| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dep. of Agron., 2004 Throckmorton Hall, Kansas State Univ., Manhattan, KS 66506
Corresponding author (bremer{at}ksu.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: BREB, Bowen ratio energy balance • DOY, day of year • ET, evapotranspiration • FIFE, First International Satellite Land Surface Climatology Project Field Experiment • G, heat flux into the soil • GR, grazed • H, sensible heat flux • Hc, sensible heat flux of canopy • Hs, sensible heat flux of soil • LAI, leaf area index • LE, latent heat flux • LEc, latent heat flux of canopy • LEs, latent heat flux of soil • PAR, photosynthetically active radiation • Rn, net radiation • Rn,c, net radiation of canopy • Rn,s, net radiation of soil • UGR, ungrazed • ß, Bowen ratio • ßc, ß of canopy • ßs, ß of soil component • 
v, volumetric soil water content
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
In grasslands, the most important climatic variable in determining ecosystem structure, function, and productivity is the balance between precipitation and ET (Stephenson, 1990; Burke et al., 1991; Frank and Inouye, 1994). However, a recent review of ET from grasslands and forests found that few canopy-scale ET data were available from grasslands (Kelliher et al., 1993). Even fewer ET data are available from grazed prairies, despite the historical importance and widespread presence of grazing around the globe (Stebbins, 1981; Axelrod, 1985; Coughenour, 1985; McNaughton, 1985).
In a grassland near the present study, during the First International Satellite Land Surface Climatology Project (ISLSCP) Field Experiment (FIFE), ET was found to be similar between grazed and ungrazed sites despite large differences in green LAI (Stewart and Verma, 1992). Other studies during FIFE, although not necessarily comparisons between grazed and ungrazed prairie per se, reported that intersite differences in ET showed no significant correlation with LAI (Shuttleworth et al., 1989). However, a recent study at the same location used for FIFE reported that greater LAI caused higher ET rates from a burned site compared with an unburned site (both ungrazed) (Bremer and Ham, 1999). In a grassland heavily grazed by prairie dogs, ET was lower than on an adjacent lightly grazed area (Day and Detling, 1994). In the latter study, ET became higher on the heavily grazed site than on the lightly grazed site as soil water became limiting, suggesting that lower ET early in the season had conserved soil water on the heavily grazed site.
Grazing, via removal of green leaf area, affects ET by several mechanisms (Fig. 1). Defoliation reduces the amount of irradiance intercepted by the canopy, thereby decreasing the amount of energy partitioned to leaves, and hence to transpiration (i.e., decreased transpiration). Conversely, irradiance at the soil surface increases after grazing, thereby raising soil temperature (Whitman, 1971; Bremer et al., 1998) and increasing evaporation of water from the soil. Therefore, grazing alters the partitioning of energy between the canopy and the soil, making less available for transpiration and more available for evaporation of water from the soil.
|
Leaf age may also become more important as the growing season progresses. Younger leaves in regrowth of grazed prairie may have improved leaf water status, lower stomatal and canopy resistances to water vapor transport, and consequently, higher transpiration per unit of leaf area (Coughenour, 1985; McNaughton, 1983, 1985; Archer and Detling, 1986; Svejcar and Christiansen, 1987; Wraith et al., 1987). The differences in the effects of leaf age on canopy-scale ET between grazed and ungrazed prairies would likely become more pronounced late in the growing season.
The effects of grazing on transpiration and evaporation of water from the soil can impact soil water content and seasonal water use. In some instances, grazing has conserved soil water content (Svejcar and Christiansen, 1987; Wraith et al., 1987; Naeth and Chanasyk, 1995), although at other times, grazing has had no effect (Coronato and Bertiller, 1996) and even caused reductions in soil water content (Whitman, 1971; Daddy et al., 1988; Naeth et al., 1991). Presumably, reductions in transpiration after grazing can result in the conservation of soil water (Parton and Risser, 1980; Day and Detling, 1994) although removal of vegetation can decrease infiltration, and consequently reduce soil water content in some cases (Bohn and Buckhouse, 1985; Abdel-Magid et al., 1987; Thurow et al., 1988).
Grazing can also alter other variables that affect ET but are not listed in Fig. 1 such as precipitation intercepted by the canopy and dew formation (Couturier and Ripley, 1973; Dunin and Reyenga, 1978; Branson, et al., 1981). However, in the interest of simplicity, we included only the mechanisms that likely have the greatest impacts on ET in grasslands over the course of the growing season.
The objectives of this study were to measure the surface energy balance and ET and to estimate cumulative ET during the growing season from adjacent areas of grazed and ungrazed tallgrass prairie. Data were collected during a 155-d period that began shortly after cattle were stocked on the grazed site and continued past the vegetational regrowth stage after the cattle were removed. Supporting measurements of green LAI, aboveground biomass, intercepted photosynthetically active radiation (PAR), soil temperature, soil water content, and reflected shortwave irradiance (albedo) were used to interpret seasonal trends in the energy balance.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
340 m above mean sea level), 5 km south of Manhattan, KS, USA. Vegetation at the site was tallgrass prairie dominated by the C4 grasses big bluestem (Andropogon gerardii Vitman), little bluestem (A. scoparius Michx.), and indiangrass [Sorghastrum nutans (L.) Nash]. Soils were silty clay loams (Benfield series: Fine, mixed, mesic Udic Argiustolls) with depths to shale and limestone fragments ranging from 0 (surface) to 1.0 m. The upper 30 cm of the soil profile had a range of bulk densities from 1.0 to 1.27 g cm-3, which were determined from volumetric samples (4.8 cm diam. x 5.0 cm). The average annual precipitation over 30 yr was 859 mm. Data were collected from two adjacent, expansive upland watersheds; one represented a GR treatment and the other represented an UGR treatment. Yearling steers (248 kg steer-1) were placed on the GR site on 8 May (DOY 128) and removed on 21 July (338 kg steer-1; DOY 202) using an intensive early stocking rate (0.81 ha steer-1; Smith and Owensby, 1978). The UGR site had been ungrazed since 1997. Before that, both sites had been grazed since the mid-1800s. Both sites had been burned annually for several decades. In the year of this study, both watersheds were burned on 19 April (DOY 109). One flux measurement tower was located on each watershed; the tower on the GR site was located approximately 450 m west of that on the UGR site.
Measurements of Energy Fluxes
The surface energy balance at each site was measured using the BREB method (Tanner, 1960), following the technique of Cellier and Olioso (1993). Differences in air temperature between 1.5 and 2.5 m above ground level were measured using shielded, fan-aspirated copper constantan thermocouples wired as a thermopile. Vapor pressure differences were determined at the same heights using a vacuum sampler described by Ham and Knapp (1998) and Bremer and Ham (1999). Air was sampled continuously from each height at 0.5 L min-1 through Excelon Bev-a-line IV tubing (Thermoplastic Processes, Stirling, NJ) with a 6.4-mm i.d. The inlet of each tube was located inside the fan-aspirated radiation shield used for air temperature measurement. Air samples were filtered (1 µm, Arco50, Gelman Sciences, Ann Arbor, MI) and routed to an environmental enclosure. To prevent condensation in the sample lines, tubing was enclosed in foam pipe insulation with a heat tape, which maintained air in the sample lines at a temperature above the dewpoint. Once inside the environmental enclosure, air samples traveled through 0.5-L ballast tanks to reduce high-frequency fluctuations in pressure and temperature. A relative humidity and temperature probe (HMP35A, Vaisala, Helsinki, Finland) attached to a sample chamber and connected to an automated valve system (Numatics, Highland, MI) was used to alternately measure the 1.5- and 2.5-m samples; measurements of vapor pressure differences were obtained every 3 min. Because humidity samples from each height were not sampled simultaneously, data were corrected for the time rate change in vapor pressure (Cellier and Olioso, 1993).
Net radiation (Rn) at each site was measured with a net radiometer (Q7, Radiation Energy Balance Systems, Seattle, WA) 2 m above the soil surface; the net radiometers were cross-calibrated immediately before the study. To protect the instruments from damage by cattle, an electric fence was placed around the masts on the GR site. However, the net radiometer on the GR site extended out beyond the fence and over a grazed area. The heat flux into the soil (G) was measured with the combination method (Kimball and Jackson, 1979). At each site, four soil heat flux plates (HFT-3, Radiation Energy Balance Systems, Seattle, WA) were positioned parallel to the surface at a depth of 0.10 m. Soil temperature and volumetric water content (v) of the soil at 0.05, 0.10, and 0.30 m were measured automatically with dual-probe heat capacity sensors (Campbell et al., 1991; Tarara and Ham, 1997). Measurements of soil temperature were logged every 60 s and then averaged and recorded every 30 min while v was estimated one to four times daily. Soil heat flux plates and dual-probe sensors were placed along a 6-m transect that was located within 20 m of the towers at both sites.
On the GR site, no significant changes in topography or vegetation were observed within 600 m of the tower in the direction of the prevailing winds (i.e., south-southwesterly). Although the topography was more broken on the UGR site, a minimum of 50 to 370 m of uniform relief extended from the tower in the direction of prevailing winds. As on the GR site, no significant changes in vegetation were observed within 600 m of the tower in direction of prevailing winds. The BREB method can be used successfully at fetch-to-height ratios as low as 20:1 when the Bowen ratio (ß) is small (Heilman et al., 1989), and fetch was sufficient on both the GR and UGR sites when winds were from the south and west.
Supporting Measurements and Data Acquisition
Air temperature and relative humidity were measured at 2.5 m in the upper aspirated radiation shield with a temperature and humidity probe (HMP35A, Vaisala, Helsinki, Finland). Wind speed and direction at each site were measured with a wind monitor (R.M. Young, Model 05103, Traverse City, MI) at a height of 3 m above ground level. Global irradiance was measured with a pyranometer (LI-200S, Li-Cor, Lincoln, NE). All data acquisition and control at each location were accomplished with two microloggers and accessories (CR10x, AM25T, AREL-12, Campbell Scientific, Logan, UT). All sensors were logged at 1 Hz; averages and the energy balance were computed and stored at 30-min intervals. Precipitation was measured with a rain gauge on the UGR site (Sierra-Misco model 2501, Nova Lynx Corp., Gross Valley, CA).
Green LAI and aboveground biomass were determined at 2-wk intervals from 18 May (DOY 138) to 30 Sept. (DOY 273) 1999. On each measurement date and at each site (i.e., GR and UGR), four 0.25-m2 areas were harvested within 100 m of the flux measurement towers. Green leaf area was measured with an area meter (LI-3100, Li-Cor, Lincoln, NE), and total aboveground biomass was determined gravimetrically after samples had been dried in a forced-air oven for 72 h at 55°C.
Intercepted PAR was measured at 1- to 4-wk intervals using a ceptometer lightbar (Sunfleck Ceptometer, Decagon Devices, Pullman, WA). On each measurement date, PAR was measured sequentially above the canopy and then below the canopy at the soil surface at six locations along a 100-m transect in both treatments (GR and UGR). Data were collected between 1000 and 1400 CST on clear days.
Albedo was measured using a pyranometer (8-48, Eppley Laboratories, Newport, RI) on the same dates and transects as intercepted PAR. Incoming and reflected irradiances were measured sequentially at each location along the transects with the same pyranometer; the height of the pyranometer was 1.2 m above ground level during measurements of reflected irradiance.
Energy Balance Definitions, Data Processing, and Computed Parameters
The energy balance for a thin layer at the surface is:
 | (1) |
 | (2) |
 | (3) |
The Bowen ratio (H/LE) was calculated in the traditional manner as:
 | (4) |
is the psychrometric constant and 
T and e are the air temperature and vapor pressure differentials, respectively, between 1.5 and 2.5 m above ground level. Available energy (Rn - G) is the amount of energy available at the surface either to evaporate water (LE) or to heat the air above the surface (H). The evaporative fraction is defined as the ratio of LE to available energy [LE/(Rn - G)]. Certain data from the BREB system were rejected based on criteria described by Ohmura (1982), including periods near sunset and sunrise or during rain. Also, an electric fence around the instrumentation on the GR site resulted in a small ungrazed area to the north and east of the tower, thus limiting fetch in that direction. The close proximity of environmental enclosures to the north and east of the sensors also may have affected air temperature and humidity gradients from those directions. Therefore, data were rejected when winds were from the north and east.
Daily and seasonal evaporation was calculated using data from the BREB system. Missing sections of data were reconstructed using the concept of self preservation of the evaporative fraction and ß as described by Crago and Brutsaert (1996). Those authors developed this concept from measurements obtained during FIFE at a site near the present study. This method estimates LE using the measured available energy at the surface (Rn - G) and the evaporative fraction from another nearby day with good ß data and similar v. This approach proved reliable in a study by Bremer and Ham (1999); a similar approach was used by Holscher et al. (1997) in eastern Amazonia. In the present study, ß was rejected on 99 entire days out of the 155-d period. Although this may sound excessive, note that in estimating ET, the evaporative fraction method did not model Rn - G, but rather modeled how Rn - G was partitioned between H and LE on certain days. We had confidence in the daily accuracy of Rn - G, which was measured everyday during the study, but elected to use ß data only when we had high confidence in the results. The evaporative fraction was typically highest after rain, decreasing steadily during the following days or weeks as the surface dried. Therefore, the evaporative fraction was similar from day to day given similar v and could be relied on to estimate LE on intermittent days when the ß data were rejected.
Cumulative seasonal evaporation was calculated from the measured daytime LE plus an estimated nighttime LE, which was assumed to be 10% of total daily evaporation (Brutsaert and Sugita, 1992). Daytime potential ET (when Rn was positive), or the maximal amount of ET given a wet surface, was calculated by the Penman equation (Penman, 1948) using data from the BREB system. Aerodynamic conductance was estimated by an empirical relationship suggested by Thom and Oliver (1977) that used wind speed at known height (e.g., 3 m in this study), zero-plane displacement, and roughness length (zM). The latter two were estimated from empirical equations based on canopy height:
 | (5) |
 | (6) |
A Note on Micrometeorological Tower Studies
One limitation of nearly all tower studies (e.g., Dugas and Mayeux, 1991; Stewart and Verma, 1992; Wofsy et al., 1993; Bremer and Ham, 1999) such as ours is the possibility that confounding site effects other than the designated treatment (e.g., grazing) may influence energy fluxes and ET. For example, variations in slope and aspect; soil depth, which affects water availability; and species composition all may impact ET (Nie et al., 1992; Stewart and Verma, 1992). After much evaluation, our two sites were chosen because they were adjacent, expansive upland watersheds with deep soils and similar slopes and aspects. Slight differences in species composition between treatments (e.g., more C3 forbs on the UGR site) were the results of grazing, and therefore represented a normal factor when comparing ET between grazed and ungrazed prairies.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
. On the same date, peak biomass was measured on both the UGR (0.637 kg m-2) and GR (0.238 kg m-2) sites. Late in the growing season, senescence caused green LAI to decline rapidly on the UGR site.
|
Early in the study, albedo was low following the spring burn that exposed the dark soil surfaces on both sites (DOY 130–145) (Fig. 2d). Albedo increased steadily on both sites during the following weeks as lighter-colored leaves emerged and began to cover the soil. However, albedo gradually became greater on the UGR site as differences in canopy sizes increased (Fig. 2a). This was similar to findings of other studies that reported higher albedo with increased vegetational cover (Ritchie, 1971; Rosset et al., 1997). Not surprisingly, the greatest differences in albedo occurred when differences in the canopies also were the greatest (UGR = 20.8% and GR = 15.3%; DOY 195). The higher amount of exposed soil surface on the GR site was probably the largest contributor to the lower albedo because optical properties of the soil are different than those of the canopy. The disparity in albedo between treatments illustrates that energy flows were affected by grazing and suggests that other variables in the energy balance (e.g., LE) also may be affected. By DOY 288, both canopies had senesced, and albedo was once again the same between the treatments.
Soil temperature also was affected by the removal of vegetation (Fig. 3). Daytime soil temperatures (avg. between 0800 and 1730 CST) were similar between the treatments early in the study but increased on the GR site as vegetation was removed by grazing. Maximum differences occurred near the end of and just after the grazing period. For example, between DOY 190 and 210, soil temperatures at 0.05 m averaged 3.8°C higher on the GR site (30.7°C) than on the UGR site (26.9°C). Thereafter, regrowth on the GR site began to moderate differences between treatments. However, soil temperature generally remained higher on the GR site throughout the remainder of the study.
|
|
 | (7) |
is the extinction coefficient of the vegetation (Massman, 1992). Assuming 
0.6 and using values of LAI from our study (e.g., LAI of 0.7 m2 m-2 on the GR site and 3.3 m2 m-2 on the UGR site), the ratios of Rn,s to Rn were 0.657 on the GR site compared with 0.138 on the UGR site. Differences in Rn,s and Rn,c between treatments had a noticeable impact on G (Fig. 4b). However, G was a small component of the overall energy budget, and differences in the total available energy (Rn - G) between sites were subtle (Fig. 5). Nevertheless, Rn - G was always larger on the UGR site, and the biggest differences occurred between DOY 195 and 220 when there was a fourfold difference in LAI between treatments. Over the entire study, available energy was 88.5 MJ m-2 more at the UGR site than at the GR site. This would be equivalent to 36 mm of ET.
|
|
v near the surface (Fig. 7a and 7b). Stewart and Verma (1992) also found no significant differences in ET between grazed and ungrazed prairie with large differences in LAI when soil water was nonlimiting. During our study, however, precipitation was below normal during July (DOY 182–212), and the upper part of the profile became dry (Fig. 7a and 7b). Thus, ßs increased as more energy was partitioned to Hs and less was partitioned to LEs. Because ßs was more dominant on the GR site, this increase in ßs caused the total ß to increase on the GR site (Fig. 6). On the UGR site, where ßc was more dominant, the larger canopy probably had access to water lower in the profile. This caused LEc to remain relatively high, which lowered ßc, and consequently lowered the total ß on the UGR site compared with the GR site during July.
|
Values of Rn, G, LE, and H from the GR and UGR sites during the study are reported in Table 2, which is provided for those who are interested in modeling ET based on the meteorological data given in Table 1.
|
|
Grazing conserved soil water in the 0- to 0.3-m profile throughout much of the study (Fig. 8b). This is consistent with other studies that reported higher soil water content under clipped or grazed prairie (Owensby et al., 1970; Buckhouse and Coltharp, 1976; Svejcar and Christiansen, 1987; Wraith et al., 1987). The largest differences in v occurred toward the end of and just after the grazing period when differences in the canopy size and ET also were maximal (Fig. 2a and 8a). For example, from DOY 190 to 215, v averaged 0.41 on the GR site compared with 0.32 on the UGR site. The trend reversed near the end of the study, and v became higher on the UGR site. This illustrates the impact of delayed senescence, higher ET, and possibly reduced infiltration on v in grazed prairie late in the growing season.
The v on the UGR site was depleted faster early in the growing season (DOY 150–160; Fig. 8b) when basically no differences in ET existed between treatments (Fig. 8a). The reduction in v during this period may have been specific to the small area where soil moisture sensors were placed and not representative of the much larger area of ET measurement. At the beginning of the study, heavy rain occurred immediately after installation of the soil moisture sensors on the GR site but before installation on the UGR site. The small trench on the GR site that was cut for installation of the sensors was filled with water. Thus, equilibration of the probes with the surrounding soil may have taken up to 2 wk during which the water in the trench artificially inflated v. Compounding this was a higher fraction of limestone fragments in the surface layer on the UGR site, which may have reduced the water-holding capacity in its 0- to 0.3-m profile. Thus, soil water in this profile may have been depleted faster on the UGR site given similar ET rates between treatments. Lower water-holding capacity in the 0- to 0.3-m profile also may explain why v was lower on the UGR site between DOY 235 and 275 when similar or even lower ET rates on the UGR site might have been expected to equilibrate or conserve v compared with the GR site.
Although reduced water-holding capacity may have contributed to lower v on the UGR site during July (DOY 182–212), the primary cause was likely higher ET. Evidence for this is that the magnitudes of the differences in both ET and v between treatments were much greater during July than at any other time in the study (Fig. 8b). The reduced water-holding capacity in the surface layer of the UGR site apparently did not limit canopy-scale ET either, probably because of deep soils. This was evident during July when ET remained higher on the UGR site despite the significant depletion of soil water in the 0- to 0.3-m profile compared with the GR site (Fig. 8a and 8b). Soil water was removed much faster from the lower profile of the UGR site than the GR site during July (e.g., 0.30 m; Fig. 7a and 7b), indicating that plants on the UGR site used water from lower depths as soil water was depleted in the surface layer (Anderson, 1965). Conversely, the decline in v at 0.30 m was slow on the GR site.
Cumulative Evapotranspiration
Cumulative estimates of ET during the 155-d study, which were adjusted for nighttime ET, were 526 mm on the UGR site and 494 mm on the GR site. Therefore, the removal of green LAI by grazing and its subsequent effects on energy partitioning at the surface caused seasonal ET to decline by 6.1%. Most of the decrease in ET occurred near the end of and just after the grazing period (i.e., July) when large differences in LAI between treatments and dry soil surfaces strongly impacted energy fluxes.
Year-to-year variations in climate may affect differences in cumulative ET between grazed and ungrazed prairies. In wet years, when soil water is nonlimiting, ET may be similar between treatments (Stewart and Verma, 1992), resulting in little net difference in cumulative ET. However, in years with periodic moisture stress, the reduction in ET on grazed prairie may result in substantial reductions in seasonal ET similar to the present study. In dry years, the reduction in ET under grazing may conserve soil water that would then be available later in the growing season (Day and Detling, 1994). The effects of grazing on ET also may have implications for the C balance of the prairie because water conservation under a grazed prairie, coupled with younger leaves late in the growing season, may extend the photosynthetically active period. This would be similar to the effects of elevated CO2 on this grassland where CO2-induced reductions in transpiration resulted in increased productivity in years with frequent water stress (Owensby et al., 1993, 1997, 1999). Therefore, the decrease in photosynthetic rates after defoliation in grazed prairie may be offset by improved water relations during dry periods. In a mixed-grass prairie, grazing affected photosynthetic rates and also increased soil C and N (LeCain et al., 2000; Schuman et al., 1999), suggesting that other factors in that grassland, such as ET, also may have been impacted by grazing.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
v was high from above-normal precipitation; hence, evaporation of water from the soil was increased. After the cattle were removed and significant regrowth had occurred on the GR site, ET converged between treatments despite remaining and significant differences in green LAI. Younger leaves in regrowth on the GR site likely had lower stomatal resistance, and thus higher transpiration per unit leaf area compared with their counterparts on the UGR site. Younger leaves on the GR site also senesced later, resulting in higher ET compared with the UGR site during the final month of the study (DOY 275–295). Cumulative ET during the study was reduced 6.1% by grazing. Lower ET under grazing may have implications for the hydrology of watersheds although further research is required for longer time periods and for measuring other components of the water balance besides ET (e.g., deep percolation, runoff, and soil water storage) to determine the effect of grazing on hydrology. In addition, climate change may increase aridity in some areas (Chaves and Pereira, 1992), accentuating the need to accurately predict ET in grasslands under various forms of land management such as grazing. |
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication March 17, 2000.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Su, D. J. Bremer, S. J. Keeley, and J. D. Fry Rooting Characteristics and Canopy Responses to Drought of Turfgrasses Including Hybrid Bluegrasses Agron. J., June 16, 2008; 100(4): 949 - 956. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Ingram, P. D. Stahl, G. E. Schuman, J. S. Buyer, G. F. Vance, G. K. Ganjegunte, J. M. Welker, and J. D. Derner Grazing Impacts on Soil Carbon and Microbial Communities in a Mixed-Grass Ecosystem Soil Sci. Soc. Am. J., May 29, 2008; 72(4): 939 - 948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Heitman, J. M. Basinger, G. J. Kluitenberg, J. M. Ham, J. M. Frank, and P. L. Barnes Field Evaluation of the Dual-Probe Heat-Pulse Method for Measuring Soil Water Content Vadose Zone J., November 1, 2003; 2(4): 552 - 560. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. Frank Evapotranspiration from Northern Semiarid Grasslands Agron. J., November 1, 2003; 95(6): 1504 - 1509. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Bremer Evaluation of Microlysimeters Used in Turfgrass Evapotranspiration Studies Using the Dual-Probe Heat-Pulse Technique Agron. J., November 1, 2003; 95(6): 1625 - 1632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
