Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
7 Erosion Control Practices: The Impact of Actual Versus Most Effective Use Paul E Rosenberry arid Burton C English For decades, the United States has both enjoyed and suffered from the ability of its agricultural sector to produce more than was demanded domestically or than could be sold abroad at a profit. Consumers benefited from this insofar as food and fiber prices were below those that would otherwise have prevailed. The agricultural sector, on the other hand, suffered from low prices relative to production costs and the resulting downward pressure on net farm income. Most farm programs have sought to control acreages in crop production, thus reducing the cropland base. Diminishing profits encour- aged farmers to intensify planting on acres still in production and to convert noncropland to cropland in order to qualify more acres for production control programs. Lower or nonexistent profits on noncropland uses further accelerated cropland conversion. The farmer's ability to increase yields of controlled crops has tended to offset the programs that aimed to reduce agricultural production. Individual farmers have responded well to the signals they have received. The government has encouraged maximum production on every acre that can qualify. As a result, farmers have been trying to produce their way out of a surplus situation. Acreage control programs have thus resulted in: . · intensive cropland use; increased conversion of noncropland to cropland; · increased misuse of natural resources; · increased capitalization of land and farm equipment; 204
205 ~ decreasing net incomes as prices of farm products have failed, even with price support programs, to keep pace with production costs; the expense of administering acreage control programs; and · the expense of storing surplus commodities. Since the early 1970s, increased foreign demand reduced agricultural product stockpiles and signaled farmers, through higher prices, to increase production. Financiers encouraged farmers to borrow capital. However, the ability of foreign agricultural producers to sell com- modities below the U.S. support price forced American exporters to be residual suppliers. Domestically, tightening supply-demand markets for capital (largely a result of increasing national debt and unfavorable balance of payments) caused interest rates to rise dramatically. The prospects for the rest of the 1980s suggest a continuing variable relationship between the demand for and the supply of agricultural commodities. World population continues to increase and inherent soil productivity continues to decline. Uncontrollable factors such as weather or foreign political instability are likely to keep for elan demand for U.S. farm com- modities unpredictable. A clear understanding of how SOilS are being protected or depleted is therefore essential. This paper explores the extent and severity of sheet and rill erosion, the extent to which conservation practices have reduced potential erosion, and the ability of conservation practices to resolve sheet and rill erosion problems. The national survey of lands used primarily for agricultural purposes by the Soil Conservation Service (SCS) is designed to yield statistically reliable estimates at the national, state, and Major Land Resource Area (MLRA) levels. Important parts of the National Resources Inventory (NRI) are the observations for each element in the Universal Soil Loss Equation (USLE) and the resulting estimates of the rate of sheet and rill erosion at each sample point in the survey. (For discussions of the elements of the equation--RKLSCP--see Heimlich and Bills, Pierce et al., and Runge et al., this volume.) A = RKLSCP
206 TABLE 1 Cropland Uses by Sheet and Rill Erosion Rate, United States, 1982(1, 000 Acres) USLE (Tons/Close-GrownOther Acre/Year)Row Crops Crops Hay CropsTotala 0.0--<5.0142,963.1 97,560.6 36,548.7 50,233.5327,306.2 5.0--<10.034,227.3 10,980.8 642.0 4,737.050,587.3 10.0--<15.011,048.3 3,403.5 180.1 1,608.016,240.2 15.0--<20.05,799.5 1,541.2 66.6 599.08,006.5 20.0--<25.03,409.7 722.1 25.2 339.24,496.5 25.0--<30.02,378.0 440.5 23.3 204.33,046.5 30.0--<35.01,527.7 296.8 7.5 123.11,955.3 35. 0--<40. 01 ,162. 7 215. 0 0. 5 81.01, 459.5 40.0--<50.01,470.8 177.7 5.1 87.11,741.0 50.0--<75.01,517.0 207.2 3.5 76.61,804.5 75.0--~100.0460.3 39.3 1.0 28.2529.1 100.0 & up319.5 29.5 0.0 22.6371.8 Total206,285.0 115,615.2 37,504.2 58,140.6417,545.3 Eros ion (million tons) 1,272.4 372.3 25.8 160.7 Average erosion t tons/acre/year ) 6.1 3.2 0.6 2.7 aFigures have been rounded off and do not total to an exact number. SOURCE: 19 8 2 NRI . 1,831.2 4.3 Generally, except for terracing, the first four factors in the equation reflect the natural potential of land to erode, while the last two factors reflect managerial decisions that determine how much potential erosion is actually realized. The exception is that terraces shorten slope lengths and, in certain cases, lower overall slope. Nevertheless, the assumption that RKLS represents the natural erosion potential of a soil is valid because terraces only occurred on 7 percent of the sample points in the 1982 NRI. Table 1 summarizes the sheet and rill erosion on crop- land documented in the 1982 NRI for the United States. These rates are annual averages of soil movement within a field or sample point. It should be noted that these data may differ from other published NRI results in that only privately owned land is used. Also, pastureland and
207 75 50 llJ Cal LL 25 lo acre | ton ...... .............. .............. .............. :::::::::::::: ....... ::::::: ....... , ...... ....... .............. :::::::::::::: ....... ::::::: .............. ....... ......... .............. .............. ....... ::::::: ............. ::::::: ....... ....... ....... :::::::::::::: ....... ::::::: ....... .............. ....... ................. ...... ............... :::::::::::::: ........ . . .... a.... ........ ........ ............... : :::: :::::::: ......... ........ . . ............... ,................ ,................ ................ . ,,, ............ ..2.2...',. . ~ 1 Row Crops Close Grown Hayland Other FIGURE 1 Cropland use and erosion in the United States in 1982. native pastureland with tillable cropland history are included in the other crops category. The NRI data indicate that 417.5 million acres are readily available for crop production in the United States. Soil movement from sheet and rill erosion is 4.3 tons/acre/year overall. Even land used for row crops has an erosion rate of only 6.1 tons/acre/year. Row crops are planted on almost half the cropland but account for almost 70 percent of sheet and rill erosion (see Figure 1). Although row cropland comprises almost 44 percent of U.S. soil loss at fewer than 5 tons/acre/year, it accounts for almost 87 percent of the loss at 75 to less than 100 tons/acre/year. The other cropland categories all have erosion rates below the average. Over 78 percent of U.S. cropland registers sheet and rill erosion of less than 5 tons/acre/year. For the whole country, land eroding at less than 5 tons/acre/year accounts for about one-third the soil loss in the 48 states. Lands eroding at the rate of 5 to 15
208 tons/acre/year account for another one-third. The remaining one-third of the lands have annual sheet and rill erosion rates greater than 15 tons/acre. They have one-third of total erosion, but represent only 6 percent of total cropland. If these 25 million acres could be taken out of cultivation, excess erosion could be decreased by almost one-third. It should be noted that these lands may be intermingled with other croplands, and they may be only small parts of fields. Such small tracts may seem insignificant to individual managers, and not farming them may be more costly in time or money. These lands would seem to be a prime target in a program to reduce U.S. erosion. FACTORS IN CROPLAND EROSION As Table 1 suggests, averages do not tell the whole story. Valuable insights can be gained, however, by reviewing the factors responsible for sheet and rill erosion on land used to grow row crops and small grains in 1982. The Impact of C and P Factors As indicated, the rainfall, erodibility, slope length, and slope gradient elements of the USLE can be considered as naturally occurring factors in the sheet and rill erosion process. The equation is designed so that the product of R. K, L, and S can be used, under certain assumptions, as an estimate of what sheet and rill erosion would be in the absence of the cover and management factor (C), the supporting practice factor (P), and the impact of the shorter slope length of terraces. The RKI,S product serves as an estimate of erosion in this case for two reasons: First, land is assumed to be tilled fallow with no plant or residue production. Second, tillage is assumed to occur up and down the field slopes. Tables 2 and 3 show cropland acreage and tons of erosion by RKLS grouping, indicating the distribution of natural potential for sheet and rill erosion as revealed by the NRI sample points. The average RKLS for all cropland is calculated to be about 21.8 tons/acre/year when the sample observations are weighted by the acreage
209 TABLE 2 Cropland Uses by RKLS Factor, United States, 1982 (Million Acres) RKLS ( Tons/ Close-Grown Other Acre/Year)Row CropsCropsHay CropsTotal 0.0--<5.044.838.012.0 19.4114.3 5.0--<10.056.728.06.5 13.5104.8 10.0--q5.031.616.63.0 6.758.0 15.0--<20.018.19.32.6 4.134.1 20.0--<25.010.86.11.5 2.420.8 25.0--<30.07.34.21.4 1.814.7 30.0--<35.05.42.91.0 1.310.5 35.0--<40.04.01.91.0 1.07.8 40.0--<50.06.02.61.4 1.511.5 50.0--<75.08.93.22.1 2.316.6 75.0--<OO.04.71.41.4 1.38.8 100.0 & up7.91.63.4 2.715.6 Total206.2115.837.3 58.0417.5 SOURCE: 1982 NRI . they represent, compared with the actual rate of 4.3 tons/acre/year (see Table 1). This suggests that if the 417.5 million acres used for cropland had been in fallow and had been tilled up and down the slopes, an average 21.8 tons/acre/year would be lost to sheet and rill erosion. The difference is due to the impact of current managerial decisions regarding cover and management practices and to shortened slope lengths of terraces. A comparison of the average tons per acre in Table 4 shows that row crops are being grown on nearly average cropland, a relationship explained in part by the fact that row crops comprise about 50 percent of cropland. Close-grown crops are planted on the least erodible land overall, while hay is grown on the most erodible (see Table 4). The removal of C and P factors changes total tons of erosion on hayland from 25.8 million (see Table 1) to 1.3 billion tons (see Table 3). Land in row crops would register the largest change in tonnage, with an increase from 1.2 billion (see Table 1) to 4.6 billion tons (see Table 3).
210 TABLE 3 Sheet and Rill Erosion if Cropland Were in Tilled Fallow ~ by RKLS Factor, United States, 1982 (Million Tons) RKLSRowClose~rown Other (Tons/Acre/Year)CropsCrops Hay CropsTotala 0. 0--<5140108 2652326 5. 0--C10.0415204 4899766 10.0--<15.0389204 3883713 15. 0--<20.0313161 4771592 20.0--<25.0241137 3554466 25. 0--<30. 0200114 4051404 30.0--<35.017593 3242342 35. 0--<40. 015070 3638294 40.0--<50.0 542 194131 142 1,008 50.0--<75.0 75. 0--<100 402 118129 109 758 100.0 & up 1,363 266698 621 2,949 Total 4,600 1,7841,323 1,428 9,135 Average RKLS 22.3 15.435.2 24.5 21.8 aTotals are rounded off. SOURCE: 19 8 2 NRI . Table 5 shows the distribution of cover and management conditions (the C factor) on land cultivated in 1982. The weighted average value of C reported in the NRI for cropland in 1982 was .26 (see Table 6). Thus, overall plant matter and residues reduced annual average sheet and rill erosion from the 21.8 tons/acre that would have prevailed under conditions of tilled fallow (see Table 2) to around 6 tons/acre. The C-factor value for all cropland could be expected to mirror that of row crops, given the dominance of the latter. However, a distinct difference can be seen when C-factor values greater than .45 tend to be row crops, and from there on the distributions are similar (see Figure 2). A comparison of C-factor values indicates that plant matter and residue are most effective in reducing erosion rates on hayland and least effective on land in row crops
211 TABLE 4 Impact of C and P Factors in USLE Estimates for Sheet and RillErosion on Cropland, United States, 1982 Inherent Erosion Actual Erosion Potential (RKLS) (RRLSCP) Reduction Percent Crops (Tons/Acre/Year ) (Tons/Acre/Year ) Factor Reduction l Row crops 22.3 6.1 3.7 72.6 Close- 15.4 3.2 4.8 79.2 grown crops Hay 35.2 0.6 58.7 98.2 Other 24.5 2.7 9.1 88.9 All crop- 21.8 4.3 5.1 80.2 land SOURCE: 1982 NRI. (see Tables 4 and 6). This is not surprising, but the overate magnitude of the effectiveness of plant and residue to lower erosion rates should be noted. For example, the overall impact of hay is to lower RKLS values of 125 tons per acre to 5 tons per acre. Another way to lower soil erosion is through supporting practices that lower the value of the P factor. Table 7 shows the distribution of major land uses by supporting practice groups on land cultivated in 1982. The P-factor values reflect the extent to which erosion is further reduced beyond that brought about by plant and residue conditions (the C factor). The weighted average value of P in the 1982 NRI for all cropland was .91 (see Table 6). Thus, the overall impact of supporting practices was to reduce sheet and rill erosion by 9 percent (1.0 - .91)--about 2 tons/acre. One reason for this small impact is that only 40 percent of cropland has a supporting practice that lowers P-factor values. A comparison of P-factor values across all land uses indicates that supporting practices do not vary by land uses. There are a few acres reported in the .45 to less than .70 range, reflecting contouring and stripcropping, but the dominant range is the .90 to 1.0 category (see Table 7). A comparison of weighted average C and P factors (see Figure 3) illustrates the overall impact of plant residue
212 OD U] UO a .,. p o - ~U C) Q U] a ~Q 3 C o ~4 C) o o ·~ .,, U] ·,' a ~: U~ w m o c C P4 C eq i.4 ·~1 C h P~ C o .,' U) i-~ ''' C U O .,. tq-I ~ 4: ~ ~a U] o Ll C) o ~r ~kD ~r~cor ~0~r0000 ................o co ~era'000 ~000~ ° Dr~u~0 ~00 ~u~ . ~.·.·.·· ~·~ O tn ~ ~ l- O · - er ~r ~ ~ ~ ~ c~ U. O ~ ~ O · · . · ~ ~ ~ O O O a) ~a,~r ~u~ava ~ . ~... ~...·.···.·. t-)1-{~a ~C~O ~o ~.-0000 U~ o . o o 0U~co ~ru ~u ~0~ .·.·.·.·. ~.·.·.·.~ co~r_d· ~I~In ~0000 C - C ~n I.~ '-I ,,¢ _ C o ·,' u) i.d · u £ 44 Ll o U ~o oo ~o~oooooooooooo ................o kD ~OOOOOOOOOOOO~ o o u~(D ~`~DOOOOOOOOOOOOU, .·.·.·.·······.·.~ _~r ~000000000000~ . 0 ~0 ~e:ce ~ro ~000 ...........·.....o o ~tD ~OOOO~ ~o CO ~0~COtD~D ~CO ~O ~OOO~ .·.·.·.·.·····.·.~ Or ~kD ~O0\0C~r ~0000 co ~0 ~cn ~r ~r ~, -00 ·.....·.·.·.·...o O ~O ~tDCO ~- 0~r ~0000 o icO~- ~~~\0U ~-I1~N ~OO~ .·.·.·.·.·.·.·.·.~ 0 ~c ~a) ~0 ~ut ~000tD c4 ~O ~4 ~ U~ ~- O U, O ~ O ~ O ~ O O O O O O 000_I_ ~) ~U ~coa~O ....·.·.·.·.·.~ VVVVVVVVVVVVVVVV ,8 IIIIIIIIIIIIIIII 1 ~o ~o ~o ~o 0 ° 0 0 _I ~ ~ ~ (~ (r~ ~ ~3' U~ kD O · . · . · . · . · . E~ C U X C o 4J o 4 o C o C ~4 W o C 3 C (U H Z 0` tn .,' ·e C) ~: o U)
213 TABLE 6 C and P Factor Weighted Averages by Land Use, United States, 1982 Land Use C Factor P Factor Row crops .28 .86 Close-grown crops .20 .87 Hay .04 .81 Other crops .19 .88 Total .26 .91 SOURCE: 1982 NRI. management over supporting practice on U.S. totals. The white bars representing supporting practices are all close to 1.0, thus having minimal impact on national totals. The black bars are much closer to zero, indicating greater ability to decrease erosion. Thus, from a national policy viewpoint, it would seem that land use changes reflecting plant and residue changes are more important than supporting practices. This point has received little attention because comparisons are often made between supporting practices given plant and residue cover. The Impact of Minimum Tillage, Contour Farming, Stripcropping, and Terracing There are about 167 million acres with some form of minimum tillage, contour, striparopping, and terrace systems (see Table 8). This amounts to 81 percent of row cropland and 40 percent of all cropland. Minimum tillage dominates conservation practice use and is concentrated on row-crop acres. Contour and terrace systems rank next, with about one-third as much acreage as minimum tillage. Some of the varying effectiveness of minimum tillage is shown in Table 9. The percentage distribution shows the value of the C factor across the complete range of possibilities, with a modal mean at the .15 to less than .20 category (see Figure 4). The acreage with the lowest
214 |~~\ Total cropland 10 _ i/ / Row crops \ o ,~,J 1 1 1 ~ 0 15 30 45 60 75 90 C FACTOR FIGURE 2 Distribution of row crops and total cropland by C factor groups in the United States, 1982. SOURCE: 1982 NRI. C factors reflects the cropland with high concentration of residue or sod-planted crops. The other extreme is not so easily explained. It would appear the residue and root structure of the vegetation present are not sig- nificantly different from tilled fallow. The main point is still clear: Namely, that the value of the C factor for minimum tillage systems can vary from extremely effective to only as effective as fall-plowed row crops. This variation is probably caused by the wide range of residue quantity that can be left on the soil surface and still qualify as minimum tillage. The current SCS definition of conservation tillage is to have 30 to 80 percent of the ground covered with residue after planting. This definition may not have been used in the survey. Also, it is not possible for all survey interviews to be conducted at planting time.
215 oo a~ tn U] .,, o 0 0 o C~ o o ,, ,, ~a ,' ~: E o o ~: U] o U] o C) P: 4~ a, o~ U] k] ·~ c) ~ Ll a, P. - o L, rl ,,$ - a C) o . - U] i. -, ,~c - a, v o .,' Ul a $,, -' ~ :~: .6- P4 o .,, U2 ~, .,~ ,,< _ L~ o JJ oooooo ~o~o ~C~o ··eee··eee ~·eeeeee 00000000000~r ~00o,0 ~O °°O°O° ~C~U-} ~CO[~~ ee ~eeeeeeeeeee·e 00000000000a ~o ~0 000000 ~u, ~\0er0r~a' ee ~eeeee·eeeeeeee OOOOOOOOOOO ~O ~OO a' OOOOOO ~ eeeeeeee OOOOOOOO d~tD ee ~D ~r o e o o u, ~tD~ a, e eeee OO0 O4 U, OOOOOON ~O_~_~)1- ~d~)_ ~· ~·e· ~e ~· ~·····ee OOOOOOOOOOO ~OOOOtD ~na OOOOOO_'_IOOOU-)NC~O1-I_1~) ·ee· ~eeee ~eeeee ~e OOOOOOOOOOOOOOOO OOOOOO ~O ~O ~O eee·~ OOOOOOo0000U, ~, -od' 00 000°0° ~0 ~0 ~r ~·eee·eee ~eeee 00000000000to ~U', -0r~ 0` 000000 ~c~0 ~0 ~r~\° ~ OOOOOOOOOOOU, ~OOO O°O°0°Ir ~-4C~-1kDI~-I(X}t~r] ·eeeeeeee ~eeeeee OOOOOOOOOO0- 0 ~OO °e o o ~D e U' e U' a, OO ee OO O ee (DH O~; CD ~ ~ 1 - 0 U~ O U~ O ~ O U, O O O O O O 0 ° 0 -I _1 ~ Od ~ ~) ~ ~ u~ ~ r- co a~ O ~· e · ~e V V V V V V V V V V V V V V V V t~ 1 1 1 1 1 1 1 1 1 1 111 1 111 _. 1 1 1 1 1 1 1 1 1 1 111 1 111 O C~ U~ ~ O ~ O ~ O U~ O ~ O O O O o O o O ~a ~a ~O ~· · · e e e E-t ·e C) 3 U)
216 1.0 0.9 0.8 0.7 LO ~ 0.6 _ ~ C ~ P 0.5 0.4 0.3 0.1 o Row Crop Close Grown Hay Other Total LAND USE FIGURE 3 Weighted average values of C and P factors by land use in the United States, 1982. SOURCE: 1982 NRI. Contour farming does not directly affect the C-factor value, but nevertheless has a similar distribution to minimum tillage (see Figure 4). The practice of contour stripcropping does affect the value of the C factor (through planting close-grown crops and meadow in rotation), but the modal mean is representative of a broader range, from .07 to less than .15. These values are significantly lower than minimum tillage, and are likely due to the impact of meadow and close-grown crops. Terrace systems do not directly affect the C-factor value. They have about the same modal mean and distribu- tion spread as minimum tillage and contour curves (see Figure 4). The distribution for minimum tillage, contour farming, stripcropping, and terraces by P factor is shown in Table 10. Minimum tillage has a small distribution around the typical P-factor values of .45 to less than .60 (see Figure 5). Over 84 percent of minimum tillage points
217 TABLE 8 Acreage and Distr ibution of Selected Conservation Practices in the United States, 1982 ConservationAcres Distribution Percent Use Practice (Million)(Percent)Row Crops All Cropland Minimum tillage 100.260.048.6 24.0 Contour system 34.920.916.9 8.4 Striperopping 3.42.01.6 0.8 Terrace system 28.517.113.8 6.8 Total 167.Oa100.080.9 40.0 aDouble counting is present in total. Each sample point could have up to three practices. Therefore, actual acreage is between 100.2 and 167.0 million acres. SOURCE: 1982 NRI. fall in the .90 to 1.0 category. This would indicate that minimum tillage does not tend to occur in conjunction with practices that affect P-factor values. Contouring and terracing, conversely, have similar distributions and tend to have a modal mean in the .45 to less than .60 categories (see Figure 5). Striparopping has a modal mean in the .20 to less than .30 categories, the lowest of all practices (see Table 10). Smaller peaks also occur in the .35 to less than .40, .45 to less than .50, and .90 to 1.00 categories, indicating that stripcropping does occur with contour and terraces. The record of P-factor values of .90 and above indicates that contours and striparopping occur on lands that have slopes or slope lengths that all but eliminate the impacts of the supporting practices. To some extent, these phenomena cannot be prevented due to the use of a variety of slopes and slope lengths in close proximity. Terraces were built on 28.5 million acres, about 7 percent of the cropland. The value of the P factor was in the range of .45 to less than .60 some 68 percent of the time. P-factor values of .90 to 1.00 occurred nearly 24 percent of the time on terraced acres (see Table 10). The distribution of this acreage by RKLS group is shown in Table 11. Terraces occur on all the RKLS groups studied, with a broad base from O to less than 75 tons/acre/year and some concentration at 5 to less than 25 tons/acre/year. Terraces appear to reduce slope length by about 100 feet in most RKLS groups (see Table 11) .
218 TABLE 9 Acreage and Distr ibution of Cropland Uses by C Factor, United States, 1982 c Factor Minimum Tillage Contour Farming Str ipcropping Terracing Acres Acres Acres Acres (Million) Percent (Million) Percent (Million) Percent (Million) Percent .00--<.021.51.46 0.61.78 0.26.83 0.62.17 .02--<.052.42.43 0.82.22 0.39.09 0.41.40 .05--<.071.71.73 0.61.62 0.38.25 0.20.76 .07--<.106.86.75 2.16.10 0.925.81 1.24.26 .10--<.1517.617.61 5.515.79 0.822.85 5.218.11 .15--<.2021.721.62 6.217.77 0.37.81 5.920.78 .20--<.2515.815.77 4.312.18 0.25.39 4.515.88 .25--<.3010.910.88 2.98.25 0.14.31 2.27.70 .30--<.3510.710.68 3.18.74 0.12.19 1.96.73 .35--<.405.45.38 3.39.58 0.13.96 2.37.90 .40--<.452.42.43 1.44.15 0.11.52 1.03.61 .45--<.501.71.67 1.54.33 0.00.84 0.82.73 .50--<.601.21.15 0.92.62 0.01.01 0.82.90 .60--<.700.40.37 1.64.71 0.00.15 1.44.88 .70--<.800.00.02 0.10.16 0.00.00 0.10.20 .80--<.900.00.02 0.00.00 0.00.00 0.00.00 .90--1.000.00.03 0.00.00 0.00.00 0.00.00 Total100.2100.00 34.9100.00 3.4100.01 28.5100.01 SOURCE: 1982 NRI. Row crops are planted on 41 percent of terraced land, compared with 50 percent on all cropland (see Table 12). The largest use of terraced land is for close-grown crops (45 percent, versus 28 percent of all cropland). Hay acreage accounts for 2 percent of terraced crouland, a drop from 9 percent of total cronland. "Other cropland" changed the least, with a drop from 14 percent of all land to 11 percent of terraced land. ~ ~. · · ~ · . ~ · ~ ~ The distribution of C values is not significantly Brent from that of minimum tillage and contour farming (see Table 9). For land in row crops, small grains, and hay, average RKLS values are higher for terraced cropland. For the other cropland category, however, the RKLS value is lower on terraced cropland (see Tables 12, 13, and 14). This indicates that for the major use categories, terraces were installed on the land with the highest potential for erosion. The relationship between erosion and slope lengths greater than 90 feet is shown in Table 15. About 60 percent of the acreage has slope lengths of 200 feet or
20 15 LL A ° 10 - ) \ 1 1 ~ I \ t 5 ~ _ i / ~ \ 1 ~ I:\ 219 __~ Minimum Tillage Contour Farming Terraced 15 30 45 60 75 90 C FACTOR FIGURE 4 Distribution of minimum tillage, contour farming, and terraced acreage by C factor in the United States, 1982. SOURCE: 1982 NRI. less. The next largest concentration is in the over- 350-feet category. The distribution of soil movement has the same pattern: On a per acre basis, the weighted average tons/acre generally increases along with slope length, until the 351-feet-and- larger category, where the C factor reduces soil movement more than slope length increases it. If all slope lengths were set at 90 feet, the total soil movement would be reduced by 2.5 billion tons, and 7 fewer tons/acre would be moved (see Table 15). This would require adding terraces to 343 million acres. If average terrace costs were $300/acre, the cost would be $103 billion--or $41/ton. At an average ton/acre reduction of 7 tons, the average cost would be $287/acre. Unfortunately, the tons/acre loss would still be two to five times above tolerable levels.
220 TABLE 10 Acreage and Distribution of Selected Conservation Practices by P Factor, United States, 1982 Minimum Tillage Contour Farming striPcropping Terracing P Acres Acres Acres Acres Factor (Million) Percent (Million) Percent (Million) Percent (Million) Percent .00--<.020.000.000.000.00 0.000.00 0.000.00 . 02--<. 050. 000. 000. 000. 00 0.000.00 0.000.00 .05--<.070.000.000.000.00 0.000.00 0.000.00 . 07--<. 100.000.000.000.00 0.000.00 0.000.00 .10--<~150.000.000.000.00 0.000.00 0.000.00 .15--<.200.000.000.000.00 0.000.00 0.000.00 .20--<.250.210.210.180.53 0.8525.13 0.080.28 .25--<.300.190.190.170.48 0.6820.18 0.050.19 .30--<.350.020.020,100.06 0.133.95 0.010.02 .35--<.400.090.090.100.28 0.3710.92 0.030.11 .40--<.450.040.040.030.08 0.185.16 0.010.04 .45--<.506.186.1617.4049.85 0.4011.74 11.7641.23 .50--<.604.884.8611.1932.04 0.247.16 7.5026.34 .60--<.702.982.980.441.27 0.051.48 1.906.83 .70--<.800.850.8S0.942.69 0.051.47 0.2S0.88 .80--<.900.180.180.3S1.01 0.041.20 0.030.10 .90--1.0084.S984.424.0911.71 0.3911.61 6.8423.98 Total100.21100.0034.99100.00 3.38100.00 28.46100.00 SOURCE: 1982 NRI. When slopes over 90 feet are sorted by slope (see Table 16), group A (a slope of O to less than 2 percent) accounts for 60 percent of the acres. Another 24 percent of the acreage can be found in slope group B (2 to less than 5 percent slope) and 10 percent is in slope group C (a slope of 5 to less than 9 percent). The distribution curve drops down to less than 1 percent for slope group G (25 percent slope or higher). Slope groups A to D (O to less than 14 percent slopes) account for 82 percent of the tonnage of soil movement. Within any one slope length, weighted averages of tons moved per acre rise sharply as slope increases. Within slope groups, a dichotomy exists when slope lengths are increased. Slope groups A and B have ranges from low to high that are within 3 tons/acre of their respective weighted averages (see Table 16). Slope group C has a range of 18
221 20 -`n 111 1 ~ A: a o J J - 5 i O 11\ (84.6%)` ---- Minimum Tillage | \ | - Contour Farming | \ l Terraced 15 30 45 60 75 90 P FACTOR FIGURE 5 Distribution of minimum tillage, contour farming, and terraced acreage by P factor in the United States, 1982. SOURCE: 1982 NRI. tons/acre, but the lowest and highest figures are still within 9 tons/acre of the weighted average. Slope groups higher than C have progressively larger ranges. The implications are that flat soils have a full range of slope lengths. Soils with very low RKLS (less than 5 tons/acre/year) may account for a significant proportion-- more than one-third of the acreage--of land with slope lengths greater than 350 feet.* The volatile combination *Many researchers are proposing using RKLST-1 as a statistic to assign fragile ranking to soils. Soils where RKLS values are less than 5 are ranked as nonfragile. One-third of the soils with less than 2 percent slope will have RKLS values of less than 5 tons/acre. If the soil loss tolerance limit is 4 or 5 tons/acre, then RKLSTr1 will be less than 1 and yet the soil may be experiencing twice the tolerable amount of soil movement. Testing the magnitude of the potential error is beyond this paper. The subject is being addressed in another report by the authors and will be available in the future.
222 TABLE 11 Amount and Slope Length of Terraced and Unterraced Cropland, by RKLS Factor, United States, 1982 RKLS (Tons/ Slope Length (Feet) Acre/Year) Terraced Unterraced Total Acres (Million) Terraced Unterraced 0.0--<5 192 340 338 1.4112.9 5.0--<10 208 303 297 6.698.2 10.0--<15 185 280 272 4.953.1 15.0--<20 175 273 262 4.030.1 20.0--<25 155 276 259 3.017.9 25.0--<30 146 264 247 2.212.6 30.0--<35 151 253 238 1.59.1 35.0--<40 143 241 229 0.96.9 40.0--<50 138 235 224 1.310.2 50.0--<75 148 237 229 1.515.1 75.0--<100 149 235 230 0.68.3 100.0 & up 188 234 232 0.714.8 Net average 175 295 287 SOURCE: 1982 NRI. of high slope length and high slopes can result in very high erosion rates. Fortunately, as noted earlier, cropland acreage is skewed to the lower slope groups, where soil movement can be more easily controlled. It would also appear that change in slope length and change in slope are indirectly proportional, i.e., when one goes up, the other one goes down and vice versa. Overall Comparison The interaction of supporting practices, land use, C and P factors, and RKLS groups is established by sorting the NRI cropland data into those points with one or all supporting practices (minimum tillage, contouring, strip- cropping, or terracing) and those points without any such practices. A comparison of C-factor values by land use, RKLS, and supporting practices shows that land in row and close- grown crops has slight decreases in C values as RKLS increases from 0 to 100 or more tons/acre/year (see Figure 6). The C-factor values for hayland are not
223 TABLE 12 Acreage and RKLS Factor on Terraced and Unterraced Land, by Land Use, United States, 1982 Terraced Land Unterraced Land Acres RKLS (Tons/ Acres RKLS (Tons/ Land 'Use (Million) Acre/Year) (Million) Acre/Year) Row crops 11. 8 3 0 . 6 19 4 . 5 21 . 7 Close-grown crops 12.8 19. 5 102.8 14.9 Hay 0.7 40.2 36.8 35.1 Other crops 3.2 20~8 55.0 24.7 Total 28.5 24.8 389.1 21.6 SOURCE: 1982 NRI. affected by rising RKLS values. Land in other crops is similar to close-grown crops for low RKLS values, but the C factor values are about half as high when the RKLS is high. Nationally, the impact of supporting practices on row crops causes the weighted average C-factor value to fall from 0.35 to 0.28 (see Figure 7). The two lines have almost parallel decreases, with the exception of the lowest RKLS values (less than 10 tons/acre/year). Overall, both the with and without trend lines decreased about .08 points from RKLS values of less than 5 tons/acre/year to those over 100 tons/acre/year. The change in weighted average C-factor values for close-grown crops is from .26 for without supporting Practices to .20 with them (see Fioure 8). This difference of .06 is maintained throughout the lower range of values. The upper range of RKLS values has a .07 difference. The trend lines are similar to those for row crops, except that here the C-factor values decrease faster with supporting practices than without them. The overall decreases were .056 with these practices and .042 without them. As with row crops, close-grown crops show more variability in the lowest RKLS groups. The change in weighted average C-factor values for hayland is from .04 with supporting practices to .03 without (see Figure 8). The with-supporting-practice trend line is almost flat while the without line declines from .03 to .02 as RKLS ranges from 0 to greater than 100 tons/acre/year.
224 TABLE 13 Sheet and Rill Erosion on Terraced Cropland by Land Use and RKLS Factor, United States, 1982 (Million Tons) RKLSRowClose-Grown Other (Tons/Acre/Year)CropsCropsHayCropsTotal 0--<52.61.80.10.75.3 5--<1016.124.20.58.849.6 10--q518.133.60.77.860.2 15--<2026.835.51.55.369.1 20--<2526.534.11.84.266.5 25--<3026.127.72.14.360.3 30--<3522.420.41.63.047.3 35--<4018.113.11.23.335.7 40--<5035.617.62.53.959.5 50--<7556.320.35.57.189.4 75--<10034.47.23.44.349.3 100 & up77.715.28.013.6114.5 Total360.7250.728.966.3706.7 TABLE 14 Sheet and Rill Erosion on Unterraced Cropland by Land Use and RKLS Factor, United States, 1982 (Million Tons) RKLSRowClose-Grown Other (Tons/Acre/Year)CropsCropsHayCropsTotal 0--<5137.3106.126.351.3321.1 5--<10398.6180.347.889.9716.6 10--q5370.5170.136.875.4652.8 15--<20286.2125.745.365.2522.5 20--~5214.3102.533.649.5399.9 25--<30173.886.238.046.2344.1 30--<35152.972.730.039.3294.8 35--<40132.156.834.734.9258.4 40--<50235.197.959.463.0455.4 50--<75485.3173.2125.3134.8918.7 75--<O0367.8110.6125.9104.7709.1 100 & up1,285.7251.2690.5607.52,834.9 Total4,239.61,533.31,293.61,361.78,428.3 SOURCE: 1982 NRI.
225 TABLE 15 Cropland, Current Soil Loss, and Soil Loss If Slope Lengths Reduced to 90 Feet, by Slope Length, United States, 1982 slow Length Acres Differ (Feet) (Million) Current Reduced ence Current Soil Loss (Billion Tons) Soil Loss (Tons/Acre) Dif fer Reduced ence 91--150111.7 2.4 2.1 0.3 22193 151--20062.0 1.7 1.2 0.5 27198 201--25023.6 0.8 0.5 0.3 342212 251--30041.3 1.1 0.7 0.4 281611 301--35010.0 0.4 0.2 0.2 351916 351 & up94.3 1.6 0.8 0.8 1899 Total342.9 8.0 Weighted average -- ~ - 23 16 7 SOURCE: 1 9 8 2 NRI . The change in weighted average C-factor values for other cropland is from .19 with supporting practices to .20 without them (see Figure 7). Except for the lowest RKLS values, the two trend lines are not significantly different. Like the other land uses, most of the variability and volatility was for land with an RKLS less than 20 tons/acre/year. A comparison of P-factor values by land uses, RKLS values, and supporting practices is much more varied (compare Figures 6 and 9), particularly in the lower RKLS groupings. It is not until RKLS increases to 40 tons/acre/year that separate trends develop. All land use trend lines for P when supporting practices are present decline more or less together through RKLS of 30 to 40 tons/acre/year. Land in hay and Bother crops" continue to exhibit similar trends. Row and close-grown crops also exhibit similar trends except for the RKLS range of 25 to 90 tons/acre/year. At these values, close-grown crops stop having lower P values than row crops and stabilize to have higher P values. The weighted average P-factor values for each land use are as follows: for land in row crops, .86 with supporting practices and 1.0 without; for close-grown crops, .87 with and .98 without; for hayland, .81 with and 1.0 without; and for land in other crops, .88 with supporting practices and .99 without them.
226 TABLE 16 Comparison of Ranges and Weighted Averages of Soil Loss Per Acre by Slope Group, United States, 198 2 Slope Group ( Percent of Slope) Lowest Tons/Acr e We ighted Highest Range Average A (0--<2)7 103 8 B (2--<5)20 255 22 C (5--<9)44 6218 53 D (9--<14)90 12838 106 E (14--q8)135 17641 151 F (18--<25)151 24594 194 G (25 or more)105 490385 391 SOURCE: 1 9 8 2 NRI . SUMMARY AND CONCLUS ION S For decades the U.S. agricultural sector has produced more than was demanded domestically or than could be sold abroad at a profit. As a result, consumers have benefited from low food and fiber prices while the agricultural sector has suffered from high production costs relative to low prices and declining net farm income. As farm programs have sought to control acreages and reduce the cropland base, farmers have turned to intensification and increased production as their only means to solve their situation. Because prospects for the rest of the 1980s suggest that supply will probably continue to exceed demand, such intensive cropland use will probably also continue with the accompanying effect of soil erosion. Using 1982 NRI data, this paper explores the extent and severity of sheet and rill erosion problems, the extent to which conservation practices have reduced potential erosion, and the ability of conservation Arc: tm rc~ land c!h^~ =^A -; 1 1 ~;~ ~_~1 ~ ~ ~ _~_ v ~ ~= ~ ~ ~ ~ ~ ~ ~ ~ V- ~ Ill CAL ~JLJ1~:~;:~ e The NRI data indicate that 417 e 5 million acres are readily available for crop production in the United States. Soil movement per acre from sheet and rill erosion is 4.3 tons/acre/year overall. Even cropland used for row-crop production has an erosion rate of 6.1
227 .30 ON .20 LL .10 o Row Crops Close Grown ' - - _ Other _. _~ Hay land 1 1 1 1 1 0 20 40 60 80 100 RKLS FIGURE 6 Comparison of C-factor values for selected land uses by RKLS groups in the United States, 1982. SOURCE: 1982 NRI. tons/acre/year. Row crops are planted on almost one-half of the cropland acreage but have almost 70 percent of sheet and rill erosion. Over 78 percent of all national cropland has sheet and rill erosion of less than 5 tons/acre/year, 90 percent of cropland has 10 tons/acre/year, and almost 95 percent of cropland has 15 tons/acre/year. Crop production occurs on almost 87 percent of soils. The natural potential for sheet and rill erosion is also revealed by the NRI sample. The natural potential for erosion of cropland is calculated to be about 21.8 tons/acre/year. This suggests that if the 417.5 million acres used for cropland had been in fallow and had been tilled up and down the slopes, the average annual erosion rate on these lands would have been about 21.8 tons/acre/ year. Because these factors were not present, erosion was much less severe and averaged only 4.3 tons/acre/year The difference is due to the impact of current managerial decisions regarding cover and management practices, which are reflected in the C factor of the soil loss equation; to supporting practices, which are reflected in the P factor; and to shortened slope lengths in terraces. .
228 .30 o 6 .20 .10 o Without With 0 20 40 60 80 100 RKLS FIGURE 7 Compar ison of C-factor values for land in row crops and other crops with and without supporting practices, by RKLS value, in the United States, 198 2 . SOURCE: 1982 NRI. A compar ison of the average tons per acre sheet and r ill eras ion for var ions croplands shows that row crops are grown on lands that are nearly average for all cropland . Close-grown crops are grown on the leas t erosive land, and hay is grown on the most erosive land. A compar ison of C-factor values across land uses indicates that plant matter and residue are the most effective in reducing erosion rates on hayland and the least ef fective on cropland. Th is is not surpr ising as the overall impact of hay is to lower the potential erosion value of 125 tons/acre/year to 5 tons/acre/year. About one-third of U.S. land has sheet and rill erosion rates greater than 15 tons/acre/year. These soils have one-third of total erosion but constitute only 6 percent of the total cropland. Thus, if these 25 million acres could be taken out of cultivation, excess erosion could be decreased by almost one-third. Another way to lower soil erosion is through supporting practices that lower the value of the P
229 30 cr o c: ~ 20 c: 10 o Without With `\ > Close Grown Other Cropland Hay Am_ 0 20 40 60 80 100 RKLS FIGURE 8 Comparison of C-factor values for land in close-grown crops, other cropland, and hay with and without supporting practices, by RKLS value, in the United States, 1982. SOURCE: 1982 NRI. factor. The weighted average value of the P factor reported in the 1982 NRI for all cropland was .91. Thus, the overall impact of supporting practices was to reduce sheet and rill erosion by 9 percent--a reduction of about 2 tons/acre. One reason for this small impact is that only 40 percent of cropland has a supporting practice that lowers the P-factor value. From a national policy viewpoint, it appears that land use changes reflecting plant and residue changes are more important than supporting practices. This point has received little attention because comparisons are often made between supporting practices, given plant and residue cover. Many other studies compare various systems of plant and residue management without considering the alternative of support practices. Other ways to lower erosion rates are contour farming, strip-cropping, and terracing. There are about 167 million acres in some form of minimum tillage, contour
230 .90 0 .80 A: .70 .60 _ , _ it' Close Grown 'it Row Crop ___ ~ ~ Other Hayland 0 20 40 60 RKLS 80 1 00 FIGURE 9 Comparison of P-factor values for selected land uses, by RKLS groups, United States, 1982. SOURCE: 1982 NRI. systems, striparopping, and terrace systems in the United States 1 including 81 percent of row crops and 40 percent of all cropland. Minimum tillage, the dominant conserva- tion practice used, is concentrated on 49 percent of row-crop acres. Contouring and terrace systems rank a distant second and third. For some cropland uses, terracing reduces erosion. However, for the major land uses, terraced land has a h igher potential for eras ion than unterraced land. Values that reflect the potential for sheet and rill erosion are higher for terraced cropland than unterraced when land is used for row crops, small grains, and hay. For other cropland uses, terraced cropland has a lower potential for eras ion . Erosion can also be reduced by changing slope lengths of cropland. The impact of adjusting slope lengths greater than 90 feet to 90 feet, which would happen with terracing, would be to reduce total tons of soil movement
231 by 2.5 billion tons and the tons/acre by 7 tons. This would require adding terraces to 343 million acres at an estimated average cost of $287/acre. Unfortunately, the tons/acre loss would still exceed tolerable levels by two to five times. An additional study is needed that extends the methods used in this paper to MLRAs, or to a selection of MLRAS across the United States. Such a study would reveal the regional impacts of erosion control practices and regional solutions to natural resource problems. A more detailed regional and MLRA study would have different results from this national study. Discussion Arnold R. Miller Rosenberry and English properly approach the soil erosion problem by recognizing that it is a physical process that occurs within the context of and is inextricably tied to economic forces. The engine of production that combines natural, human, capital, and technologic resources to produce agricultural outputs is the farmer. If our concern is for the optimum preservation of soil resources in a market economy, a basic fact that must be recognized and acted upon is that the value of soil in farm production is derived from the value of the commodities it is used to produce. In a market economy, every force that operates to limit the private and social value of farm outputs also operates to limit the value of soil and the volume of soil conservation that is privately and socially justifiable. Forces affecting the value of soil in farm production bear on the erosion process through their impact on the farmer. They do so by influencing the farmer's ability to convert production inputs of given value to outputs of greater value over successive cycles of production. Five basic forces interact to determine the value of unit of erosive soil in farm production at any point in time: (1) the price of farm outputs; (2) the change in output per unit of soil loss; (3) the interest rate; (4) the rate of change in the price of farm outputs; and (5) the rate of technologic advance as it applies to farm production. The value of erosive soil to the farmer operating in a market economy rises and falls with the value of the
232 commodities it is used to produce. This elementary relationship may not warrant discussion, but too often this relationship is ignored. The link between the value of farm outputs and the production value of soil was ignored for decades in the administration of federal conservation programs. Large portions of available funds and staff resources earmarked for conservation were diverted to production-oriented practices, especially drainage and irrigation. The cumulative effect of these diversions was to increase production on millions of acres. The increased production aggravated crop surpluses and increased downward pressure on farm prices and on the value of soil in farm production. As for crop yield response to soil loss, three general outcomes are possible: (1) yields may not change in response to soil loss, at least not at present soil depths; (2) yields may decline in response to soil loss; and (3) yields may rise in response to soil loss, such as when underlying soil material provides a better rooting medium than surface material. In situations where crop yields do not decline in response to soil loss, one has to question what is achieved in the way of preserving production capacity by expending resources to reduce erosion. Where yields decline in response to soil loss, one must distinguish between soil deterioration and soil depletion. Following Schickele (1937) and Bunce (1942), soil deterioration refers to the permanent impairment of the ability of soil to support plant growth. Soil depletion refers to the removal of plant nutrients and organic matter by any means when they can be replaced and fertility restored. Soil deterioration involves the permanent loss of pro- ductive capacity in the sense of consuming an exhaustible resource. By contrast, soil depletion involves the sacrifice of renewable productive capacity. The literature suggests that where crop yields decline in response to erosion, the range of decline is from 1 to 9 percent per acre inch of soil loss and the average decline is about 5 percent. Assuming corn yields of 120 bushels/acre and soil weight of 150 tons/acre inch, 1 ton of soil loss coverts to a yield reduction of 0.04 bushel annually. For lack of a better term, this value will be called the annual yield equivalent of a ton of soil loss. When the annual yield equivalent of a ton of soil loss is multiplied by crop price at the farm gate, an estimate of the annual decline in gross income resulting from ton of soil loss is obtained.
233 The annual decline in gross income due to erosion can be capitalized to estimate the value of erosive soil in agricultural production. The techniques used in the capitalization process are widely used in the private sector and elsewhere when the objective is to obtain maximum performance over time from available productive resources. The interest rate is a key item in the capitalization process. Experience suggests that parts of the conven- tional wisdom of soil conservation view the interest rate as a financial ogre feeding on the flesh of unborn generations by discounting future food supplies to present values. The emotional appeal of this view is strong. Its error, however, is equally strong and operates to reduce future resource endowments rather than enhance them. Few will argue that parents and grandparents do not serve their descendants best by seeking the highest annual yield on investments set aside for them. Although the relationship between current investment yield and future resource endowments is most commonly thought of in terms of financial assets, it holds equally well for physical assets, such as soil. When investment yields are reinvested, they compound over successive periods. Thus, by demanding high performance from investments set aside for the future, unborn generations stand to benefit from larger rather than smaller resource endowments. Although not always intuitively obvious, the arithmetic of the situation demonstrates that the effect of dis- counting is simply to reverse the effect of compounding. Thus, insofar as we accept low rates of return on investments in soil preservation, we not only retard their current performance but also reduce future resource endowments. A major impact of applying artificially low interest rates to soil conservation practices and programs, or any program, is to penalize future resource endowments in favor of poor current performance. I submit that insofar as the purpose of soil conservation is to ensure maximum availability of resources for future generations, conservation programs and practices should be planned and administered using interest rates at least as high private market rates. The annual rate of change in the price of farm outputs is also central to the valuation of erosive soil. Insofar as prices received by farmers can be expected to rise at a given rate, the return on conservation invest- ments will increase. The easiest way to integrate
234 expected price increases properly into the valuation process is to subtract the annual rate of price increase from the interest rate. Conversely, the easiest way to integrate expected price declines properly into the valuation process is to add the annualized rate of price decline to the interest rate. The relationship between the rate of change in farm prices and the interest rate is critically important to soil conversation in a market economy. Half a century ago, Harold Hotelling tl931) demonstrated that the value of exhaustible resources must rise at an annual rate equal to the interest rate if the privately and socially optimum rates of exhaustion are to coincide. Applying Hotelling's "law" to agriculture, if the price of farm outputs rises at annual rates equal to the interest rate, market forces will give farmers every incentive to conserve erosive soil to the extent socially justifiable. The sobering secular trend, however, is for inflation-adjusted farm prices to decline rather than to rise, and therein lies the primary cause of the erosion problem. A "cheap food" policy manifests itself as a "cheap resource" policy where erosive soil and farm production in a market economy are concerned. Fortunately, technical advance can have the same effect as rising farm prices in determining the agri- cultural production value of erosive soil. Where technical advance is "neutral" with respect to soil, the benefits of advancing technology are realized without expenditure of resources to "adapt" soil to successively higher levels of technology. . . . . . When expressed as an annual rate or change in output per acre, technical advance is properly integrated into the soil valuation process by subtracting the rate of advance from the interest rate. Conversely, technical decline is properly integrated into the valuation process by adding the rate of decline to the interest rate. Other things being equal, higher rates of technical advance translate to higher unit values of erosive soil. Hence, research to speed the rate of technical advance in agriculture can act to offset the exploitive impact of a "cheap food" policy on erosive soil resources. The determinants of the value of erosive soil in agricultural production interact to present the farmer figuratively with a basic question as to the effectiveness of erosion control practices. Are erosion control practices efficient enough to reduce soil loss at costs that are less than the capitalized value of the
235 soil assets they preserve? Casual empiricism suggest that under recent conditions, the capitalized value of erosive soil in agricultural production ranges from a few cents to a few dollars per ton. ~ ~ ~ Studies indicate that the cost of erosion control can range from almost nothing to many dollars per ton. These observations suggest that erosion control practices can be applied selectively in such manner as to be both "effective" as measured in tons/acre of erosion reduction and "efficient" in the sense of expending resources of lesser value than that of the soil they preserve. At risk of over-simplification, one can observe the history of federal erosion control programs as a series of phases. Without attempting to attach dates to phases, it is safe to characterize the earliest phase as having a tremendous zeal for the application of conservation practices coupled with Depression-era make-work programs. The result was the application of erosion control prac- tices almost without regard to the scarcity of resources other than soil itself. The second phase can be characterized as striving for effectiveness in controlling erosion. Unfortunately, effectiveness was defined in the absolute terms of the soil loss tolerance, or T value. This was an improvement over the first phase, however, in that the T value concept recognizes that there may be situations where the expenditure of scarce resources to reduce erosion may not be justifiable. Great progress was made during the second phase in terms of technical capability to quantify erosion problems and the impact of alternative erosion control practices on them both at the farm level and in terms of national policy. Too often, however, the drive to reduce erosion rates to absolute T values established without regard to nonsoil resources caused farmers to question whether the federal government really knew much about "scarce resources." It is time for a third phase in the erosion control effort. The third phase should strive for balance, in the sense of recognizing the scarcity of all types of resources, natural and otherwise. Balance means two things with respect to erosion control practices, effectiveness in reducing erosion and efficiency in both getting the job done and in determining how far to go.
236 REFERENCES Bunce, A. C. 1942. Economics of Soil Conservation. Ames: Iowa State. Hotelling, H. 1931. The economics of exhaustible resources. Journal of Political Economy 39:137-75. Schickele, R. 1937. Economics of Agricultural Land Use Adjustment. 1. Methodology in Soil Conservation and Agricultural Adjustment Research. Research Bulletin 209. Iowa Agricultural Experiment Station.
