Soil water depletion by C3 and C4 pasture grasses in central Appalachia.
Trees, however, have higher summer evapotranspiration (ET) rates because of greater radiation interception relative to grass (Baumgartner) and deeper soil water extraction. Pastures facilitate higher streamflow than forests in the fall and winter since subsurface reservoirs are less depleted and more quickly recharged by rainfall once seasonal ET decreases. Over a year, pastures generally yield more water to stream flow than hardwood forests, which in turn yield more than pine forests (Swank et al.).
In recent years there has been considerable interest in including [C.sub.4] species in pastures in Appalachia (D'Souza et al.). In addition to interest in [C.sub.4] grasses as pasture, there is developing interest in these grasses as a biomass source for ethanol production for use as liquid fuel (Cherney et al.). [C.sub.4] grasses usually begin growing after the last frost, which delays soil water use relative to [C.sub.3] grasses. Growth of [C.sub.4] grasses ceases at the first frost in the fall, unlike [C.sub.3] grasses, which continue to grow until cold weather becomes severe. Even within the overlapping growing season, there is evidence for differences in water use between [C.sub.3] and [C.sub.4] pastures (Stout; Stout et al.).
A thorough understanding of soil water depletion patterns of [C.sub.3] and [C.sub.4] grasses is needed to predict the hydrologic consequences of pasture composition. Therefore, this study was designed to evaluate differences in soil water depletion in Appalachia by one [C.sub.3] grass and two [C.sub.4] grasses at two moderately high elevation sites [920 m and 615 m (3,000 ft and 2,000 ft)].
Methods and materials
The experiment was conducted at two sites located about 27 km (16 mi) apart. One site was at an elevation of 920 m (3,000 ft) and located near Cool Ridge, West Virginia (37 [degrees] 39[minutes]N, 81 [degrees] 02[minutes]W) on a Gilpin loam (Typic Hapludult; fine loamy, mixed, mesic) and the other was at an elevation of 615 m (2,000 ft) and was located near Talcott, West Virginia (37 [degrees] 40[minutes]N, 80 [degrees] 48[minutes]W), on a Tilsit silt loam (Typic Fragiudult; fine silty, mixed, mesic). Both sites were nearly level ridgetop fields that had been cut for hay and received minimal inputs for many years preceding the study.
Three grass species were planted in the early summer of 1985 in 6 x 12 m (20 x 40 ft) randomized plots replicated four times at each site. The [C.sub.3] grass, orchard-grass (Dactylis glomerata L var 'Pennlate'), was grown to typify soil water use in regional pastures. The [C.sub.4] grasses were switchgrass (Panicum virgatum L. var 'Cave-in-rock'), which is a tall growing grass, and bermudagrass (Cynodon dactylon L. var 'Quicksand Common'), which grows in a short dense mat.
Plots were limed to achieve a surface soil pH of 6.0 to 6.5 and fertilized each year with 550 kg [ha.sup.-1] (490 lb [ac.sup.-1]) of 0-25-25 (0% N, 25% [P.sub.2][O.sub.5], 25% [K.sub.2]O) as recommended by the Pennsylvania State University Soil Testing Laboratory. Nitrogen was applied as ammonium nitrate at a rate of 55 kg [ha.sup.-1] (49 lb [ac.sup.-1]) in split applications, half at the onset of growth in the spring and half immediately after the [TABULAR DATA OMITTED] first harvest. This low nitrogen level was applied since little nitrogen is generally applied to hill land pastures in Appalachia and determining production potential was not an objective. The [C.sub.4] grass plots were sprayed each spring with 3.4 kg [ha.sup.-1] (3 lb [ac.sup.-1]) atrazine (2-chloro-4-ethylamine-6-isopropylamine-S-triazine) immediately prior to the beginning of growth to control [C.sub.3] weeds. Plots were managed to harvest a mature, fully headed hay crop. Dry matter yield was determined by taking three 1 [m.sup.2] (10 [ft.sup.2]) samples per plot, leaving a 5 cm (2 in) stubble height.
Two neutron probe access tubes were installed in each plot 2 m (6.5 ft) from the center of one end. Weekly soil water measurements were made at 20, 40, 60, and 80 cm (8, 16, 24, and 32 in) at the 615 m (2000 ft) elevation site and somewhat more frequently at the 920 m (3000 ft) elevation site. Starting at the 80 cm depth, the soil at both sites contained a substantial amount of parent material fragments.
Total soil water depletion was calculated by summing the 0-30 increment's change from field capacity, determined by the 20 cm (8 in) reading with subsequent 20 cm interval depletion amounts determined by their mid-point reading. Field capacity was defined as the content of the soils on day 130 of 1989. This gives moisture depletion to a 90 cm (35 in) depth. Lack of significance in soil water depletion differences between the three grass species was tested using Duncans Multiple Range Test.
Precipitation was measured with a Belfort weighing recording rain gauge, and air temperature and relative humidity were measured with a Belfort recording hygrothermograph.
At the conclusion of the study, four 5 cm (2 in) diameter soil cores were taken at each site with a Eijkelkanp closed ring holder at 5 cm increments to 70 cm (28 in). Samples were analyzed for particle size distribution, bulk density, and pH.
Results and discussion
The 615 m (2000 ft) site had an average growing season (May-September) daily maximum temperature that was 4.4 [degrees] C (7.9 [degrees] F) warmer than the 920 m (3000 ft) site over the 4 years of the study. However, the daily minimum temperatures only averaged 1.9 [degrees] C (3.4 [degrees] F) warmer over that period. During clear nights, under high pressure systems, the temperature was frequently cooler at the 615 m site. For this reason the growing season had the same duration at both sites each year because the sites were subjected to heavy frosts simultaneously.
Precipitation was similar at the two sites during all years except 1986. During that year precipitation was 28% lower than average at the 615 m site but near average at the 920 m site. During the 1987 and 1988 growing seasons, both sites were about 20% lower than average, but in 1989 both sites were a little above average.
The largest soil water deficits were measured around Day of Year (DOY) 220 during 1987 and DOY 190 during 1988. Maximum soil profile water deficits of about 18 cm (7 in) was measured at the 615 m site and 14 cm (6 in) at the 920 m site. Since the 615 m site was considerably warmer, this difference in soil water depletion is not surprising although soil properties may have also been a factor as will be discussed later.
Since soil water was measured under three grass species for a large number of dates, the following system was devised to illustrate when the soil water depleted by a grass species was not significantly different from other species using Duncan's Multiple Range Test.
It is evident from Figure 1 that there was no significant difference in soil water depletion level between species most often at the 615 m site. For the 4-year growing seasons, at the 615 m site, lines a, b, and c, respectively, extend over 77, 83, and 87% of the measurement periods. However, at the 920 m site the respective lines only extend over 35, 50, and 26% of the measurement periods.
During most of the periods when lines a and c were not present in the 920 m elevations plots of Figure 1, the depletion of soil water was in the range of 2 to 4 cm greater for orchardgrass than for the two [C.sub.4] grasses. These periods were late spring and late summer, which was when temperatures favored [C.sub.3] grass growth. The exception was in late summer of 1987 when depletion under bermudagrass was not different from depletion under orchardgrass.
There was no significant difference in soil water depletion between switchgrass and bermudagrass half of the time at the 920 m site. During 1986 and 1989, soil moisture levels were essentially identical throughout the measurement period. During the latter part of 1987, soil moisture levels were lower under the bermuda-grass, and during the early part of 1988, soil moisture levels were lower under the switchgrass.
At the 615 m site there were no significant differences between soil water levels under the three grasses most of the time nor any clear trends during the remaining periods.
Soil water levels at the four measurement depths were fairly constant during wet periods. During the driest period (DOY 212 of 1987), depletion of soil water decreased almost linearly with depth at the 615 m site (Tilsit silt loam). The 920 m site (Gilpin loam) showed a much smaller relative change in volumetric water content between the wet and dry periods at 40 cm and below.
An examination of soil properties revealed that clay content increased with depth at both sites with the 920 m site (Gilpin loam) having the greatest overall clay content. The 615 m site (Tilsit silt loam) had a higher bulk density with values reaching 1.7 at the 30 cm depth, while those at the 920 m site had a maximum of about 1.5. However, the pH at the 920 m site dropped precipitously to around 3.2 below the 45 cm depth. At the 615 m site, the pH averaged 4.6 at 45 cm and was 3.8 at 70 cm. The pattern of soil water extraction exhibited in Figure 2 suggests that extraction of water by all three grasses was more severely impacted by the low pH at the 920 m site than by the high bulk density at the 615 m site.
Yields were measured not to determine production potential but as an indication of relative growth at these sites under low intensity management. Orchardgrass yields were significantly higher at the cooler, 920 m site all but the last year, supporting the idea that the site had a climate more suitable for growth of [C.sub.3] grasses. Switchgrass yields were similar at both sites while bermudagrass yields were significantly higher at the warmer, 615 m site in 2 of the 4 years.
Soil water depletion differed between the two elevation sites. At the 920 m site (Gilpin loam), orchardgrass generally maintained soil water at lower levels than switchgrass or bermudagrass. This trend was not evident at the 615 m site (Tilsit silt loam). Orchardgrass growth was also greater at the higher, cooler site.
There was no constant pattern of differences in soil water utilization between switchgrass and bermudagrass even though there were substantial differences in canopy height between them. Switchgrass grew to be more than 1.5 m (4.9 ft) tall while the bermudagrass remained under 0.5 m (1.6 ft). Yields of the [C.sub.4] grasses were generally higher at the warmer 615 m site as expected although the difference was statistically significant in only 3 of the 8 crop years.
A greater degree of soil water extraction was achieved by the grasses at the 615 m site (Tilsit silt loam) than at the 920 m site (Gilpin loam). The greatest difference between sites came in 1987 and 1988, which were dry years and during which both sites had similar precipitation. The [TABULAR DATA OMITTED] difference was probably caused in part by higher evaporative demand at the warmer 615 m site. However, the distribution of soil water with depth during the driest period suggests water uptake was hindered more by the low soil pH at the 920 m site (Gilpin loam) than by the high bulk density at the 615 m site (Tilsit silt loam).
Table 3. Percent send, percent clay, bulk density, and pH of soil cores at depths up to 70 cm at the plot sites
Depth Sand Silt Clay Bulk Density pH (cm) (%) (%) (%) (gm cm-3)
920 m (Gilpin loam)
6 43 31 26 1.21 6.3 19 36 31 33 1.43 4.6 32 35 25 40 1.52 3.8 44 34 24 42 1.53 3.2 57 30 25 45 1.39 3.2 70 23 25 52 1.30 3.2
615 m (Tilsit silt loam)
6 67 25 8 1.25 6.6 19 61 26 13 1.60 5.9 32 57 22 21 1.71 5.6 44 54 22 24 1.70 4.6 57 59 15 26 1.70 4.4 70 60 16 24 1.64 3.8
[TABULAR DATA OMITTED]
It is well known that in hot climates such as the southern United States, [C.sub.4] grasses thrive and give high levels of forage production (Chamblee and Spooner). These data suggest that introducing [C.sub.4] grasses into medium elevation pastures within central Appalachia is unlikely to have any significant hydrologic impact. However, at high elevation sites, [C.sub.4] grasses may result in a decrease in soil water depletion, which will reduce the soil water reservoir size available for capturing storm precipitation. Therefore, higher water yield may be obtained from the higher elevation sites when [C.sub.4] grasses are grown rather than [C.sub.3] grasses.
Baumgartner, A. 1967. Energetic basis for differential vaporization from forest and agricultural lands. Int. Symp. For. Hydr. Proc. pp. 381-389.
Burt, T.P., and W.T. Swank. 1992. Flow frequency responses to hardwood-to-grass conversion and subsequent succession. Hydrology Processes 6:179-188.
Chamblee, D.S, and A.E. Spooner. 1973. Hay and pasture seedings for the humid South In: M.E. Heath, D.S. Metcalfe, and R.E. Burns (eds.) Forages: The Science of Grassland Agriculture. Iowa St. Univ. Press, Ames, Iowa, USA.
Cherney, J.H., K.D. Johnson, J.J. Volenec, E.J. Kladivko, and D.K. Greene. 1990. Evaluation of potential herbaceous biomass crops on marginal crop lands: 1) agronomic potential. Final Report 1985-1989. ORNL/Sub/85-21412/5&Pl. 34 pp.
D'Souza, G.E., R.F. Romero, and D.K. Smith. 1988. The investment potential of warm-season grasses for hill-land beef producers. Northeast Journal of Agricultural Economics 17:56-63.
Sharma, M.I., R.J.W. Barron, and D.R. Williamson. 1987. Soil water dynamics of lateritic catchments as affected by forest clearing for pasture, Journal of Hydrology 94:29-46.
Stout, W.L. 1992. Water-use efficiency of grasses as affected by soil, nitrogen, and temperature. Soil Science Society of America Journal 56:897-902.
Stout, W.L., G.A. Jung, J.A. Shaffer, and R. Estepp. 1986. Soil water conditions and yield of tall rescue, switch grass, and Caucasian bluestem in the Appalachian Northeast. Journal of Soil and Water Conservation 41(3):184-186.
Swank, W.T., L.W. Swift, Jr., and J.E. Douglass. 1988. Streamflow changes associated with forest cutting, species conversions, and natural disturbances. In: W.T. Swank and D.A. Crossley, Jr. (eds). Ecological Studies. Vol.66: Forest Hydrology and Ecology at Coweeta. Springer-Verlag, New York, NY.
C. M. Feldhake is a soil scientist and D. G. Boyer is a hydrologist with U.S. Department of Agriculture, Agricultural Research Service, Appalachian Soil and Water Conservation Research Laboratory, Beckley, West Virginia 25802-0867.
|Printer friendly Cite/link Email Feedback|
|Author:||Feldhake, C.M.; Boyer, D.G.|
|Publication:||Journal of Soil and Water Conservation|
|Date:||Jan 1, 1995|
|Previous Article:||Converting hydric cropland to wetland in Missouri: a geoeconomic analysis.|
|Next Article:||Terrace formation in cropping strips protected by tall wheatgrass barriers.|