeter for collecting s o i l water drainage under f i e l d conditions. In his system, a constant suction approximating s o i l water tension at f i e l d capacity i s applied externally to reduce the s o i l - a i r interface problem. However, since the s o i l water tension i n a natural s o i l changes during periods of drainage, errors of measurement are induced by application of a fixed suction to the lysimeter. This error i s particularly large during the early stage of drainage following a major r a i n f a l l . I f this source of error can be reduced, the tension l y s i -meter could be very useful i n studies of s o i l water drainage, because i t not only can be used to measure s o i l water drainage directly, but also provides a means of collecting s o i l water samples for chemical analysis. The estimation of s o i l water drainage indirectly through measuring the s o i l water content changes by neutron s o i l moisture probe or other equipment i s possible. However, this method i s not sufficiently sensi-tive to changes i n rate of s o i l water drainage, particularly i n coarse-structured s o i l where a relatively large change i n s o i l water drainage rate can accompany only a small change i n s o i l water content. Another indirect method i n calculating s o i l water drainage i s based on Darcy's equation for s o i l water movement. Several investigators (Rice, 1967; Van Bavel et a l . , 1968; and Black et a l . , 1970) feel that 4. Darcy's equation, which relates the flux of water i n the s o i l to the driving force or hydraulic gradient and the s o i l hydraulic conductivity, offers the greatest potential for calculating root zone drainage. In the past, this method has relied on f i e l d measurements of s o i l water tension at various depths and on laboratory determination of relations between s o i l water tension and hydraulic conductivity. Alternatively, f i e l d measurements of s o i l water content have been followed by laboratory determination of water content and water tension relations and, f i n a l l y , of the relation between s o i l water tension and hydraulic conductivity. The s o i l sampling and laboratory work have been required because of d i f f i c u l t i e s i n f i e l d determination of s o i l hydraulic conductivity. The objectives of this study were as follows: 1. To measure drainage from a forested s o i l (root zone) by use of a tension lysimeter system designed to f a c i l i t a t e the application of suctions on lysimeter i n close conformity with the tensions existing i n the surrounding s o i l so as to minimize errors resul-ting from applying fixed suction to the lysimeter. 2. To modify and apply a system for f i e l d determinations of s o i l hydraulic gradient (hydraulic head gradient) and hydraulic con-ductivity, and to use the data so derived to calculate by Darcy's equation the drainage from the same forested s o i l . 3. To compare the drainage from the forested s o i l as measured with tension lysimeter system with that calculated from Darcy's equation and with that calculated from a water balance equation for the forested s o i l layer. 5. 4. To explore relations between precipitation ( r a i n f a l l ) , s o i l water drainage and streamflow. Both the tension lysimeter method and the Darcy's equation method used i n this study are simple and inexpensive. It i s intended that both methods w i l l be suitable for the f i e l d conditions of B.C. forest lands, and be compatible with the fact that large numbers of sampling points are required for evaluating the quantity and quality of s o i l water drainage and their changes resulting from forest management. 6. B. PHYSICAL BASIS In this study attention was restricted to one dimensional, iso-thermal, vertical flow of s o i l water. The equation of Darcy's law for such flow can be written as: V = - K (2) z z dZ where V (cm day 1) i s the rate of vertical drainage at depth Z (cm). K (cm day i s the s o i l hydraulic conductivity which expresses the a b i l i t y of s o i l to transmit water and i s strongly dependent on i t s s o i l 3 ~3 water content, w (cm cm' ). H (cm of water) i s the hydraulic head at depth Z (cm) and i s given by H = - T - Z, where T (cm of water) i s the matric tension of s o i l water and the depth Z (cm) i s positive downward dH and i s measured from the s o i l surface. Thus, the term i s the driving force or the hydraulic gradient (hydraulic head gradient) of s o i l water. The drainage term D (cm), at depth Z can be given by: D z J 2 V dt (3) z where (t^ - t^) i s the time period between observations. It i s observed from equation (2) that the calculation of the drainage rate from a given s o i l requires the monitoring of hydraulic dH gradient, ^ r , and the measurement of hydraulic conductivity, K , of the s o i l under investigation. The f i e l d determination of s o i l hydraulic conductivity was one of the purposes of this study and can be accomplished by use of theory 7. developed by Rose et a l . (1965). Under the condition where there i s no lateral flow and that any water flux (precipitation and evapotranspira-tion) through the s o i l surface i s prevented, the water balance equation for a given volume of s o i l with depth (Z) equal to zero at the s o i l surface and equal to L at the lower l i m i t of the s o i l volume i s : - D T = AS (cm) (4) Z=L or J 1 J - t ^ o (^jr) dz dt (cm) (5) where & dz dt = AS dt o By substituting (2) into (5) and integrating with respect to time, an expression for the rate of drainage at the lower li m i t of the s o i l volume where Z = L i s given as follows: L ( f ) dZ = K, <§) (6) o h The hydraulic conductivity, K, of the s o i l at depth L, can be calculated from the following equation: -L c & d z I dZ I L Lysimeters are devices for measuring the extraction of s o i l water from a given volume of s o i l . The conduction and retention of water by 8. s o i l i s strongly dependent on the s o i l water tension resulting from cohesive and adhesive forces of the s o i l water system. As can be seen from equation (2), the water i n the s o i l w i l l only move when a hydraulic gradient exists which i s determined by s o i l water tension and gravity. For water to move from the s o i l into an open space, the s o i l water tension within the s o i l must be zero. In zero tension lysimeters the s o i l must reach saturation before water w i l l move out of the s o i l and into a collection system. This i s the so-called s o i l - a i r interface (the region between the s o i l face and the lysimeter) problem. Some early s o i l water drainage studies of zero tension lysimeters (Neal et a l . , 1937; Richards et a l . , 1939; Walliham, 1940 and Colman, 1946) showed that i n order to maintain s o i l water drainage similar to natural condi-tions, a suction must be applied to the lysimeter face where s o i l water extraction occurs. However, the s o i l water tension i s not a constant value but varies with the s o i l water content. In theory, i f suction on the lysimeter i s equal to the s o i l water tension within the s o i l i n con-tact with the lysimeter, natural movement of water w i l l occur along the hydraulic gradient into the porous lysimeter plate. Thus, the suction applied to the lysimeter face must be varied to match the s o i l water tension i f natural drainage i s to occur. The tension lysimeter system used i n this study was so designed that suction on the lysimeter plate could be varied manually. The unconfined type lysimeter (Cole, 1958) has no sidewalls and has the added advantages of allowing free development of root systems i n and around the lysimeter and of permitting minimal s o i l disturbance during installation. This unconfined feature was adopted i n the design of the tension lysimeter plate used i n this study. 9. C. . INSTRUMENTATION AND MEASUREMENTS This study was conducted i n the summer of 1971 on the Jamieson Creek Experimental Watershed located near the headwaters of the Seymour River basin of the Greater Vancouver Water District and within the wetter subzone of the coastal western hemlock biogeoclimatic zone (Krajina, 1965). The study watershed i s within the Southern Pacific Coast Section (C-2) of Coast Forest Region (Rowe, 1959). A f l a t bench on sloping terrain was chosen as study site to ensure minimal inflow from the upslope but to provide the f a i r l y steep downslope required by the instrumentation to be used. The forest at the study site consisted mainly of western hemlock (Tsuga heterophylla Raf. Sarg.) with scattered western red cedar (Thuja plicata Donn.). The s o i l was a Strachan gravelly sandy loam5'* developed from mixed ablation t i l l and colluvium over basal t i l l . (a) Tension lysimeter system. The tension lysimeter system (Figure 1) included tension l y s i -meter plate, a hanging water column leading to an outflow unit, graduate collection cylinders and a system for maintaining conformance between the applied suction and that present i n the surrounding s o i l . As shown i n Figure 1, a rectangular p i t about 150 cm i n depth was dug into the s o i l and then a side excavation was made under the volume of s o i l above 80 cm depth that was to be studied. During the "B.C. Department of Agriculture, Soil Survey Division, Kelowna, B.C. unpub. A,A' Neutron probe access tubes B Tension lysimeter plate C,C' Tensiometers D Continuous water column E Manometers F Outflow unit G Collection graduate cylinder H Barrel I Rain gauge D-TK Z E : IDZ TF Figure I- The layout of field instrumentation at the study site-11. digging of the former rectangular p i t , s o i l core samples were taken for the determination of s o i l bulk density and water retention character-i s t i c s i n the laboratory. In the side excavation, the tension lysimeter plate was placed so that the lysimeter plate surface was i n good hydraulic contact with the excavation roof. A tensiometer was placed at 80 cm depth i n the adjacent s o i l together with three more tensiometers at different depths (Figure 1). The tension lysimeter plate was a modification of the apparatus designed for laboratory use by Dr. J. de Vries*. A cross section of the lysimeter plate i s shown i n Figure 2. The plate consists of two chambers, one lying above the other. The lower chamber constructed of 8 mm acrylic p l a s t i c , i s 8 mm deep. The top plate of this chamber i s perforated with a fine grid of 2 mm holes. In two corners of the lower chamber, 9.5 mm diameter acrylic plastic tubes were inserted: one tube for the outlet of water from the chamber and the other for de-airing the lysimeter. On the top of the perforated plate i s , f i r s t , a piece of nylon screen, second, a piece of nylon netting, and, f i n a l l y , a m i l l i -pore membrane. The nylon netting permits free flow of water to the holes i n the plate. The netting and the screen together serve to prevent the millipore membrane from being pulled into the holes of the plate and rupturing when suction i s applied. The millipore membrane was of the nylon type having a pore size of 7 microns with an a i r intrusion value of approximately 200 cm of water suction. The membrane allows for the "Personal communication. Assistant Professor, Department of Soil Science, University of British Columbia. A- Upper chamber B- Millipore membrane 0 Nylon net D- Nylon screen E- Rubber gasket F Outlet tube for releasing air bubbles G- Lower chamber H- 2 0 mm holes 1 • Outlet tube for draining soil water J- Upper rim Figure 2- Diagram of a tension lysimeter plate-application of this maximum of suction to the lysimeter plate, and also prevents colloidal particles from entering the collected water. Above the membrane i s an upper chamber formed by gluing an acrylic plastic rim ., 1.2 cm high on a frame 2 cm in width and 8 mm thick that extends around the perimeter of the tension plate. To ensure that no a i r leaks into the plate and to prevent the membrane from being punctured by the frame, a soft rubber gasket i s placed between the frame and the membrane and then the frame i s fastened to the tension plate by brass machine screws. Both the lower chamber and the upper chamber have a cross-sectional 2 area of 450 cm . The volume provided by the upper chamber i s f i l l e d with glass beads 29 microns i n diameter. These beads rest on top of the millipore membrane and ensure good hydraulic contact with the natural s o i l while also protecting the millipore membrane. Because of the par-t i c l e size of the glass beads, the lysimeter has an a i r intrusion value of 160 cm of water suction. Cole (1958) incorporated a fused alundum disc i n his tension lysimeter. Such a disc ensures no contamination of s o i l water leachate i n passage through the lysimeter and permits the application of suction to minimize the s o i l - a i r interface problem. However, a fused plate pressed against the s o i l tends to cause problems with respect to hydrau-l i e contact between the lysimeter and the s o i l . The use of glass beads to provide better hydraulic contact with the s o i l i s the main difference i n the lysimeter plate used i n this study. I f water quality i s to be tested, an inert material l i k e s i l i c o n carbide (Bourgeois, 1969) could be used because the glass beads tend to contaminate the leachate. Tozer 14. (1970) used a lysimeter plate with similar design to that used i n this study but of smaller dimensions. The use of larger dimension i n this study was with a view to relatively reducing the disturbance of s o i l by the upper rim of the lysimeter plate during installation. The hanging water column i s held by 6 mm O.D. flexible plastic tube. This water column i s connected to the outlet of the lower chamber of the lysimeter and permits a continuous water contact from the l y s i -meter plate to the outflow unit. A cross section of the outflow unit i s shown i n Figure 3. The outflow unit i s constructed of a block of acrylic plastic with cylinders and tubing inserted as shown i n the diagram (Figure 3). The larger cylinder of the outflow unit, with rubber stopper and capillary tube, provides a way of minimizing evapo-ration of the leachate. One collection graduate cylinder used i s 100 cc and marked i n 1 cc intervals. The other collection graduate cylinder i s 500 cc and i s marked i n 5 cc intervals. The s o i l water tension at the 80 cm depth of s o i l was determined from a tensiometer-manometer system which i s to be described i n the next section. A draining trench of 150 cm deep was dug around the study site so as to reduce the lateral flow effects, particularly when the s o i l was very wet. The s o i l immediately above the tension lysimeter plate was covered with plastic sheeting to reduce the evaporation from the side wall of the pedon. The actual assembling of the components of the tension lysimeter plate was done a short time before f i e l d installation for the reason A- Capillary tube B- Rubber stopper C- Outer cylinder D- Inner cylinder E- Outflow tube F- Rubber tube G- Inflow tube from tension plate H- Acrylic block 0- Bench mark level of outflow unit Figure 3- Outflow unit-16. that the millipore membrane must remain saturated when i n position and not allowed to dry out, otherwise the continuity of the hanging water column w i l l be lost. A detailed description of the assembly and i n s t a l l a -tion of lysimeter plate has been given by Tozer (1970). After insta l l i n g the assembled and saturated tension lysimeter plate i n the desired position at the study s i t e , the continuous water column was established by f i l l i n g a long piece of tubing with water and connecting one end to the short rubber tubing of the outlet of the lower chamber of lysimeter plate and the other end to the inflow tube of outflow unit (Figure 1 and Figure 3). The outflow unit was then moved 6 meters downslope and attached to a vertical iron pole connected with a board marked i n centimeters. The height on the board corres-ponding to the elevation of the lysimeter was i n i t i a l l y determined by allowing the water i n the tubing to seek i t s own level and by a hand level. After this level was determined, the tension lysimeter system was ready for measurement. By obtaining the s o i l water tension at the depth of 80 centi-meters from the reading of the appropriate manometer, suction equal to the s o i l water tension was applied to the lysimeter plate by lowering the outflow unit with bench mark level (0) as a reference point (Figure 3) to an appropriate distance below the elevation of the tension lysimeter. Following the application of a suction to the lysimeter plate, the drainage from s o i l moved out of the lysimeter, passed through the hanging water column, spilled over the inner cylinder of the outflow unit and f i n a l l y flowed out from the outflow tube to a collection graduate cylinder. 17. Field service, which included adjusting the suction applied to the lysimeter and measuring the drainage rate from the lysimeter, was provided for each daytime from August 22 to September 26 with excep-tions of September 6, September 17 and September 21. The suction applied to the lysimeter was adjusted according to the s o i l water tension at the depth of 80 cm. The s o i l water drainage rate was measured with the 100 cc graduate cylinder and the reading was taken every hour or less, depending on the trend of change i n s o i l water drainage rate. Nighttime f i e l d operation and measurement were also conducted during the f i r s t period of s o i l wetting by r a i n f a l l on August 31 and September 1. Whenever f i e l d service was not available, the suction applied to the lysimeter was the approximated medium value of the s o i l water tension when f i e l d service was terminated and that when f i e l d service was to be resumed. This approximated suction was determined by the experience of previous time trends of s o i l water tension. During the non-field service period, one 500 cc graduate cylinder or two 500 cc graduate cylinders connected by a syphon device replaced the 100 cc graduate cylinder. (b) The method of Darcy's equation 1. Hydraulic gradient determination The s o i l hydraulic gradient was measured with a tensiometer-manometer system. As shown i n Figure 1, the s o i l profile adjacent to the site excavation contained four tensiometers i n 20 cm increment from 40 cm to 100 cm depth. A l l tensiometers were constructed of a 18. ceramic cup (2 cm long, 1 cm O.D., 0.65 cm I.D.) cemented to a brass f i t t i n g with epoxy resin. The use of this type of ceramic cup restricts the range of measurement of s o i l water tension to approxi-mately 850 cm of water. Each of the four tensiometers was connected to a 200 cm high, water f i l l e d manometer 5 meters downslope by hydraulic lines of 0.32 cm bore nylon tubing. The use of nylon tubing has the advantages of allowing detection of entrapped a i r i n the measuring system and providing reasonable f l e x i b i l i t y . The tee-joint arrange-ment (Figure 4) enabled the insertion of spaghetti tubing for the rapid and efficient de-airing of the tensiometer-manometer system. Each of the hydraulic lines from the second brass f i t t i n g to the tensiometer cup was cut to a length of 45 cm and then this tensio-meter cup with 45 cm hydraulic line was installed i n the desired position one month before the actual connection with manometer. Each tensiometer cup with 45 cm hydraulic line was connected to i t s mano-meter by another long piece of hydraulic line three days before the beginning of measurements and l e f t for equilibration. Any a i r bubble detected i n the measuring system was taken out by inserting spaghetti tubing through the tee-joint and flushing with de-aired water. Soil water tensions were read every two hours or more often, depending on their time trends of change, for each daytime from August 22 to September 26 with the exceptions of September 6, September 17 and September 21. During the f i r s t period of s o i l wetting by r a i n f a l l on August 31 and September 1, nighttime f i e l d observations were also made. Hydraulic head values were obtained by adding values of s o i l water tension to the depth of each tensiometer. 19. •removable cap for de-airing by spaghetti insertion to manometer brass tee-joint hydraulic line brass f i t t ing ceramic cup Figure 4- Details of tensiometer cup, brass fitt ing and hydraulic line unit-20. 2. Hydraulic conductivity determination To evaluate the s o i l hydraulic conductivity i n the f i e l d i n a manner appropriate to Darcy's theory and that developed by Rose et a l . (1965), a 4-0 gallon barrel with both ends removed was pushed gradually into the s o i l as the s o i l beneath i t s lower rim was carefully removed. The barrel was inserted vertically to a depth of 85 cm at 3 meters distant from the tension lysimeter plate to create a cylinder of essentially undisturbed s o i l devoid of lateral s o i l water flow. After backfilling, an access tube for a neutron s o i l moisture probe was driven into the center of the s o i l column inside the barrel (Willington, 1971). The barrel with the s o i l inside was then l e f t to stabilize for two months before commencement of measurements. A tensiometer-manometer system similar to that described above was installed before the begin-ning of the s o i l hydraulic conductivity determination. In order to prevent the wetting of s o i l by r a i n f a l l during the period of s o i l hydraulic conductivity determination, a trench and a draining system about 150 cm deep were dug around the barrel. A plastic sheet of 16 square meters covered the top of the barrel and area around the barrel. The area adjacent to this 16 square meters area was also tramped to reduce i n f i l t r a t i o n and create surface runoff to the nearby trench. The measurements for determination of s o i l hydraulic conductivity were conducted i n the late September and early October of 1971. In the procedure of this determination, the s o i l i n the barrel was f i r s t satur-ated with water from two 20 gallon water containers and then allowed to drain while the s o i l surface was covered with plastic sheeting to prevent evaporation. Hydraulic heads at depths of 20, 40, 60 and 80 centimeters were measured for three weeks with the tensiometer-manometer system. Changes i n s o i l water content were determined for these four depths by a neutron s o i l moisture probe (Troxler) whenever the s o i l water tension changed by approximately 5 cm head of water. With the values of s o i l water content and correspondent hydraulic head so derived, the s o i l hydraulic conductivity were then calculated by equation (7). Details of procedure and results are given i n Appendix 1. (c) Supplementary instrumentation and measurements Rainfall input to the study site was measured with one tipping bucket recording rain gauge and one standard rain storage gauge. These two gauges were used to determine both the amount and timing of r a i n f a l l event. An access tube for neutron s o i l moisture probe was installed to a depth of 120 centimeters at the study site (Figure 1). Soil water contents were measured for depths of 30, 50, and 70 cm at the beginning and the end of this s o i l water drainage study, and measurements were also made whenever a significant change occurred i n the s o i l water status from drying to wetting or vice versa. A 120 V-notch weir existed 1 kilometer from the study site for measuring the streamflow discharge from the watershed of Jameson Creek. 22. D. RESULTS (a) Soi l hydro-logic characteristics Bulk densities, of s o i l samples from three different s o i l layers are given i n Table 1. This s o i l bulk density was evaluated i n the laboratory by the core method (Black, 1965). Water retention character-i s t i c s of s o i l were determined from the same s o i l samples with a small round tension plate lysimeter connected to a hanging water column (Black, 1965). The water retention curves for three different s o i l layers are presented i n Figure 5. As indicated i n the diagram, s o i l water tension increases as the s o i l water content decreases. Hydraulic conductivity characteristic for s o i l of the 60-80 cm interval, as determined from the method described above, i s shown in Figure 6. The plot i s on a log/log scale and i t i s apparent that as s o i l water tension increases, hydraulic conductivity decreases even more rapidly. For example, when the s o i l water tension i s equal to 5 cm of water the conductivity value i s about 100 times that at a tension of 50 cm of water. Table 1. Values of bulk density for three s o i l layers Soil layer 0-40 cm 40-80 cm 80-100 cm - mean values and ranges -Bulk density 0.87 ± 0.03 1.44 ± 0.05 1.58 ± 0.06 (gm/crn^ ) 23. 0 - 6 O - O L 0 • 1 0 - 2 0 - 3 0 - 4 0 - 5 0 - 6 0 - 7 0 - 8 0 - 9 0 - 1 0 0 - 1 1 0 - 1 2 0 T E N S I O N ( C M - H 2 0 ) Figure 5- Water retention curves for the 0 - 4 0 cm-, 4 0 — 8 0 cm-and 8 0 - 1 0 0 cm- soil layers of Strachan gravelly sandy loam-24. lOOOf < o > t-o Q Z o o 3 < Q: o >-x 100 -100 -10 TENSION ( C M - H 2 0 ) Figure 6- Hydraulic conductivity characteristic of the 6 0 - 8 0 cm-layer of Strachan gravelly sandy loam-(b) Soil water drainage measured by tension lysimeter Because the suctions applied to the lysimeter plate were manually adjusted to conform to s o i l water tensions at the depth of the lysimeter plate i n the ambient s o i l , natural movement of s o i l water should have occurred to the lysimeter plate. Implicit i n this assumption i s that s o i l water tensions right above the lysimeter plate were equal to those determined by the adjacent tensiometer at that depth. The drainage rates (cm/hour) were obtained by dividing the 3 lysimeter outflow rate (cm /hour) by the area of the lysimeter (450 cm2). The values of mean daily drainage rates were determined i n such a way that a l l recorded hourly drainage rates were taken into considera-tion. For example, the mean daily drainage rate for August 31 was derived from both daytime and nighttime hourly drainage rates. The mean daily drainage rates were obtained by averaging recorded hourly drainage rates of a given day to give values i n centimeters per hour. These mean daily drainage rates were then converted to centimeters per day for the convenience of comparison with those calculated by Darcy's equation which are to be presented later. The mean drainage rate for a given period when f i e l d service was not available was obtained through dividing total drainage i n centimeters by hours of this period to give a value i n centimeters per hour. This mean drainage rate was then converted to centimeters per day. 26. Values of mean drainage rate measured by the tension lysimeter when f i e l d service was unavailable, as given i n Table 2, can be compared with those of the corresponding mean daily drainage rate when f i e l d service was available as presented i n Table 3. Slight differences were found during drying cycles and greater differences during wetting cycles. These differences may be due to the time trends of s o i l water drainage and to deviations of applied suction from ambient s o i l water tension during periods when f i e l d service was unavailable. The differences caused by this l a t t e r factor were expected to be s i g n i f i -cant during wetting cycle periods. Because of the evidence of drainage rate errors during non-service periods, only values from periods of continuous servicing were used i n the later comparisons with drainage rates calculated by Darcy's equation. (c) So i l water drainage calculated by Darcy's equation The mean daily hydraulic head profiles were derived from a l l recorded hydraulic head values of various depths as recorded at two hour intervals (Appendix 2). For August 31 and September 1, the mean daily hydraulic head profiles were derived from both daytime and nighttime values. The mean daily profiles for August 22 to September 1 are given i n Figure 7 as an example. A noticeable aspect of these profiles i s that the change of hydraulic head with time i n the upper s o i l layers was usually more rapid than that of the lowest layer. This difference represents the withdrawal of s o i l water through evapo-ration from ground surface and transpiration by vegetation. The delay 27. Table 2. Mean drainage rates at the 80 cm depth of s o i l as values measured by tension lysimeter for various periods when f i e l d service was not available Month Date Hour Drainage rate values Month Date Hour measured by tension lysimeter (cm/day) August 22 1600* to August 23 0600- 0.853 23 1600 ii 24 0600 0.558 24 1800 ti 25 1000 0.387 25 2000 II 26 0800 0.251 26 1400 it 27 0700 0.220 27 2000 it 28 0800 0.140 28 1800 it 29 0700 0.128 29 1800 II 30 0700 1.013 30 1400 II 31 0700 3.176 02 1600 to September 03 0700 1.561 03 1000 II 04 1200 1.313 04 1800 ti 05 0700 3.304 05 1800 II 07 1000 1.713 07 1400 ti 08 0800 1.135 08 1600 ti 09 0800 1.062 09 1600 I I 10 1000 1.411 10 1600 it 11 1000 3.503 11 1800 tt 12 1200 2.816 *Pacific Standard Time 28. Table 2, continued. Drainage rate values Month Date Hour Month Date Hour measured by tension lysimeter. (cm/day) September 12 1600 to September 13 1200 " 13 1800 " 14 0800 " 14 1600 " 15 1000 11 15 1600 11 16 1000 " 16 1600 " 18 0700 11 18 1200 11 19 0800 " 19 1400 " 20 1000 " 20 1600 " 22 0700 22 1600 " 23 1000 " 23 1600 11 24 0700 " 24 1400 " 25 0800 " 25 1600 " 26 0800 2.334 1.781 1.476 1.027 0.584 0.381 0.324 0.272 0.204 0.183 0.161 0.123 91 8-23 8-22 8-24 831 8-25 8-26 8-27 8-28 8-29 8-30 HYDRAULIC HEAD (CM- H 2 0 ) Figure 7- Mean daily hydraulic head profiles for August 22 to September I-30. i n response of hydraulic head changes to r a i n f a l l events on August 31 and September 1 also increased with s o i l depth. The drainage rates at the 80 cm of s o i l depth, V^, were calcu-lated for every two hours for a given day by equation (2). The mean values of s o i l hydraulic conductivity were obtained from Figure 6, using the mean values of s o i l water tensions for every two hours at the depth dH of 80 cm. The average hydraulic gradient (hydraulic head gradient), for each two hour period was derived from hydraulic head profiles of 60 centimeter and 80 centimeter s o i l depth interval, and used for the hydraulic gradient value at s o i l depth of 80 centimeters. The values of drainage rate for every two hours i n centimeters per day of a given day were averaged to give the mean daily drainage rate. As shown i n Figure 7, only positive downward drainage from the designated s o i l layer (0 cm to 80 cm) was found by the theory of Darcy's equation. Judging from the time trends of hydraulic gradient, i f the dry weather had continued, negative upward drainage, such as that found by Willington (1971) would have occurred. (d) Comparison between two methods of drainage evaluation Mean daily drainage rates at the 80 cm depth as derived from tension lysimeter data and as derived by Darcy's equation during August 22 to September 26 are compared i n Table 3. I t i s apparent that drainage rates as derived from tension lysimeter data were consistently higher than those calculated by Darcy's equation. Table 3. Mean daily . drainage rates at the 80 cm depth of s o i l from values measured by tension lysimeter, and from values calculated by Darcy's equation, for August 22 to September 26, 1971. X, Drainage rates Y, Drainage rates measured by calculated by tension lysimeter Darcy's equation (cm/day) (cm/day) (cm/day) August 22 23 24 25 26 27 28 29 30 31 September 1 2 3 4 5 7 0.864 0.672 0.425 0.250 0.213 0.209 0.158 0.132 0.125 0.900 2.755 1.556 1.023 1.351 2.955 1.224 1.102 0.951 0.831 0.493 0.341 0.242 0.191 0.129 0.124 0.105 0.083 0.484 1.494 1.064 0.760 0.770 1.852 0.871 0.803 0.676 0.033 0.179 O.Q8t». 0.008 0.022 0.080 0.034 0.027 0.042 0.416 1.261 0.492 0.263 0.581 1.103 0.353 0.299 0.275 32. Table 3, continued. Date X, Drainage rates measured by-tension lysimeter (cm/day) Y, Drainage rates calculated by Darcy's equation (cm/day) X-Y (cm/day) September 10 2.844 1.512 1.332 11 3.955 2.410 1.545 12 2.344 1.897 0.447 13 2.152 1.439 0.713 14 1.583 1.151 0.432 15 0.901 0.841 0.060 16 0.852 0.544 0.308 18 0.382 0.379 0.003 19 0.353 0.323 0.030 20 0.345 0.265 0.080 22 0.244 0.200 0.044 23 0.189 0.152 0.037 24 0.163 0.106 0.057 25 0.145 0.080 0.065 26 0.123 0.062 0.061 A similar comparison of the two methods i s possible through regression of one on the other. Several regression equations were derived from data of different periods of observation. (a) V,. • , • + v =-0.0792 + 1.5902 V,^ , . . * (tension lysimeter) (Darcy's equation) This equation, based on a l l observations (33) during the period of this study, has a standard error of estimate of ± 0.2088 cm/day and an r of 0.9566. This regression i s shown i n Figure 8. ^ ^(tension lysimeter) ~ + 1.3020 ^(j^pqytg equation) This equation, based on 22 observations during two separate drying periods (from August 22 to August 30 and from September 12 to September 26), has a standard error of estimate of ± 0.1081+ cm/day and an r 2 of 0.973U. (c) V,. . , . . v = -0.1"+16 + 1.7473 V / T, , . . v (tension lysimeter) (Darcy's equation) This equation, based on 11 observations of a period of three mixed wetting-drying cycles from August 31 to September 11, has a standard error of estimate of ± 0.2017 cm/day and an r 2 of 0.9633. (d) V,. . -, . . x = +0.0398 + 1.0595 , . . s (tension lysimeter) (Darcy's equation) This equation, based on 9 observations during a drying cycle and continuous daily daytime f i e l d service, has a standard 2 error of estimate of 0.0498 cm/day and an r of 0.9669. 34. V D .DRAINAGE RATE CALCULATED USING DARCY'S EQUATION (CM- DAY") Figure 8- Comparison of mean daily drainage rales at the 8 0 - c m - depth of soil from measured values by tension lysimeter,and values calculated using Darcy's equation,for August 2 2 to September 26,1971-35. It should be realized that these above regression equations are for the purpose of comparison and not for the prediction of drainage rate as measured by tension lysimeter from that determined by use of Darcy's equation. (e) Soil water drainage evaluated by water balance equation As mentioned previously, s o i l water drainage can also be evalu-ated by the water balance equation for a given s o i l layer. For a given time interval and at a single s i t e , i f precipitation (P), evapo-transpiration (ET), surface runoff (R) and the change i n s o i l water content (AS) are measured and horizontal s o i l water flux can be neglected, the net vertical drainage of s o i l water (D) across the terminal depth of measurements (L) can be calculated. An attempt was made here to obtain the net vertical s o i l water drainage amount by a water balance equation for two rainless drying periods (from August 22 to August 30, and from September 12 to September 25). Surface runoff and precipitation were zero for these two periods, the s o i l water contents were determined at the beginning and the end of each of these two periods. Evapotranspiration at the study site was not measured because of the lack of a readily applicable method. Observed sunny conditions prevailing during much of these two periods made an approximation of evapotranspiration values possible through results from an earlier study i n the vi c i n i t y . The values of evapo-transpiration for these two periods were approximated from measurements by micrometeological method aborea forest site at U.B.C. Research 36. Forest near Haney, B.C., during a drying period of July, 1970 (Black and McNaughton, 1972). An evapotranspiration rate of 0.35 cm/day was adopted here. A comparison of s o i l water drainage amounts during these two periods from values derived from measurements by tension lysimeter, values calculated by Darcy's equation and values obtained by the water balance equation i s presented i n Table 4. The drainage amount values derived by tension lysimeter and Darcy's equation method are based on the assumption that the mean daily drainage rate represents the actual daily drainage rate. The drainage rates for September 6, September 17 and September 21 obtained by these two methods were average values of the preceding day and the following day. Because the actual evapotranspiration rate at the study site was not actually measured and also because the errors of measure-ments of other components were possibly confounded with the s o i l water drainage term i t s e l f , the drainage amounts calculated from the water balance equation can only be considered approximately comparable with drainage amounts obtained by the other two methods. (f) Relationships between r a i n f a l l , s o i l water drainage and streamflow An example showing the relationship of r a i n f a l l intensity to the s o i l water drainage rate at the 80 cm depth as measured by the tension lysimeter and from values calculated by Darcy's equation whenever data were available from August 22 to September 2 i s presented i n Figure 9. In the diagram solid lines connect the points representing two hour 37. Table 4. Drainage from the 80 cm depth of s o i l during two drying periods as measured by tension lysimeter as calculated by Darcy's equation and as derived by a water balance equation" Periods Drainage amount measured by tension lysimeter (cm) Drainage amount calculated by Darcy's equation (cm) Drainage amount calculated by a water balance equation (cm) August 22 to August 30 3.048 2.539 2.770-" September 12 to 10.565 8.070 7.900"** September 25 *The average s o i l water content (w) i s obtained by averaging the s o i l water contents (w) at the depths of 30, 50 and 70 cm by the neutron s o i l moisture probe (cnvVcmS). The integrated s o i l water content change (AS) i s calculated by multiplying the average change of s o i l water content (w-^  - w2) with the total depth under investigation (L) (cm). ** S = 80.000 (cm) x (wx - w2) (cm3/cm3) S = 80.000 (cm) x (0.310 - 0.236) = 5.920 (cm) ET = 9.000 (day) x 0.350 (cm/day) = 3.150 (cm) D = 5.920 (cm) - 3.150 (cm) = 2.770 (cm) S = 80.000 (cm) x (0.405 - 0.245) = 12.800 (cm) ET = 14.000 (day) x 0.350 (cm/day) = 4.900 (cm) D = 12.800 (cm) - 4.900 (cm) = 7.900 (cm) 38. icr o I 2 o UJ < or UJ < z g l O - 2 cr UJ 5 O CO 14 —' CM 5N 5 cr — ./I x Soil water drainage rate calculated using Darcy's equation-• Soil water drainage rate measured by tension lysimeter-ill I, h 11111. ii ~22 I 23 r 24 I 25 I 26 I 27 AUGUST, 1971 28 I 29 I 30 I 31 I I I 2 SEPTEMBER, 1971 Figure 9 Drainage rates from values calculated using Darcy's equation and from values measured by tension lysimeter for the 8 0 - c m depth of soil,and rainfall intensity-39. evaluations of drainage rate. As i n f i l t r a t e d r a i n f a l l passes through the s o i l p r o f i l e , both s o i l hydraulic conductivity and hydraulic gradient change. This i n turn tends to increase s o i l water drainage. The degree of influence on s o i l water drainage by r a i n f a l l input depends on r a i n f a l l intensity and duration, antecedent moisture condition of s o i l , the characteristics of the vegetation system and hydrologic properties of various s o i l layers. As shown i n the diagram, a time lag exists i n the response of s o i l water drainage at the 80 cm depth to the r a i n f a l l input. A further comparison of r a i n f a l l intensity, drainage rate at the 80 cm depth of s o i l calculated by Darcy's equation, and the stream-flow from the watershed of Jamieson Creek i s given i n Figure 10, for the period from September 2 to September 18. A comparison of daily streamflow discharge and mean daily drainage rate for the 80 cm depth of s o i l at the study site i s also presented in Figure 11 for the period from September 2 to September 26. In this diagram, solid lines connect the points representing one day evaluations of s o i l water drainage rate or streamflow discharge. As illustrated i n Figure 10 and Figure 11, the s o i l water drainage at the 80 cm depth of s o i l has many features similar to those of streamflow. During a recharge period from r a i n f a l l , perco-lation i n the s o i l rapidly increases, maximizes and then gradually decreases to a minimal condition. This pattern i s markedly similar to the ascension^naximun, and recession found i n streamflow hydrographs. These results of this study agree very well with the observation of Figure 10 Streamflow discharge from Jamieson Creek Experimental Watershed .drainage rate calculated using Darcy's eauation for the 8 0 - c m - deDth of soil at the s t u d v s i t e , a n d r a i n f a l l int*»ns'itv-i o r L I I I I I I 1 I 1 I 1 I I I I I I I I I I 1 I I 1 -4 10" A-o Soil water drainage rate calculated using Darcy's equation t Soil water drainage rate measured by tension lysimeter x Streamflow discharge measured by 120° V-notch weir which is located I ki lometer downstream from the soil water drainage study site I I I I M I 1 I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18 19 2021 22 SEPTEMBER,1971 24 26 10 1 0 10 Figure II- The daily streamflow discharge from Jamieson Creek Experimental Watershed and mean daily drainage rate for the 8 0 - c m - depth of soil at the study site-Cole (1967) that at the 36 inch depth in s o i l , the time distribution of s o i l water flow rate resembles the time distribution of streamflow. E. DISCUSSION As indicated i n Table 3, Figure 8 and above regression equations, the degree of agreement between values of drainage rate obtained by-tension lysimeter and by method of Darcy's equation seems to relate to s o i l water status and frequency of f i e l d service. The best agree-ment i n values of drainage rate appears i n a period when the s o i l was dry and daily daytime f i e l d service was continuous. An overall compari-son shows that s o i l water drainage rates measured by tension lysimeter were consistently and slightly higher than those calculated by Darcy's equation. The differences may be attributed to some factors which are systematically associated with these two methods and some factors which are not easy to identify. During nighttime and other periods when conformance between tension on the lysimeter plate and that i n the ambient s o i l was not maintained, an obvious source of error i n lysimeter drainage measure-ments existed. During continuous service periods, probably a major source of error was from instrumental inadequacy to prevent a time lag i n the adjustment of suction on the lysimeter plate. The tensio-meter-manometer system had response time of approximately 30 minutes. The error resulting from this inherent time lag must have been most significant during the wetting cycles when changes i n s o i l water tension were rapid. The most rapid change of s o i l water tension during the study period was + 3 cm of water i n 30 minutes. The correc-tion of the error inherent i n the time lag was not conducted because the change of s o i l water tension during wetting cycles depended on the r a i n f a l l amount and r a i n f a l l intensity. 44.-Errors i n drainage rate calculated by Darcy's equation probably came from errors i n determinations of both s o i l hydraulic conductivity and hydraulic gradient. Because of the slow response time and thermal sensitivity of the tensiometer-manometer system, the determination of hydraulic head values i s subjected to an error of ± 2 cm of water. Errors i n the determination of s o i l hydraulic conductivity probably originated i n errors irievaluations of both hydraulic gradient and s o i l water content. Soil water content between Z = 0 cm and Z = 80 cm can only be estimated to within ± 2.40 mm with the neutron s o i l moisture probe because the calibration curve has a standard error of estimate 3 3 of ± 0.003 cm /cm (Appendix 3). The s o i l hydraulic conductivity curve presented i n Figure 6 was determined during a drying cycle of s o i l water status. Obtaining hydraulic conductivity from this curve for calculation of s o i l water drainage of wetting cycles could induce some errors because of the hysteric phenomena on relationship between s o i l water tension and hydraulic conductivity (Topp and M i l l e r , 1966). i The tension lysimeter and the tensiometer array were only 30 cm apart, and thus the variations i n hydrologic properties of these two relevant s o i l profiles can be considered to be negligible. The appli-cation of suctions to the lysimeter plate with values only approxi-mately equal to "those of ambient s o i l , as occurred during periods of non-field service could, however, cause a small amount of horizontal water flow between the two s o i l profiles. This disturbance of natural s o i l water status could, i n turn, have created errors i n values of hydraulic gradient for use i n Darcy's equation and i n measurements of 45. drainage rates by tension lysimeter obtained during the following period when f i e l d service was available. The maximum deviation of suction applied to the lysimeter plate and that of ambient s o i l during non-service periods was ± 3 cm of water. (The error from this source w i l l be insignificant i f there i s no lysimeter installation and only the method of Darcy's equation i s used). The errors associated with human operations i n making reading of various measurements were guarded against by double checking. Errors from this source should have been of small magnitude and random. Because the discontinuity of f i e l d service and the delay i n response of the tensiometer-manometer system were probably the chief sources of error i n measurements of drainage by tension lysimeter, continuous daytime and nighttime f i e l d service as necessary would promote greater accuracy. Furthermore, for s t i l l better accuracy, an electronic device incorporating a tensiometer-differential pressure transducer system to regulate a source of negative pressure could be developed"*". Because of the high cost, such a refined tension lysimeter system would probably not be suitable for extensive spatial sampling of s o i l water drainage but good for accurate point measurements. The usefulness of a tension lysimeter system with an a i r intru-sion value of approximately 160 cm of water i s also limited to measurement of s o i l water drainage at tensions below this value. A 'Personal communication with Dr. T.M. Ballard, Assistant Professor, Faculty of Forestry, University of British Columbia. 46. further disadvantage of the tension lysimeter system used in this study i s that only downward s o i l water drainage can be measured. Two l y s i -meters could be used i n opposite directions to overcome this latter disadvantage. To evaluate downslope s o i l water drainage, which i s very important i n the mountainous forest land of B.C., a vertical series of small tension lysimeters might be installed to intercept downslope flow. The method based on Darcy's equation can be used to determine both positive downward and negative upward s o i l water drainage up to a s o i l water tension of the a i r intrusion value of the tensiometer cup. This method i s more expensive than the tension lysimeter system because i t requires a neutron s o i l moisture probe. However, since one neutron s o i l moisture probe i s enough for many measurements i f i t s transporta-tion to measurement sites i s feasible, this method i s s t i l l relatively inexpensive. It should be realized, however, that the method of Darcy's equation can only measure quantity and rate of s o i l water drainage while the tension lysimeter can also supply water samples for analyses of chemical content. The use of a water-filled tensiometer-manometer system i n the method of Darcy's equation has the advantage of low cost and easy maintenance. However, this system i s slow i n response to change i n s o i l water tension and subjected to significant thermal effects. A further disadvantage of this tensiometer-manometer system i s that as with 47. the lysimeter system, continuous f i e l d service i s necessary during periods of in s t a b i l i t y of s o i l water tension. The tensiometer-manometer system can be replaced by a recording type of tensiometer-pressure transducer system, which has a higher resolution, a lower thermal sensitivity and a fast response time. This type of system i s capable of continuously monitoring s o i l hydrau-l i c head values for the calculation of s o i l water drainage (Willington, 1971; Chow and de Vries, 1972). However, the cost i s much higher than that of tensiometer-manometer system. In order to reduce the cost but s t i l l continuously monitor the s o i l hydraulic head values, one pressure transducer for several tensiometer cups can be used (Rice, 1967 and KLute et a l . , 1968). Because of the d i f f i c u l t i e s i n obtaining a desired volume of undisturbed s o i l for evaluating s o i l hydraulic conductivity i n the laboratory, the i n s i t u method of determining hydraulic conductivity described i n this report and that of Willington (1971) appear to be an attractive alternative. The agreement i n s o i l water drainage values calculated by Darcy's equation and as measured by tension lysimeter and estimated by a water balance equation suggests that the hydraulic conductivity values were reasonably accurate. In the future, a high degree of accuracy i n evaluating the basis values of (1) s o i l water content (w); (2) s o i l water tension (T); (3) the time (t) and measuring depths (Z) of tensiometer measurements, w i l l probably be achieved. With the improvement i n the measurements of these basic values, the determination of s o i l hydraulic conductivity i n the f i e l d 48. w i l l be more accurate. Although the tension lysimeter and the method of Darcy's equation used in this study have limitations i n quantifying s o i l water drainage with very high degree of accuracy, their advantageous features of being simple and inexpensive make them suitable for use i n extensive spatial sampling. These two methods may be particularly useful i n developing countries where a cheap labor source permits extensive and frequent s o i l water drainage sampling within a watershed. The exploration of relationship between r a i n f a l l , s o i l water drainage and streamflow i n this study and the results of other studies (Nixon and Lawless, I960; Hewlett, 1961; Sartz, 1964; Cole, 1967; and Weyman, 1970) suggest a s o i l water-streamflow model i n which water-shed geological, edaphic, topographical and climatological conditions favor large contributions to streamflow from s o i l water drainage and relatively small or negligible amounts from saturated groundwater aquifers. For a given s o i l water system, the drainage rate w i l l be related to s o i l hydraulic conductivity and hydraulic gradient. As r a i n f a l l penetrates the s o i l , water moves rapidly downward i n response to an increased hydraulic conductivity and hydraulic gradient. This high flow of s o i l water drainage results, i n turn, i n increased stream-flow. In addition to r a i n f a l l interception by stream channel and surface runoff, the contribution of high s o i l water flow to the stream-flow may account for flood discharges. The decrease i n s o i l water content, resulting from evapotranspi-ration by vegetation and s o i l water drainage leaving the root zone, 49. cause decreases i n s o i l hydraulic conductivity and hydraulic gradient. This w i l l influence recession curves for both s o i l water drainage and streamflow. Hewlett (1961) stated that low flow rates through the s o i l were sufficient to provide minimum streamflow values. Therefore, s o i l water flow leaving the root zone observed may be very influential on both high and low stages of streamflow. F. SUMMARY AND CONCLUSION The evaluation of s o i l water drainage component of a water balance equation for the root zone of forested s o i l i s one of the fundamentals i n hydrology research of forest watersheds. With some complicated and expensive instruments which are currently available, the s o i l water drainage may be accurately evaluated at a point. However, a satisfactory understanding of the s o i l water drainage i n watershed hydrology and i t s possible modifications by forest manage-ment can only be achieved by extensive sampling with a relatively cheap but accurate method because of the spatial variations i n s o i l hydrologic characteristics even i n a small watershed. The results presented i n this paper are from the writer's attempts to calculate and measure s o i l water drainage completely in s i t u by two simple and inexpensive f i e l d techniques, and to obtain some knowledge about the s o i l water drainage and i t s relationship to r a i n f a l l and streamflow i n the humid mountainous forest area of southwestern British Columbia. A modified tension lysimeter system and a method based on Darcy's equation were used i n this study. This study was conducted at a forest site on Jamieson Creek Experimental Watershed, located at the headwater of Seymour River, near Vancouver, B.C. , i n the summer of 1971. i The tension lysimeter system incorporated a simple capability for manually regulating suctions to the lysimeter plate corresponding to tensions i n the surrounding s o i l and also a lysimeter plate that ensured satisfactory hydraulic contact between i t and the natural 51. s o i l . The application of various suctions to the lysimeter plate was accomplished by controlling the height of a continuous water column hung from the lysimeter plate. The s o i l water tension for the reference of applying suction to the lysimeter plate was obtained from the reading of a water f i l l e d manometer connected by hydraulic line to a ceramic tensiometer buried i n adjacent s o i l profile at the depth of lysimeter plate. Application of Darcy's equation to s o i l water drainage evaluation was based on i n s i t u measurements of hydraulic , gradient rwith •./ _ a tensiometer-manometer system and i n si t u determination of hydraulic conductivity through data from another tensiometer-manometer system and a neutron s o i l moisture probe. Soil water drainage rates derived from tension lysimeter data and those calculated from the method of Darcy's equation were compared for both drying and wetting periods of this study. Cumulative s o i l water drainage amounts for each of the two drying periods obtained by above two methods were compared with those calculated by a water balance equation. The relations of s o i l water drainage to r a i n f a l l rates and streamflow were explored. From the results reported i n this paper one may conclude that: 1. Judging from the good agreement between s o i l water drainage rates derived by tension lysimeter system and by the method of Darcy's equation, tension lysimeter probably can be used to quantify s o i l water drainage provided that: 52. (a) lysimeter plate i t s e l f has a good hydraulic conductive capability; (b) a satisfactory hydraulic contact i s formed between the undisturbed s o i l and the lysimeter plate; (c) suctions applied to the lysimeter plate are adjusted according to those of ambient s o i l . 2. With adequate determinations of both hydraulic conductivity and hydraulic gradient of s o i l i n s i t u , the evaluation of s o i l water drainage by calculation using Darcy's equation i s possible. 3. Both the tension lysimeter system and the method based on Darcy's equation gave reasonable estimate of s o i l water drainage during the two drying periods of entire measure-ments of this study. 4. Both methods used i n this study have their limitations i n quantifying s o i l water drainage with a very high degree of accuracy, particularly during the wetting stage of s o i l water status. With daytime and nighttime f i e l d service as frequent as necessary, one can only increase the accuracy of measurements to a limited extent. The limitations of the tension lysimeter system and Darcy's equation method may be overcome by u t i l i z i n g necessary automatic recording equipment. However, the cost associated with the refinements i s usually much higher than that for the method used i n this study. 5. The i n s i t u method for determination of s o i l hydraulic con-ductivity described i n this paper and that of Willington (1971) appears to be an adequate f i e l d technique for use i n forest s o i l conditions. 6. In humid coastal region of southwestern B.C., s o i l water drainage appears to be a major component of the water balance for the root zone of forested s o i l . In the study watershed, the time trends of s o i l water drainage were similar to those of streamflow discharge suggesting that the geological, edaphic, topographical and climatological conditions favor a large contribution of s o i l water drainage leaving the root zone to the streamflow. 54. G'.. LITFJ^TURE CITED Black, C.A. (Editor) "Methods of Soi l Analysis". 1965. Part I. Agronomy No. 9 Am. Soc. of Agron. Inc. Black, T.A., W.R. Gardner and C.B. Tanner. 1970. Water storage and drainage under a row crop on a sandy s o i l . Agron. Jour. 62: 48-51. Black, T.A. and K.G. McNaughton. 1972. Average Bowen-ratio methods of calculating evapotranspiration applied to a Douglas f i r forest. Boundary-Layer Meteorol. (in press). Bourgeois, W.W. 1969. A study of soils and leachates from two forest sites using tension lysimeters. Unpub. M.Sc. Thesis. University of British Columbia. 74 pp. Chow, T.L. and J. de Vries. 1972. Dynamic measurement of hydrologic properties of a layered s o i l during drainage and evaporation, followed by wetting. Paper presented at the International Symposium on Fundamentals of Transport Phenomena i n Porous Medium. Cole, D.W. 1958. Alundum tension lysimeter. Soil Sci. 85: 293-296. Cole, D.W. 1967. The forest soil-retention and flow of water. Proc. Soc. Amer. For., 1966: 150-154. Colman, E.A. 1946. A laboratory study of lysimeter drainage under controlled s o i l moisture tension. Soil Sci. 62: 365-382. Hewlett, J.D. 1961. Soil moisture as a source of base flow from steep mountain watersheds. U.S. Forest Serv. Southeast. Forest Exp. Sta. Pap. 132. 11 pp. KLute, A. and D.B. Peters. 1968. Hydraulic and pressure head measure-ment with strain gauge pressure transducers. Symposium on water in unsaturated zone (Wageningen). Int. Ass. Sci. Hydrol. Krajina, V.J. 1965. Biogeoclimatic zones and classification of British Columbia. Ecol. of Western North America. 1: 1-17. Neal, O.R., L.A. Richards and M.B. Russell. 1937. Observations of moisture conditions i n lysimeter. Soil Sci. Soc. Amer. Proc. 2: 35-44. 55. Nixon, P.R. and G.P. Lawless. 1960. Translocation of moisture with time i n unsaturated s o i l profiles. Jour. Geophys. Res. 65: 655-661. Rice, R. 1967. A fast response f i e l d tensiometer system, A.S.E.A. Transactions. 10: 430-438. Richards, L.A., O.R. Neal and M.B. Russell. 1939. Observations on moisture conditions. i n lysimeters: I I . Soil Sci. Soc. Amer. Proc. 4: 55-59. Rose, C.W. and W.R. Stern. 1965. The drainage component of water balance equation. Aust. Jour. Soil Res. 3: 95-100. Rose, C.W., W.R. Stern and J.E. Drummond. 1965. Determination of hydraulic conductivity as a function of depth and water content for s o i l i n situ. Aust. Jour. Soil Res. 3: 1-9. Rouse, W.R. 1970. Effects of s o i l water movement on actual evapo-transpiration estimated from the s o i l moisture budget. Canadian Jour, of Soil Sci. 409-417. Rowe, J.S. 1959. Forest Regions of Canada. Bull. 123 Forestry Branch, Canada Department of Northern Affairs and Natural Resources, Ottawa. 71 p. Sartz, R.S. 1964. Duration of percolation from a Loess s o i l . U.S. Forest Ser. Res. Note. LS-40. Topp, G.C. and E.E. Miller. 1966. Hysteresis moisture characteristics and hydraulic conductivity for glass-bead media. Soil Sci. Soc. Amer. Proc. 30: 156-162. Tozer, R.R. 1970. The f e a s i b i l i t y of using tension lysimeter for the study and measurement of water relations of humid coastal mountain s o i l s . Unpub. B.S.F. Thesis. University of British Columbia. Van Bavel, C.H.M., G.B. Sterk and K.J. Brust. 1968. Hydraulic properties of a clay loam and the f i e l d measurement of water uptake by roots. I-III. Soil Sci. Soc. Amer. Proc. 32: 310-326. Walliham, E.F. 1940. An improvement i n lysimeter design. Jour. Amer. Soc. Agron. 32: 395-404. Weyman, D.R. 1970. Throughflow on hillslopes and i t s relations to Streamflow Hydrograph. Int. Ass. Sci. Hydrol. B u l l . , 15 (2), 25-33. Willington, R.P. 1971. Development and application of a technique for evaluating root zone drainage. Unpub. Ph.D. Thesis, University of British Columbia. APPENDIX I THE DETERMINATION OF THE SOIL HYDRAULIC CONDUCTIVITY CHARACTERISTIC 58. ADDENDUM 1. The hydraulic conductivity, K ( T ) , was calculated according to the following equation: n C.L2Q EjCdCIVy/dtJ K ( T ) = 1 = 1dH x 1 4 4 0 m i x i / d a y dZ where C = the slope of the neutron moisture meter calibration curve 0.491 (cm3cm~3)/(I/I_) o L o n = depth interval associated with measured I./I , 20 cm. 20 c i s dt = time interval (min.) n = number of depth interval, 4. d ( l V l ) = change of neutron meter counts min to neutron standard counts min~l for time interval dt. Z = depth (cm) T = tension at 80 cm dH 2. The hydraulic gradient, ^ r , i s computed by dividing the mean differ-ence i n H at 60 cm and 80 cm depths by 20 cm. This i s assumed to be the hydraulic gradient at 80 cm. 3. The above equation i s a simplified form of that used by Willington (1971) which i s derived by differentiating the neutron c a l i -bration equation with respect to time and using Rose's equation. 4. The graph of K vs. T appears i n Figure 6. Determination of Hydraulic Conductivity Characteristic dH 4 Time Z = 60 cm Z=80cm ^ I d(I./I ) (minute) T* H* T* H* i=l 1 S K (T) 0 5.3 - 65.3 15 11.6 - 71.6 45 19.6 - 79.6 90 25.2 - 85.2 150 30.9 - 90.9 480 36.5 - 96.5 1320 41.6 -101.6 2700 48.6 -108.6 4630 57.9 -117.9 7180 66.8 -126.8 10380 77.8 -137.8 14340 86.9 -146.9 19080 95.9 -155.9 24780 3.5 - 83.5 0.9100 7.3 - 87.3 0.7850 13.5 - 93.5 0.6950 18.3 - 98.3 0.6550 23.4 -103.4 0.6250 27.6 -107.6 0.5550 32.2 -112.2 0.5350 38.6 -118.6 0.5000 47.3 -127.3 0.4700 55.3 -135.3 0.4200 63.5 -143.5 0.2850 71.7 -151.7 0.2400 79.6 -159.6 0.1850 0.2833 293.4863 0.2193 131.6805 0.0337 15.2372 0.0256 9.2113 0.0646 4.4290 0.1001 3.0362 0.0998 1.9114 0.1067 1.5635 0.0981 1.1574 0.0882 0.9279 0.0588 0.7367 0.0501 0.6227 0.0406 0.5444 " cm of water APPENDIX II THE RECORDED HYDRAULIC HEAD VALUES FOR THE PERIOD OF AUGUST 22 to SEPTEMBER 26, 1971. 6i. Hydraulic head values for four s o i l depths (cm of water) Date Time (hours) 40 cm depth 60 cm depth 80 cm depth 100 cm depth August 22 800* -104.9 -105.1 -114.9 -124.0 1000 -103.6 -105.8 -115.5 -124.3 1200 -102.3 -106.7 -116.1 -124.6 1400 -101.3 -107.9 -117.0 -124.8 1600 -100.5 -108.8 -118.0 -125.0 August 23 800 - 99.7 -114.0 -122.2 -127.6 1000 -100.0 -114.8 -122.7 -128.0 1200 -100.4 -115.6 -123.2 -128.4 1400 -101.6 -116.8 -123.9 -129.0 1600 -102.4 -117.5 -124.6 -129.7 August 24 800 -108.2 -121.4 -128.6 -133.0 1000 -109.1 -122.0 -129.1 -133.6 1200 -110.0 -122.9 -129.7 -134.1 1400 -110.7 -123.9 -130 ..2 -134.6 1600 -111.3 -124.7 -130.9 -135.2 1800 -111.5 -125.0 -131.2 -135.5 August 25 1200 -119.0 -129.2 -134.6 -137.9 1400 -119.6 -129.6 -135.0 -138.1 1600 -120.3 -129.9 -135.3 -138.2 1800 -120.7 -130.3 -135.6 -138.4 2000 -121.1 -130.5 -135.9 -138.5 August 26 1000 -126.6 -133.4 -138.6 -140.4 1200 -127.0 -133.8 -138.9 -140.5 1400 -127.4 -135.3 -139.2 -140.7 -'Pacific Standard Time 62. Date Time 40 an 60 cm 80 cm 100 cm (hours) depth depth depth depth August 27 800 -130.7 -138.4 -141.9 -143.5 1000 -131.3 -138.8 -142.4 -143.7 1200 -131.9 -139.2 -142.8 -144.0 1400 -132.5 -139.7 -143.3 -144.4 1600 -133.0 -140.1 -143.7 -144.7 1800 -133.5 -140.7 -144.0 -145.2 2000 -134.1 -141.3 -144.4 -145.4 August 28 1000 -139.6 -14 3-. 6 -147.6 -148.6 1200 -140.3 -144.1 -148.0 -149.0 1400 -141.0' -144.7 -148.5 -149.5 1600 -141.6 -145.3 -149.0 -150.0 1800 -142.1 -145.8 -149.4 -150.4 August 29 800 -146.9 -149.3 -152.9 -153.8 1000 -147.6 -149.8 -153.4 -154.2 1200 -148.3 -150.3 -153.8 -154.6 1400 -149.0 -150.8 -154.3 -155.0 1600 -149.5 -151.5 -154.9 -155.6 1800 -150.3 -151.8 -155.3 -156.0 August 30 August 31 800 -157.2 -155.4 -158.8 -159.4 1000 -157.6 -155.8 -159.0 -159.7 1200 -158.3 -156.7 -159.5 -160.0 1400 -158.5 -156.7 -159.5 -160.2 800 -140.0 -154.0 -158.9 -161.4 1000 -133.0 -147.0 -152.0 -159.9 .1200 -127.0 -134.0 -141.2 -153.6 Cl400 -117.0 -127.0 -134.'3 -141.8 1600 -111.7 -118.8 -127.5 -135.8 2000 -100.1 -109.8 -118.9 -131.0 2200 - 96.3 -107.2 -116.7 -129.9 2400 - 90.0 -103.0 -112.1 -127.4 Date Time (hours) 40 cm depth 60 cm depth 80 cm depth 100 cm depth September 1 400 - 87.0 -101.0 -109.0 -121.0 800 - 81.7 - 98.0 -106.1 -118.1 1000 - 82.0 - 98.0 -106.0 -115.0 2000 - 81.7 - 95.7 -105.7 -114.8 2200 - 81.0 - 94.9 -105.9 -114.8 2400 - 80.9 - 95.0 -106.0 -115.0 September 2 200 - 86.9 -101.2 -110.0 -117.9 800 - 85.8 -100.3 -109.7 -117.9 1000 - 85.5 -101.3 -110.5 -117.7 1200 - 85.4 -101.9 -110.9 -117.8 1600 - 87.6 -101.9 -111.9 -117.8 September 3 800 - 91.0 -105.8 -113.5 -119.8 1000 - 91.5 -106.8 -114.3 -120.5 September 4 1400 - 88.0 -107.9 -116.6 -123.7 1600 - 87.7 -107.5 -116.3 -123.9 1800 - 87.2 -107.0 -116.0 -123.8 September 5 800 - 75.8 - 93.0 -107.3 -117.0 1000 - 76.0 - 92.0 -105.2 -114.0 1200 - 76.3 - 93.7 -103.4 -112.1 1400 - 76.6 - 95.2 -102.7 -111.1 1600 - 77.2 - 95.6 -103.5 -110.4 1800 - 77.7 - 96.4 -104.0 -110.4 September 6 September 7 1200 - 85.7 -102.4 -109.4 -114.2 1400 - 86.5 -102.9 -109.7 -114.4 September 8 1000 - 86.9 -103.0 -109.3 -116.4 1200 - 86.5 -103.0 -109.2 -115.7 1400 - 86.0 -102.9 -109.0 -115.5 1600 - 85.3 -102.9 -106.8 -115.3 64. Date Time (hours) 40 cm depth 60 cm depth 80 cm depth 100 cm depth September 9 1000 - 87.3 -104.5 -109.5 -115.6 1200 - 87.8 -103.9 -109.0 -115.9 1400 - 88.4 -103.6 -108.8 -115.3 1600 - 88.7 -103.6 -108.7 -115.4 September 10 1200 - 78.3 - 96.6 -103.6 -109.0 1400 - 77.7 - 96.1 -103.5 -110.7 1600 - 77.7 - 95.6 -103.6 -110.3 September 11 1200 - 68.4 - 89.6 -101.1 -107.5 1400 - 70.5 - 91.1 -101.2 -107.8 1600 - 71.8 - 92.6 -101.6 -108.3 1800 - 72.2 - 93.8 -101.9 -108.5 September 12 1400 - 74.5 - 94.6 -102.4 -110.0 1600 - 75.0 - 94.9 -103.0 -111.0 September 13 1400 - 79.4 - 97.3 -104.5 -112.5 1600 - 79.8 - 97.5 -104.7 -112.6 1800 - 80.2 - 97.7 -104.8 -112.7 September 14 1000 - 84.8 - 98.8 -105.2 -114.2 1200 - 85.2 - 99.3 -105.6 -114.4 1400 - 85.4 - 99.6 -105.8 -114.6 1600 - 85.9 - 99.8 -106.0 -114.8 September 15 1200 - 90.2 -102.8 -108.2 -116.3 1400 - 90.4 -103.0 -108.4 -116.5 1600 - 90.6 -103.2 -108.5 -116.6 September 16 1200 - 94.6 -106.4 -110.6 -117.9 • 1400 - 94.9 -10 6-. 8 -110.9 -118.0 1600 - 95.3 -107.1 -111.1 -118.1 September 17 65. Date Time (hours) 40 cm depth 60 cm depth 80 cm depth 100 cm depth September 18 800 -105.0 -111.8 -115.4 -120.5 1000 -105.5 -112.1 -116.0 -120.6 1200 -105.0 -112.2 -116.2 -120.8 September 19 1000 -108.6 -113.3 -117.0 -121.7 1200 -108.9 -113.5 -117.1 -121.7 1400 -109.1 -113.9 -117.5 -121.8 September 20 1200 -112.0 -115.0 -118.0 -122.2 1400 -112.2 -115.0 -118.1 -122.2 1600 -112.3 -115.1 -118.0 -122.2 September 21 September 22 800 -115.1 -121.5 -121.5 -123.2 1000 -115.4 -121.7 -121.7 -123.3 1200 -115.6 -121.8 -121.8 -123.5 1400 -115.9 -121.9 -121.9 -123.6 1600 -115.0 -122.0 -122.0 -123.8 September 23 1200 -117.0 -121.4 -123.6 -125.3 1400 -117.2 -121.7 -123.9 -125.4 1600 -117.4 -121.8 -123.9 -125.6 September 24 800 -119.4 -123.3 -125.1 -126.9 1000 -119.5 -123.6 -125.2 -126.9 1200 -119.7 -123.6 -125.1 -127.0 1400 -119.8 -123.7 -125.2 -127.1 September 25 1000 -121.3 -125.0 -126.3 -127.2 1200 -121.3 -125.1 -126.4 -127.1 1400 -121.4 -125.2 -126.4 -127.3 1600 • -121.4 -125.2 -126.5 -127.3 September 26 1000 -123.4 -126.3 -127.3 -128.3 1200 -123.4 -126.4 -127.4 -128.4 APPENDIX III THE CALIBRATION CURVE FOR THE NEUTRON SOIL MOISTURE PROBE 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0101732/manifest

Comment

Related Items