ABSTRACT

Soil moisture plays a major role in regulating the energy balance at the land surface and the growth of plants. Research is ongoing to develop new procedures to monitor soil moisture across large regions of the country. A promising tool appears to be passive microwave sensors on airborne or space platforms. Passive microwave instruments, however, are affected by the vegetation on the land surface, which acts to degrade the accuracy of the soil moisture estimates. To further explore the use of passive microwave sensors to monitor soil moisture, a large multi-disciplinary research program was conducted in the Southern Great Plains region of Oklahoma in the summer of 1997. Included in the land surface monitoring were extensive measurements of the vegetation in grass/pasture and wheat fields located in the Little Washita watershed near Chickasha, at the Agricultural Research Service facility near El Reno, and at the Atmospheric Radiation Measurement /Cloud and Radiation Testbed (ARM/CART) Central Facility near Lamont, Oklahoma. This report presents these vegetative measurements and an analysis of the differences between the three sampling areas and the vegetative types. Green and brown standing biomass and surface residue biomass were sampled in a total of 48 fields, including one corn field. From these biomass samples the water content of the aboveground biomass was determined. Leaf area index (LAI), fraction of absorbed photosynthetically active radiation (fAPAR), and plant height were also measured. The greatest differences were observed between different vegetative covers. Green biomass was greatest in the grass/pasture fields, while brown biomass and surface residue were greatest in the harvested wheat fields. The most water was found in the grass/pasture canopies, with the majority of water located in the green standing biomass. Vegetation in the Central Facility and Little Washita sampling areas was very similar. Most of the grass fields in the El Reno sampling area were ungrazed and had significantly greater biomass, LAI, and fAPAR than the grass fields in the other two sampling areas. These data represent a snapshot of the vegetation conditions during the 24 June - 5 July 1997 period.

CONTENTS

INTRODUCTION

METHODS

Vegetation Sampling Scheme

Vegetation Sampling Procedures

Tilled Wheat Field Sampling

Derived Variables

Quality Control

Analysis Procedures

RESULTS

Wet Biomass

Dry Biomass

Aboveground Water Mass

Leaf Area and Plant Height

Photosynthetically Active Radiation

DISCUSSION

ACKNOWLEDGMENTS

REFERENCES

APPENDIX I: VEGETATION SAMPLING DATA

LIST OF TABLES

Table 1. Proposed and Actual Distribution of Vegetation Sampling Sites 2

Table 2. Variables for Which Means and Variances were Computed 7

Table 3. Mean and Standard Error of Measured Variables for Each Field at the Central Facility, El Reno, and Little Washita Locations

Table 4. Means and Standard Errors of the Means for Wet Biomass by Cover Type in the Central Facility (CF), El Reno (ER), and Little Washita (LW) Areas

Table 5. Means and Standard Errors of the Means for Dry Biomass by Cover Type in the Central Facility (CF), El Reno (ER), and Little Washita (LW) Areas

Table 6. Means and Standard Errors of the Means for Aboveground Water Mass by Cover Type in the Central Facility (CF), El Reno (ER), and Little Washita (LW) Areas

Table 7.Means and Standard Errors of the Means for Total Leaf Area Index (LAI), Green LAI, and Specific Foliage Area (SFA) by Cover Type in the Central Facility (CF), El Reno (ER), and Little Washita (LW) Areas

Table 8.Means and Standard Errors of the Means for Fraction Absorbed PAR (fAPAR), Fraction Reflected PAR by Soil (fRs), Fraction Transmitted PAR through the Canopy (fTc), and Fraction Reflected PAR by the Canopy (fRc) by Cover Type in the Central Facility (CF), El Reno (ER), and Little Washita (LW) Areas

LIST OF APPENDIX TABLES

Table AI.1. Vegetation Characteristics of Each Field, and Geographic Co-ordinates of Samples Taken at the ARM/CART Central Facility (CF), El Reno (ER), and Little Washita (LW) Locations

Table AI.2. Central Facility at Lamont Weather and Surface Conditions as Described by the Sampling Teams at the Time of Sampling

Table AI.3. El Reno Weather and Surface Conditions as Described by the Sampling Teams at the Time of Sampling

Table AI.4. Little Washita Weather and Surface Conditions as Described by the Sampling Teams at the Time of Sampling

Table AI.5. Vegetation Plant Height, Green and Brown Standing Biomass, Surface Residue Biomass, and Water Content at the Central Facility Fields. Values are the Mean for Each Field Sub-sample

Table AI.6. Vegetation Plant Height, Green and Brown Standing Biomass, Surface Residue Biomass, and Water Content at the El Reno Fields. Values are the Mean for Each Field Sub-sample

Table AI.7. Vegetation Plant Height, Green and Brown Standing Biomass, Surface Residue Biomass, and Water Content at the Little Washita Fields. Values are the Mean for Each Field Sub-sample

Table AI.8. Leaf Area Index, Percent of Photosynthetically Active Radiation Transmitted to the Soil (fT_C), Reflected above the Canopy (fR_C), Reflected from the Soil Back into the Canopy (fR_S), and Absorbed by the Canopy (fAPAR) at the Central Facility Fields. Values are the Means of 10 Measurements for Each Plot.

Table AI.9. Leaf Area Index, Percent of Photosynthetically Active Radiation Transmitted the Soil (fT_C), Reflected above the Canopy (fR_C), Reflected from the Soil Back into the Canopy (fR_S), and Absorbed by the Canopy (fAPAR) at the El Reno Fields. Values are the Means of 10 Measurements for Each Plot.

Table AI10. Leaf Area Index, Percent of Photosynthetically Active Radiation Transmitted to the Soil (fT_C), Reflected above the Canopy (fR_C), Reflected from the Soil Back into the Canopy (fR_S), and Absorbed by the Canopy (fAPAR) at the Little Washita Fields. Values are the Means of 10 Measurements for Each Plot.

INTRODUCTION

Vegetation cover on the soil surface affects the accuracy of remotely sensed soil moisture estimates from passive microwave instruments (Jackson and Schmugge, 1991; Jackson, 1997a). This effect is due to the attenuation of the microwave emission from the soil and the additional radiative flux emission from the vegetation (Engman and Chauhan, 1995). While vegetation affects the sensitivity of the brightness temperature to soil moisture changes at all microwave frequencies, the effect is greater at higher frequencies (Jackson, 1997a).

When estimating soil moisture over a large region, it is necessary to develop a complete characterization of the surface vegetation over the region (Schultz, 1988). Such a vegetation characterization includes vegetation type, vegetative biomass above the surface, water content of the vegetation (Jackson et al., 1982), and surface residue and its water content (Schmugge et al., 1988). While remote sensing techniques can be used to detect leaf area index (Price and Bausch, 1995; Carlson and Ripley, 1997) and vegetation biomass (Wigneron et al., 1995), ground observations are needed to calibrate the models for different vegetation types and growth stages.

This document describes the vegetation sampling procedures and the vegetation data collected as part of the Southern Great Plains (SGP) 1997 Hydrological Experiment conducted in central Oklahoma from 18 June - 18 July 1997. The major objectives of the SGP 1997 Hydrological Experiment were to measure soil moisture using the L-band Electronically Scanned Thinned Array Radiometer (ESTAR), to evaluate the influence of soil moisture on the local surface energy budget, and to evaluate the influence of the mesoscale variability of the surface energy budget on the development of the convective boundary layer. The ESTAR, a passive microwave radiometer with an operation frequency of 1.4 gigahertz (GHz), was flown on an aircraft. Satellite data from the Special Sensor Microwave Imager (SSM/I) and Landsat Thematic Mapper (TM) were also acquired to produce a 30 meter (m) vegetation classification and a 30 m vegetation parameter database.

Vegetation sampling was concentrated in three sampling areas in central Oklahoma, the Little Washita watershed (LW) southwest of Chickasha, Oklahoma; at the United States Department of Agriculture (USDA) Agricultural Research Center (ER) near El Reno, Oklahoma, just west of Oklahoma City; and at the Atmospheric Radiation Measurement /Cloud and Radiation Testbed (ARM/CART) Central Facility (CF) near Lamont, Oklahoma. Vegetation sampling included measurements of green standing biomass, brown standing biomass, surface residue biomass, leaf area index (LAI), and plant height. Four independent photosynthetically active radiation (PAR) flux density measurements--two above the plant canopy and two below the canopy--were made at each sampling location. From these field samples, water content, fraction absorbed PAR (fAPAR), green and brown LAI, and specific foliage area (SFA) were derived to further describe the surface vegetation. This section describes the field sampling protocol and procedures used to compute the derived variables.

A detailed project description and sampling plan can be found in Jackson (1997b). The sampling scheme proposed in Table 1 was designed to collect samples from prairie/pasture (80%), wheat (10%), and other crops (10%). The initial scheme split prairie and pasture. However, during the sampling process, this separation was not clear so the prairie and pasture classifications were combined into a single class called grass/pasture. Samples were collected from a total of 48 fields, including one corn (Zea mays L.) field. Approximately 59 percent of the fields were grass/pasture, 35 percent wheat, and 6 percent other crops. The corn field and a Bermuda grass field harvested for hay (LW14) were the two other crops sampled.

Vegetation samples were collected from the fields used for gravimetric and profile soil moisture measurements. A single sample of ripe, unharvested wheat was taken from ER15 near the edge of the unharvested wheat to minimize disturbance to the standing crop. Fields with mixed vegetation included LW01 and CF06. Vegetation sampling crews were trained in LW01 which included short Bermuda grass and wheat stubble. The CF06 site was a partially tilled,

harvested wheat field. Two samples were taken from the harvested but untilled portion of the field and one residue estimate was taken from the tilled area of the field.

Vegetation measurements were taken during the period from 24 June-5 July. From 5-10 July, the drying samples were measured daily until all the samples were dry as determined by no further reduction in mass.

Vegetation Sampling Procedures

Each field was sampled at three different locations. Two sub-samples were collected at each sample location, resulting in six green standing biomass, six brown standing biomass, and six surface residue biomass measurements from each field. The leaf area index (LAI) and the components of fAPAR were measured at five locations for each sub-sample (one in the sampling frame and four within 3 m of the frame) resulting in 30 LAI and fAPAR measurements for each field. In addition, vertical and oblique photographs were taken of each sample location. Latitude and longitude were recorded as well as the general weather conditions at the time of sampling. Coordinates for the nearest meteorological station were also recorded. The meteorological stations were operated by either the Oklahoma Mesonet, the ARM/CART program, or the Agricultural Research Service (ARS) Micronet program. A summary of the procedures used to collect the different vegetation variables is presented below.

Sampling locations were approximately 100 m apart at a minimum of 100 m from the field edges. Once the sample location was identified, a three-sided square metal frame (0.71 m on a side) was pushed through the vegetation at the soil surface. A Global Positioning System (GPS) receiver (PLGR+, Rockwell International) provided latitude and longitude with an accuracy of 3 to 5 m. The second sub-sample was located within 5 m of the first sub-sample.

Descriptions of the vegetation type, growth stage, and conditions of the sky, vegetation, and soil during the sampling time were recorded on the data form (Figure 1). An oblique photograph centered on the sampling frame was taken from a distance of 3 to 5 m. A vertical photograph centered on the sampling frame was taken by holding the camera at shoulder height with the lens facing the surface. The roll and frame number of each picture were recorded on the data form.

Leaf area index (LAI) was measured in the sample frame with a plant canopy analyzer (LAI-2000, LiCor, Inc., Lincoln, Nebraska). A reading above the canopy was followed by five readings below the canopy as described by Welles and Norman (1991). During the measurements the canopy and LAI-2000 were shaded from the direct sun using a large umbrella. This procedure was repeated at four locations around the sampling frame, within 3 m of the

frame. The LAI-2000 actually measures the Afoliage area index@ and cannot distinguish between leaves, stems, and other structures that block incoming radiation. Leaves were predominate in the grass fields, thus it was assumed that the LAI-2000 was measuring the LAI. Standing wheat stubble with few or no leaves present dominated the the harvested wheat fields. In this case it was assumed that the LAI-2000 was measuring the foliage area index. This report refers to all LAI-2000 measurements as LAI.

A Sunceptometer (Decagon Devices, Pullman, Washington), an AccuPAR (Decagon Devices, Pullman, Washington), and a Line Quantum Sensor (LI-191, LiCOR Inc., Lincoln, Nebraska) were used to measure PAR flux densities. Although each sampling team used a different instrument, the errors introduced by the instruments should be small (Acock et al., 1994). Care was taken to level the instrument before each reading. Incoming photosynthetically active radiation (S) was measured above the canopy with the instrument level and facing upward. The PAR transmitted (T_C) through the vegetation canopy was measured near the soil surface with the sensor level and facing upward. The PAR reflected from the canopy (R_C) was measured with the instrument facing downward at 1 m above the vegetation. The PAR reflected by the soil (R_S) was estimated as the product of bare soil reflectance (R_BS) from tilled wheat fields and T_C (Daughtry et al., 1992).

Vegetation height was measured at five spots within the sampling frame and recorded on the data form. After the various measurements of the canopy within and around the sampling frame were completed, a meter stick was used to form the fourth side of the sampling frame, and the standing vegetation was clipped at the soil surface. All vegetation within the volume defined by the sampling frame was clipped and collected. If a plant extended from outside the frame into the frame volume, or from inside the frame to outside, only the part of the plant within the frame volume was clipped and included in the sample. All clipped vegetation was then separated into either green or brown vegetation that was bagged and weighed separately. The vegetative surface residue was collected from the soil surface and placed in a separate bag and weighed.

The collected samples were weighed in an area sheltered from the wind as soon as the team exited the field. From the time that the first sample was collected until it was weighed was approximately 90 minutes. At the end of each day, the samples were placed in a forced-air dryer at approximately 50EC until dry. After four days, several representative bags were weighed, allowed to dry for another day, and re-weighed. This procedure was repeated daily until there was no further decrease in mass.

Tilled Wheat Field Sampling

Vegetation samples were not taken from harvested wheat fields that had been tilled. Instead, a measure of the crop residue cover was obtained using a line-transect (Laflen et al., 1981; Morrison et al., 1993). The 15.2 m line had 100 beads or orange marks evenly spaced. At each sample location, the line-transect was stretched diagonally across the direction of tillage, and coincidences of the markers and pieces of crop residue on the soil surface were visually counted. The line-transect was moved to a different area and another count taken. This procedure was repeated until five counts were taken at each field sample location. The number of coincidences or Ahits@ divided by the total number of points observed (usually 500) is the fraction of residue cover.

The water content of the green and brown standing vegetation, and surface residue were computed as grams per square meter (g m^-2) and as percent of wet biomass. The percent water content (%Water) was computed as

where B_w is wet biomass and B_d is dry biomass. The difference between wet and dry biomass gives the mass of water held in the green and brown standing vegetation, and surface residue. Water content (W_m) in g m^-2 was computed as

where A_S is the area sampled: 0.5 m².

Absorbed PAR (APAR) is the algebraic sum of incoming and outgoing flux densities measured above and below a plant canopy (Asrar et al., 1989). Determination of APAR by the vegetation requires measuring four streams of radiation: 1) PAR incoming at the top of the canopy (S), 2) PAR transmitted through the canopy to the soil surface (T_C), 3) PAR reflected by the soil back into the canopy (R_S), and 4) PAR reflected by the canopy (R_C) (Asrar et al., 1989; Daughtry et al., 1992). Absorbed PAR of the canopy may be computed as

Since APAR is strongly affected by incident flux variations, the PAR flux measurements were normalized by S as follows:

where fT_C is the fraction of PAR transmitted to the soil surface, fR_C is the fraction of PAR reflected above the canopy, and fR_S is the fraction of PAR transmitted through the canopy and reflected by the soil back into the canopy. The mean soil PAR reflectance factor (R_BS /S) from the tilled wheat fields was 0.1026. The fraction of absorbed PAR (fAPAR) was calculated as

The green leaf area index (LAI_g) was computed as total LAI weighted by the ratio of green standing biomass divided by total standing biomass. Specific foliage area (SFA) was computed by dividing total leaf area index by total standing biomass. Specific foliage area provides a conversion factor to estimate biomass coverage from leaf area derived from reflectance data.

Quality Control

Data collected by the PAR sensors (AccuPAR, Sunceptometer, Line Quantum Sensor) and LAI-2000 leaf area meters were downloaded daily and scanned for missing observations and obviously erroneous data using a spreadsheet. The most common and easily corrected errors were measurements taken in the wrong order, e.g., transmitted PAR measured before incoming PAR. Data from the LAI-2000 were exported in both text and spreadsheet formats. The spreadsheet format was used to compute the field mean and standard errors.

Vegetation data from the SGP97 Vegetation Data Sheet were entered into a Paradox database and manually checked for entry errors. The vegetation data were further checked during data analysis by comparing means and variances of the samples within a field. When questionable data were found, data were checked against the original data sheets to determine if an entry error had been made. The data analysis was completed by importing the vegetation data from the Paradox database into a spreadsheet where calculations of the derived variables were made.

Analysis Procedures

The means and standard error of the means for the measured and computed variables (Table 2) were calculated for each field. The standard error of the mean is the standard deviation

divided by the square root of the number of observations used to compute the mean and standard deviation.

The vegetation variables (Table 3) were analyzed by computing the sum of the two sub-samples from each sample site, and the mean and standard error of the three sample sites in each field. An analysis of variance was conducted using the SAS PROC MIXED routine. The least significant differences computed from the analysis of variance were used to determine the significance of the means of vegetation within each location, and the differences in vegetation means across locations.

The vegetation biomass, LAI and PAR means and standard errors are presented for each vegetative type, and sampling area. A brief discussion of the differences between sampling areas and vegetation type is also presented. Appendix I, Table AI.1 presents vegetation type, growth stage, and latitude and longitude of each sample. Appendix I includes the weather and surface conditions at the time of sampling at the Central Facility (Table AI.2), El Reno (Table AI.3), and Little Washita watershed (Table AI.4); vegetation height and biomass measurements at the Central Facility (Table AI.5), El Reno (Table AI.6), and Little Washita watershed (Table AI.7); and LAI, fAPAR for the Central Facility (Table AI.8), El Reno (Table AI.9), and Little Washita watershed (Table AI.10).

Wet Biomass

Differences in wet biomass were greatest between vegetation cover types (Table 4). Wet green standing biomass was greatest in the grass/pasture fields and least in the harvested wheat fields. Significantly more wet green standing biomass was measured at El Reno than at either the Central Facility or the Little Washita watershed. Wet green standing biomass was slightly greater in the Little Washita watershed grass/pasture fields than at the Central Facility. However, this difference was not significant.

While not statistically significant, the wet green standing biomass in harvested wheat fields was greatest at the Central Facility and least in the Little Washita watershed. During the sampling period, the Central Facility was the wettest of the three sample areas and the harvested wheat fields had more weeds and volunteer wheat than the other two sampling areas.

Wet brown standing biomass was greatest in the wheat fields and least in the grass/pasture fields. There were no significant differences in the wet brown standing biomass among the three sampling areas in either the grass/pasture fields or the wheat fields. The greatest wet brown standing biomass in both the grass/pasture and harvested wheat fields was measured at El Reno, and the least at the Central Facility.

Wet surface residue showed greater differences between sampling areas than between cover types. The most surface residue was measured in the harvested wheat fields at El Reno,

and the least in the grass/pasture fields in the Little Washita watershed. These two extreme surface residue measurements were significantly different from each other. However, they were not significantly different from the other cover/sampling area combinations.

Total wet biomass at El Reno was significantly greater than at either the Central Facility or Little Washita areas. Total wet biomass in the El Reno grass/pasture and harvested wheat fields were not significantly different. Across sampling areas, the total wet biomass in the wheat fields was not significantly different. The greatest total wet biomass was found in the corn field, followed by the grass/pasture, and unharvested wheat fields.

Dry Biomass

Dry green standing biomass in harvested wheat fields at all three sampling areas was significantly less than in grass/pasture fields (Table 5). However, the dry green standing biomass in harvested wheat fields was not significantly different across the areas. Dry green standing biomass in grass/pasture fields was significantly greater in the El Reno area than in either the Central Facility or the Little Washita areas.

Dry brown standing biomass showed the same pattern across cover type and sampling areas as the wet brown standing biomass. Dry standing biomass in wheat was significantly greater than dry brown standing biomass in the grass/pasture fields.

Dry surface residue at El Reno, in both the harvested wheat and grass/pasture fields, was significantly greater than in grass/pasture fields in the Little Washita watershed. The remaining cover type and sampling area combinations were not significantly different.

Total dry biomass was greatest in the El Reno grass/pasture fields and least in the Little Washita grass/pasture fields. Both the grass/pasture fields in the Little Washita watershed and the harvested wheat fields at the Central Facility had significantly less total dry biomass than the El Reno grass/pasture fields. The greatest total dry biomass was measured in the mature unharvested wheat field. The corn field had the least total dry biomass.

Aboveground Water Mass

The water mass in the green standing biomass was greater in the grass/pasture cover types than in the harvested wheat (Table 6). This is due mostly to the greater green biomass in the grass/pasture fields. Water mass in the El Reno grass/pasture fields was significantly greater than in the grass/pasture fields in the Little Washita watershed and at the Central Facility. Water mass in the Central Facility grass/pasture fields was not significantly different from that in the harvested wheat fields at all three sample areas. The percent water content, computed by Eq. 1, was greater in the green biomass in harvested wheat fields than in the green biomass in grass/pasture fields. Percent water content in the wheat fields of the three sample areas ranged from 67 percent in the Little Washita wheat fields to 72 percent in the El Reno and Central

Facility wheat fields. The water content was 56 percent in the grass/pasture fields at the Central Facility, 66 percent in the El Reno grass/pasture fields, and 61 percent in the Little Washita watershed grass/pasture fields. The greatest water mass and water content (79 percent) in green standing biomass was measured in the Little Washita watershed corn field.

Water mass in the brown standing biomass ranged from 7 to 21 g m^-2 (Table 6). Water content in the brown standing biomass in the grass/pasture fields was 33 percent in the El Reno sample area, 23 percent in the Little Washita sample area, and 15 percent at the Central Facility. Water content of the brown standing biomass in harvested wheat fields was 5 percent at both the Central Facility and El Reno sample areas, and 4 percent in the Little Washita area. The differences between the percent water content in the two different cover types is due to the effect of the standing green biomass. Because the green standing biomass is transpiring, the humidity in the canopy is greater than in the harvested wheat fields, thus allowing the brown standing biomass to retain more water.

Water mass in the surface residue was greatest in the El Reno sampling area. The water mass in the surface residue in both the grass/pasture and harvested wheat fields was significantly greater than water mass in the surface residue of both vegetative types in the Little Washita sampling area (Table 6) . The absolute mass of water in the surface residue at both the Central Facility and in the Little Washita sampling area are approximately the same. However, the water mass in the Central Facility grass/pasture and harvested wheat surface residue was not significantly different from the grass/pasture surface residue water mass at El Reno. This was due to the large variations of water mass at both the Central Facility and El Reno sampling sites. The water content in the surface residue ranged from 21 to 40 percent in the grass/pasture fields at the three sampling areas, and from 6 to 50 percent in the harvested wheat fields.

Total water mass in the aboveground vegetation and surface residue was significantly greater in the grass/pasture field in the El Reno sampling area (Table 6) than at the two other sampling areas. Total water mass in the grass/pasture fields at both the Central Facility and Little Washita sampling areas was only slightly less than the total water mass in the harvested wheat fields at El Reno. There are large differences among the total water mass across vegetation type and sampling area. However, there are few significant differences due to the large variability of the total water mass as seen in the large standard errors of the means. Total water content of the aboveground biomass ranged from 6 to 38 percent in the harvested wheat fields to 43 to 55 percent in the grass/pasture fields in the three sampling areas.

Leaf Area and Plant Height

Consistent with the greater amounts of green and brown standing biomass in the grass/pasture fields in the El Reno sampling area, both total and green leaf area index (LAI) were significantly greater than the LAI of the grass/pasture fields at the other two sample areas (Table 7). The green LAI from the harvested wheat fields was significantly less than the grass/pasture fields. The largest average LAI was measured in the corn field. The mean grass/pasture LAI at the El Reno sampling area was only slightly less than the corn field LAI. Several El Reno grass/pasture fields had mean LAIs greater than the corn field (Table 3).

The specific foliage area of the grass/pasture fields was significantly greater than of the harvested wheat fields (Table 7) due to the absence of leaves in the harvested wheat fields. Although the specific foliage area was greater at the Central Facility sampling area than at the other two areas, it was not significantly greater.

The El Reno grass/pasture fields had the tallest average plant height, approximately 1.7 times taller than either the wheat fields in the three sampling areas or the grass/pasture fields in the Little Washita and Central Facility sampling areas (Table 7). The grass/pasture fields at the Central Facility and Little Washita sampling areas were not significantly different from each other or from the height of the harvested wheat fields. Plants in the corn and unharvested wheat fields were both taller than the harvested wheat and grass/pasture fields. With more samples, it would most likely demonstrate that the corn plants were significantly (statistically) taller than the grass/pasture fields at the El Reno area.

Photosynthetically Active Radiation

The plants in the corn field absorbed the greatest fraction of PAR followed by the grass/pasture fields within the El Reno sample area (Table 8). This is consistent with the greater

LAI and green biomass measured within these fields. The fraction of PAR reflected from the soil (fRs) was measured in harvested and tilled wheat fields in each sample area, and it is equal to the fraction of PAR reflected from the canopy (fRc). For the grass/pasture, harvested wheat, mature wheat, and corn fields, fRs is estimated using Eq. 6 with fRc of the bare soil for each sampling area used as R_BS.

Significant differences were found in the measured vegetation variables as a function of location and vegetation type. As expected, these differences were associated with differences in the amount of vegetative growth on the surface. The greatest variability was observed in the grass/pasture sites across the three sampling areas. Standing brown biomass in the harvested wheat fields was relatively constant across the three sampling areas. This would be expected because harvest practices tend to be rather constant and leave approximately the same length of stubble standing after the grain is harvested. Differences in the wheat green standing biomass were a function of the length of time after harvest and soil moisture between the time of harvest and sampling. In those areas where the soil moisture was high and more than a week had passed between harvest and sampling, considerable green biomass was observed in the harvested wheat fields. The largest green biomass in harvested wheat was measured in the fields at the Central Facility. These wheat fields had been harvested before the last week in June. Rainfall in the Central Facility area was quite high during the period from 20-30 June. This resulted in

significant growth of weeds and volunteer wheat. Conditions were drier in the western regions of the Little Washita sampling area where most of the wheat fields were located, resulting in very little volunteer wheat or weed growth in the standing stubble.

Water content as a percent of wet green biomass was a function of the age of plants and rainfall. This is supported by the large water content in the volunteer wheat and weeds sampled from the harvested wheat sites compared to the water content in the grass/pasture fields. Water content in the corn was also quite high compared to the grass/pasture fields.

Water content in the brown standing biomass was higher in the grass/pasture sites than in harvested wheat fields. This difference was due to the different exposures of the brown standing biomass to the atmosphere. Green standing biomass surrounded the brown standing biomass in the grass/pasture fields reducing the exposure of the brown standing biomass to the wind and sun. Brown standing biomass, the predominate form of vegetation in the harvested wheat fields, was not sheltered from the wind or sun. In addition, the evapotranspiration would be less in the harvested wheat fields, so the humidity would be lower in the canopy than in the grass/pasture field canopies.

This work was sponsored by a cooperative agreement with the United States Department of Agriculture-Agricultural Research Service. The views expressed in this report are those of the authors and do not necessarily reflect the views of the sponsor or the Illinois State Water Survey. The authors thank Harold Anthony, Troy Curry, Ahmed Fahsi, Jan Grothe, Jon Luman, Nate Luman, Lanita McGraw, Wichaune Porter, Andy Russ, Alan Ward, Valerie Williams for assisting with the collection of the vegetative data.

REFERENCES

Acock, M. C., C. S. T. Daughtry, G. Beinhart, E. Hirschmann, and B. Acock. 1994. Estimating leaf mass from light interception measurements on isolated plants of Erythroxylum species. Agronomy J. 86:570-574.

Asrar, G., R. B. Myneni, and E. T. Kanemasu. 1989. Estimation of plant-canopy attributes from spectral reflectance measurements. In Theory and Applications of Optical Remote Sensing (G. Asrar, ed.). Wiley, New York, pp. 252-296.

Carlson, T. N., and D. A. Ripley. 1997. On the relation between NDVI, fractional vegetation cover, and leaf area index. Remote Sens. Environ. 62:241-252.

Daughtry, C. S. T., K. P. Gallo, S. N. Goward, S. D. Prince, and W. P. Kustas. 1992. Spectral estimates of absorbed radiation and phytomass production in corn and soybean canopies. Remote Sensing Environ. 39:141-152.

Engman, E. T., and N. Chauhan, 1995. Status of microwave soil moisture measurements with remote sensing. Remote Sens. Environ. 51:189-198.

Jackson, T. J. 1997a. Soil moisture estimation using special satellite microwave/imager satellite data over a grassland region. Water Resour. Res. 33:1475-1484.

Jackson, T. J. 1997b. Southern Great Plains 1997 (SGP97) Hydrology Experiment Plan. http://hydrolab.arsusda.gov/sgp97 (verified 12 May 1999).

Jackson T. J., and T. J. Schmugge, 1991. Vegetation effects on the microwave emission of soils. Remote Sens. Environ. 36:203-212.

Jackson, T. J., T. J. Schmugge, and J. R. Wang. 1982. Passive microwave sensing of soil moisture under vegetation canopies. Water Resour. Res. 18:1137-1142.

Laflen, J. M. M. Ameniya, and E. A. Hintz. 1981. Measuring crop residue cover. J. Soil Water Conservation 36:341-343.

Morrison, J. E., Jr., J. Lemunyon, H. C. Bogusch, Jr., and C. S. T. Daughtry. 1993. Residue cover measurement techniques. J. Soil Water Conservation 48:470-483.

Price, J. C., and W. C. Bausch. 1995. Leaf area index estimation from visible and near-infrared reflective data. Remote Sens. Environ. 52:55-65.

Schmugge, T. J., J. R. Wang, and A Asrar. 1988. Results from the push broom microwave radiometer flights over the Konza Prairie in 1985. IEEE Trans. Geosci. Remote Sens. GE-26:590-596.

Schultz, G. A. 1988. Remote sensing in hydrology. J. of Hydrology 100:239-265.

Welles, J. M., and J. M. Norman. 1991. Instrument for indirect measurement of canopy architecture. Agronomy J. 83:818-825.

Wigneron, J. P, A. Chanzy, J. C. Calvet, and N. Bruguier. 1995. A simple algorithm to retrieve soil moisture and vegetation biomass using passive microwave measurements over crop fields. Remote Sens. Environ. 51:331-341.

Last update:Mon Oct 18 13:30:51 EDT 1999 Page Author: Hydrology Data Support Team -- hydrology-disc@listserv.gsfc.nasa.gov Web Curator: -- Stephen W Berrick NASA official: Steve Kempler, DAAC Manager -- Steven.J.Kempler@nasa.gov