2002-07-08_GENERAL DOCUMENTS - M2002004 (4)

Home Browse Search

Fig. 1-Field plot design. to 70% of the aboveground biomass, thus simulating a moderate level of grazing. Stubble height was kept as high as possible in an effort to trap snow and increase moisture availability the following growing season. During harvesting, samples of each grass species were col- lected at each plot and placed in a sealed container. In the laboratory, these samples were oven -dried at 70 °C, and the wet -to -dry factors thus obtained were used to convert the field data to a dry- weight basis. A gamma- neutron probe was used to measure moisture and bulk density at four locations in each plot. The moisture profile was ob- tained by measuring both moisture and density at 15 -, 30 -, 50 -, 70 -, 90-, and 110-cm depths. Moisture data were converted to percentage moisture by weight and the profile with the highest moisture content (usually the May profile) at a given location compared with the lowest moisture profile (usually the August profile) at that location. It was assumed that correlations existed between water depletion and the relative quantity of roots present; a number of studies have used this technique to indicate depth of rooting activity (Bohm, 1979; Power et al., 1981). Coring at selected plots appeared to support the relation between water depletion and rooting depth. Laboratory and Statistical Methods At the time of plot construction, samples of spoil and of soil were collected. Following sieving to remove fragments > 2 mm in diam- eter, standard methods were used to determine chemical and physical properties of the materials. Hydrogen ion activity was determined in a water- saturated paste using a glass electrode. Values of electrical con- ductivity (EC) and of sodium adsorption ratio (SAR) were determined using methods developed by the U.S. Salinity Laboratory Staff (1954). Iron and zinc were determined by the DTPA method (Lindsay & Nor- veil, 1978) and organic C was measured using the Walkley -Black pro- cedure described by Nelson and Sommers (1982). The hot water extract method was used to determine B (Bingham, 1982), the hydrometer method was used to determine texture (Day, 1965), and clay minerals were identified by X -ray diffraction (Whittig, 1965). The rate of water infiltration into spoil and soil (where soil depth was 152 cm) was de- termined using an infiltrometer as described by Haise et al. (1956). The response of perennial grass production to increasing soil depth was expressed on a relative basis as a percentage of the maximum production (or yield), thus eliminating the influence of plot location on the production response pattern and, when desired, eliminating the influence of precipitation and species traits on the response pattern. A replicate (n) was defined as production for a given species in a given plot for a particular year. The data were pooled in various ways, i.e., 400 J. Environ. Qual., Vol. 13, no. 3, 1984 by spoil traits, by years, by species, and by growing season precipita- tion. Standard statistical techniques, as described by Steel and Torrie (1960), were used throughout this study to calculate confidence inter- vals, to test hypotheses, and to calculate various statistical parameters. RESULTS AND DISCUSSION Influence of Spoil Type The response patterns of cool- season grass production to increasing soil depth were dependent on chemical and physical traits of the underlying spoil. Therefore, data were pooled on the basis of spoil traits, with four spoil types (generic, sodic, acid, and soil -like) being recognized. Generic Spoil Generic -type spoil lacked distinguishing traits (such as sodicity and acidity) and differed from soil in terms of origin, biological activity, and ability to support plant growth. Traits useful in describing this spoil type and the applied soil are presented in Table 1. Spoil pH was near neutral and had a narrow range from 6.6 to 7.6. This spoil type was commonly saline, but the highest salinity level was only 5.08 dS m '. Generic spoil was subjected to a comprehensive nutrient analysis and, with exception of N and P which were applied as fertilizers, generic spoil contained moderate levels of nutrients. Organic C in spoil ranged from 0.4 to 16.7 g kg but this material was dominated by coal fragments and should not be considered as biologically active plant residue. Spoil was generally classified as a clay loam, and kaolinite was the dominant clay mineral at most plots. Plots with generic spoil were found at 10 locations in the NGP: six plots were located in Wyom- ing, three plots in Montana, and one plot in North Dakota. Soil used in the plots was characterized as mildly alka- line, nonsaline, nonsodic, and loam to sandy loam in texture (Table 1). With exception of N and P, the soil nutrient status was not considered restrictive to plant growth. Infiltration, density, and plant establishment were other traits pertinent to evaluating the growth media. Water infiltration into soil was approximately three times greater than that into spoil; these values were mid- range for reclamation situations. Bulk density of soil Table 1- Selected traits of the four spoil types and of soil used in the plots. Trait pH EC, dS m - ' SAR Zn Fe, µg g B, g" Organic C, g kg ' Sand, g kg Clay, g kg' Kaolinite, g kg' of clay Infiltration, cm h ' Density, Mg Establishment, % Spoil type Generic Sodic Acid Soil -like Soil 7.2 8.2 4.0 7.4 7.2 3.91 2.94 2.57 2.81 1.85 3.6 28.4 2.9 0.9 1.4 2.6 2.4 3.0 0.8 0.9 28 21 61 13 9 0.9 0.9 1.7 2.0 0.7 3.4 3.0 10.2 0.7 1.9 28 28 47 22 48 32 28 36 26 21 41 30 38 18 27 2.4 0.4 5.5 3.0 8.9 1.58 1.65 1.52 1.57 1.62 18.3 5.3 17.4 46.7 41.5