Infiltration and Soil Erosion As Influenced by Vegetation and Soil in Northern Utah

Infiltration and Soil Erosion as Influenced by Vegetation and Soil in Northern Utah

Item Type Article; text

Authors Meeuwig, R. O.

Citation Meeuwig, R. O. (1970). Infiltration and soil erosion as influenced by vegetation and soil in northern Utah. Journal of Range Management, 23(3), 185-188.

DOI 10.2307/3896384

Publisher Society for Range Management

Journal Journal of Range Management

Download date 06/10/2021 05:53:47

Item License http://rightsstatements.org/vocab/InC/1.0/

Version Final published version

Link to Item http://hdl.handle.net/10150/649915 soil washed from each plot by rain- Infiltration and Soil Erosion as Influenced by drop splash and overland flow dur- Vegetation and Soil in Northern Utah’ ing the 30-minute period is considered an index of soil erodibility.

RICHARD 0. MEEUWIC? To obtain a wide difference in initial soil moisture, half (40) of Soil Physicist, Intermountain Forest and Range Experiment Station, the plots were pre-wet the day be- Forest Seroice, USDA, Ogden, Utah. fore infiltration tests were made by applying 0.5 inch of simulated rain Highlight diverse topography and soil char- during a 15-min period. Initial acteristics. For instance, soils that moisture content of the surface 2 The influences of vegetation, soil properties, and slope gradient on in- have high inherent erodibility need inches of soil was determined from filtration capacity and soil stability more protective cover than do soils two 240 cc soil samples taken just of high-elevation herbland on the that have low inherent erodibility. outside the boundaries of all the Wasatch Front in northern Utah were The quantitative effects of cer- plots a few minutes before the in- investigated under simulated rainfall conditions. Results emphasize the im- tain soil properties on infiltration filtration tests. portance of vegetation and litter cover and soil erosion have been pre- Protective cover in each plot was in maintaining infiltration capacity sented by Packer (1963) for the measured with a point analyzer and soil stability. Infiltration is also Gallatin elk winter range in south- (Levy and Madden, 1933). First affected significantly by soil properties, notably bulk density, aggregation, and western Montana, by Dortignac and strikes of 100 mechanically spaced moisture content. Love (1961) for ponderosa pine points were recorded as plants (by ranges of Colorado, and by Meeu- species), litter, stone, or bare soil. wig (1965) for a subalpine range in A day or two after the tests, all Control of overland flow and central Utah. Similar information vegetation on the plots was clipped soil erosion is a basic requirement was developed on the Davis County flush with the soil surface and for effective watershed protection. Experimental Watershed in north- all litter was removed. Litter and Such control is particularly impor- ern Utah during the summer of vegetation were air-dried for 2 weeks tant on herbaceous high-elevation 1962. and weighed. rangelands of Utah, where summer Two days after the application floods not only cause damage to the Study Area and Methods of simulated rain, when the soil range itself but also inflict severe The Davis County Experimental moisture had approached field ca- damage to cities and towns at the Watershed is situated on the Wa- pacity, two 120 cc cores were ob- mouths of flood-producing water- satch Front, midway between Salt tained from the 0- to l-inch and the sheds. Lake City and Ogden, Utah. Study l- to Z-inch depths of the soil man- The importance of vegetative plots are located on ungrazed, her- tle. Also, two 240 cc cores were ob- cover in maintaining soil stability baceous sites at an elevation of ap- tained from the Z- to 4.inch depth and permeability is well known. proximately 9,000 ft in the head- and one 240 cc core was obtained For example, on the Davis County waters of Farmington and Parrish from the 4- to 6.inch depth. Soil Experimental Watershed, Marston Creeks. Soils are derived from par- bulk density and 0.06 bar moisture (1952) found that amounts of sum- ent rock materials varying from content of each of these cores were mer storm overland flow and soil metamorphic gneisses and schists measured. Bulk soil samples were erosion, during 15 years of observa- to conglomerates, sandstones, and obtained from the surface inch tion, were slight on sites where veg- shales. of soil concurrently with the soil etation and litter covered 65% or The rainfall-simulating infiltrom- core samples. Texture of the soil more of the soil surface. Overland eter described by Dortignac (1951) was measured by the hydrometer flow and erosion from plots with was used to apply approximately method. The dichromate method less than this amount of cover were 2.5 inches of simulated rain to (Peech et al., 1947) was used to de- judged to be excessive. There is, each of eighty ZO- by 30.inch termine organic matter content. however, considerable uncertainty plots during a 30-min period. Re- Size distribution of soil particles about the amounts of vegetative sultant overland flow and soil ero- and aggregates was determined by cover needed to prevent excessive sion from each plot were measured. wet-sieving. overland flow and erosion under The difference between applied The above data were analyzed by rainfall and overland flow in area a series of multiple regressions to l Received April 21, 1969; accepted for inches closely approximates infil- determine which combinations of publication July 22, 1969. tration capacity, but includes plant the measured site variables were 2 Stationed at the Renewable Resources canopy storage and ground depres- most closely associated with infil- Center, University of Nevada, Reno, Nevada 89502. sion storage as well. The weight of tration and eroded soil. 185 186 MEEUWIG

-0.3 J I I I I 0 25 lnitiaf soil moiihe (pe$ent by $eht]

-0 5,- FIG. 3. Effects of changes in initial soil inches of water retained. :::t moisture on 20 30 40 50 60 70 66 90 100 Plant and litter cover [percent] 1.0 1.1 1.2 1.3 1.4 I.5 Soil bulk density (G/CC] FIG. 1. Inches of water retained during proportion of particles and aggre- FIG. 2. Changes in inches of water re- application of 2.5 inches of simulated gates larger than 0.5 mm, and soil tained due to variations of bulk density rain in 30 min in relation to plant and of the surface 4 inches of soil and per- moisture content are shown in Fig- litter cover and vegetation weight. centage of particles and aggregates ures 2 and 3 as deviations from the Initial soil moisture, soil bulk density, larger than 0.5 mm in the surface inch and soil particles and aggregates are values in Figure 1. Water retention of soil. held constant at their sample means. decreases with increasing bulk density. However, this relation is strongly affected by the coarseness Results 0.5 mm, B is bulk density of the surface 4 inches of soil in grams per of the soil as expressed by the pro- Infiltration cc, C is the proportion of the soil portion of particles and aggregates Plant and litter cover accounted surface protected from direct rain- larger than 0.5 mm in diameter for 73% of the variance in the drop impact by vegetation and (Fig. 2). When a large proportion amount of water retained by the litter, V is air-dry weight of live of the soil is composed of particles study plots during the 30-min simu- vegetation in tons per acre, and M and aggregates larger than 0.5 mm lated rainfall tests. No other single is the initial moisture content of the effect of bulk density is mini- measured variable was this highly the surface 2 inches of soil expressed mal, but when the soil is fine and correlated with retained water. in grams of water per gram of oven- poorly aggregated, water retention Four other site factors in combi- dry soil. This equation is expressed decreases sharply as bulk density in- nation with plant and litter cover graphically in Figures 1, 2, and 3. creases. One might expect soil bulk density accounted for 82% of the Shown in Figure 1 are the in- density and aggregate distribution variance in water retained. These fluences of plant and litter cover to be closely related, but such was factors are: bulk density of the sur- and of vegetation weight on amount not the case on this study area. The face 4 inches of soil, proportion of of water retained when soil bulk correlation between bulk density particles and aggregates larger than density, soil particles and aggregates and percentage of particles and ag- 0.5 mm in diameter in the surface larger than 0.5 mm, and initial gregates larger than 0.5 mm was inch of soil, initial moisture content moisture content are held constant not significant (r = -0.04). of the surface 2 inches of soil, and at their respective averages of Initial moisture content exerts air-dry weight of live vegetation. 1.22 g/cc, 50%, and 11%. These a small, but significant, effect on The multiple regression equation results are in general agreement amount of water retention (Fig. is: with findings of other studies; the 3). Retained water decreases about j, = 4.44 - 8.74A + 7.25A2 - 4.45B amount of water retained decreases 0.1 inch for each 5% increase in + 0.3 1 6B2 + 9.20AB-7.64A2B as cover and vegetation weights de- initial moisture content of the sur- + 3.03C - 1.46C2 + 0.362V - crease. However, the rate at which face 2 inches of soil. 0.192V2 + 0.419CV - 3.35M - water retention increases as vege- Inches of water retained by sites 4.55M2 tation weight increases is small during the 30-min simulated rain- in which p is the difference between when vegetation weight exceeds 1 fall can be estimated if values for applied rain and resultant runoff ton per acre. the five site factors are known. For in inches, A is the proportion of the The changes in water retention example, if 70% of the soil surface surface inch of soil composed of associated with variations from av- is protected by plants and litter, aggregates and particles larger than erage values of soil bulk density, and vegetation weight is 0.5 ton per INFILTRATION AND SOIL EROSION 187

equation upon which these curves soil surface protected from direct are based. raindrop impact by plants and lit- The standard error of estimate of ter, L is air-dry weight of litter in this equation is 0.343 inch of water. tons per acre, H is organic matter The relation between water reten- content (in grams) of the surface tion actually measured on the 80 inch of soil per gram of soil, and T plots and estimated by the regres- is plot slope gradient in percent. sion equation is shown in Figure The curves in Figures 5, 6, and 7 4. were calculated from this equation to show the relations graphically. Soil Erosion The effect of differences in cover The weights of soil eroded during percentage and litter weight (tons 0’0 0.5 1.5 2.0 2.5 3.0 the 30-min simulated rainfall tests 1.0 per acre) when organic matter con- Estimated water retained [inches] were related exponentially to in- tent of the surface inch of soil is 4. fluential site factors. Accordingly, FIG. Inches of water retained as esti- at its average of 5% and slope grad- mated by regression and as actually logarithms of eroded soil weights ient is at its average of 20% are measured. were used as the dependent variable shown in Figure 5. The rate at in regression analyses rather than which soil erosion increases as cover acre, 1.58 inches of water are re- actual tons per acre of eroded soil. decreases is reduced substantially As in the case of infiltration or retained without correction for bulk by increasing weights of litter. tained water, plant and litter cover density, aggregate distribution, or Organic matter content is the is the single site factor most closely initial moisture (Fig. 1). If 30% of most important soil factor influ- correlated with eroded soil (r = the surface soil consists of particles encing soil erosion. Variations of -0.87). Differences in cover explain and aggregates larger than 0.5 mm organic matter content from 5% 76% of the variance of the loga- and bulk density is 1.3 g per cc, the alter the relations shown in Figure rithm of eroded soil. depth of water retained is reduced 5. The amount of such alteration Three other site factors-litter by 0.29 inch (Fig. Z), changing the is shown by the curve of correction weight, slope gradient, and soil or- total to 1.29 inches. Finally, if ini- factors in Figure 6. For example, ganic matter-in combination with tial moisture content is 5%, the if organic matter content is 100/o, cover, account for 83% of the vari- depth of water retained is 0.15 inch the weights of soil eroded are only ance of the logarithm of eroded greater (Fig. 3), increasing total one-half of those indicated in Fig- soil. The multiple regression equa- water retention to 1.44 inches. This ure 5. tion is: is the same value that would be ob- Variations of slope gradient from 9 = 0.553 - 0.0621C - 1.91C? - tained by solving the regression 20% alter the plant cover-litter 0.341L + 0.194CL - 5.24H - weight-soil erosion relation shown 1.41H2 + 0.0183T in Figure 5. The magnitude of 47 in which j, is the common logarithm such alteration due to variations in of weight in tons per acre of soil eroded during the 30-min simulated rain test, C is the proportion of the 2.51

2.0

1.5

Z z z E 1.0 .3”

f s

0.5

FIG. 5. Weight of soil eroded during 0 I I 1 I I I 2 10 12 application of 2.5 inches of simulated Orgafiic mattir (percenBtJ ,L, 0 10 20 30 40 rain in 30 min as a function of plant Slope gradient [percent] and litter cover and air-dry weight of FIG. 6. Correction of estimated soil litter on 20% slopes with 5% organic eroded for variations in organic matter FIG. 7. Slope correction of estimated soil matter in the surface inch of soil. content of the surface inch of soil. eroded. 188 MEEUWIG slope gradient is shown by the 20‘- 0 proximating those of the simulated 0 curve of correction factors in Figure rainfall tests, They furnish a basis lo- 00 7. Values from the curves in Figure C- 0 for comparing the relative flood 5 should be multiplied by the car- ~ z - 0 0 oo0” and erosion potentials of different rection factors obtained from the 2 lo,- sites. They also provide a basis for curve in Figure 7 to correct for $_ OS determining the changes in influ- slope. For instance, at a slope gra- .E “f - ential site factors that must be dient of 2%, the weights of soil “, ,05_ achieved by management to insure 0 eroded are only one-half of the l “$ 00 that flood and erosion potentials Z oz- 0 0 co 0 0 0 0 considered to be “acceptable” are weights indicated by the curves in 2 Ol- 0 not exceeded. Figure 5. Similarly, at a slope gra- ,005- 0 Cl0 dient of 36%, the weights of soil 00 %oov0 0 n*= 0 83 ,002‘- Literature Cited eroded are twice those indicated in 0.0, 1 ’ 1 ’ ’ ’ ’ ’ ’ ’ 1 1 0.0,002 ,005 01 02 .05 0 I 0 2 0 5 1.0 2 5 10 DORTIGNAC, E. J. 1951. Design and Figure 5. Estimated soil erosion (tons/acre] operation of Rocky Mountain in- Weights of soil eroded under con- FIG. 8. Soil eroded as estimated by regres- filtrometer. U.S. Forest Serv., Rocky ditions similar to those of the simu- sion and as actually measured, plotted Mountain Forest and Range Exp. lated rainfall tests can be estimated logarithmically. Sta., Sta. Pap. 5, 68 p. from the curves in Figures 5, 6, DORTIGNAC, E. J., AND L. D. LOVE. and 7. If, for example, a site is relation between soil erosion ac- 1961. Infiltration studies on ponderosa pine ranges of Colorado. Rocky characterized by a plant and litter tually measured on the 80 plots Mountain Forest and Range Exp. cover of 60%, litter weight of 1 ton and estimated by the regression Sta., Sta. Pap. 59, 34 p. per acre, 2% organic matter in the equation is plotted logarithmically LEVY, E. B., ANDE. A. MADDEN. 1933. surface inch of soil, and slope gra- in Figure 8. The point method of pasture anal- dient of lo%, the expected weight ysis. New Zealand J. Agr. 46:267- Application of eroded soil is estimated as fol- 279. lows: a combination of 60% cover Results of this study do not pro- MARSTON,R. B. 1952. Ground cover and 1 ton per acre of litter, holding vide absolute values of flood and requirements for summer storm run- soil organic matter content and sedimentation potentials of a water- off control on aspen sites in northern Utah. J. Forest. 50:303-307. slope gradient constant at 5% and shed under natural rainfall condi- MEEUWIG, R. 0. 1965. Effects of 20%, respectively, yields 0.50 ton tions. However, they identify those seeding and grazing on infiltration per acre of eroded soil (Fig. 5). site factors that exert important in- capacity and soil stability of a sub- However, the correction factor for fluences on both detention of pre- alpine range in central Utah. J. 2% organic matter content is 1.46 cipitation and soil erodibility, de- Range Manage. 18: 173-l 80. (Fig. 6), and the correction factor fine these relations quantitatively, PACKER, P. E. 1963. Soil stability re- for 10% slope is 0.66 (Fig. 7). Ap- and indicate how these factors inter- quirements for the Gallatin elk win- plying these correction factors, the act. They provide quantitative cri- ter range. J. Wildlife Manage. 27: 401410. estimated weight of soil eroded is teria for estimating site-by-site in- PEECH, M., L. T. ALEXANDER, L. A. 0.50 x 1.46 x 0.66 = 0.48 ton per filtration and erosional behavior DEAN, ANDJ. F. REED. 1947. Meth- acre. of high-elevation herbaceous water- ods of soil analysis for soil-fertility The standard error of estimate of shed lands that are subjected to investit-zations. U.S. Dep. Agr. Circ. this equation is 0.455 log unit. The natural rainstorm conditions ap- 757, 2; p.

Summer Meeting

The Board of Directors and Advisory Council will hold their official mid- year meeting immediately prior to the AIBS Interdisciplinary Meeting, Univer- sity of Wyoming, Laramie, June 26-27, 1970.