Relative efficacy of two biophysical approaches to assess soil loss

Relative efficacy of two biophysical approaches to assess soil loss tolerance for Doon Valley soils of India D. Mandal, V.N. Sharda, and K.P. Tripathi

Abstract: One of the major factors in favor of soil conservation measures is the prevention of top fertile soil removal, which adversely affects the crop productivity, depending upon the type of crop, soil, and erosion intensity.This reasoning has generally been assigned qualitatively and has rarely been supported through a quantitative relationship between soil loss and crop productivity. The soil loss tolerance limit (T value) is one of the several indicators to properly explain this phenomenon. The T value can be used as a guide to decide the maximum soil loss that can be removed before the long-term soil productivity is adversely affected. In this paper, two methods have been compared in determining T value on a regional scale for Doon Valley conditions in India.The first approach is based on assessment of the productivity index (PI) that considers permissible soil productivity loss rate (δ) and planning horizon (H) for sustainable land use. Productivity index is assessed and then related with tolerable rate of soil loss. The second approach is based on a quantitative weighted additive model, which has been used to define the current state of the soil resource. Both methods have been found to be good indices of soil loss tolerance value. However, the PI-based approach requires a complicated depth-wise dataset, including available water capacity, bulk density, and pH, which at present is not available for most of the ecological regions of India. Generating such a database may require long time and large investment. On the other hand, the weighted additive model requires a minimum data set of six soil attributes, which are readily available. Using the sensitivity index, the different T value at each of the study sites was separately compared. The overall mean of the sensitivity index was statistically insignificant at p < 0.05 for each location. Both methods were able to provide a reliable estimate of T value at different locations. However, the weighted additive model proved to be more reasonable as it requires a readily available dataset. Key words: Doon valley—India environmental sustainability—erosion tolerance—productivity index (PI)—quantitative weighted additive model Soil erosion resulting mainly from agricultural land use is associated with environmental impacts (Clark II et al. 1985) and crop productivity loss (Lal 1995; Pimentel et al. 1995; Bakker et al. 2004). It is, therefore, imperative to properly understand the erosion tolerance of a given soil to ensure food (Daily et al. 1998) and environmental security (Matson et al. 1997). Soil conservation programs need to be justified in terms of productivity and environmental sustainability (Tiwari 1991; Ponzi 1993). Soil scientists assign soil loss tolerance limits (T value) to a soil based on the soil’s properties and the potential for the soil to lose produc-

404

jan/feb 2010—vol. 65, no. 1

tivity over time from erosion. Erosion depletes or eliminates root-explorable soil depth and crop-available water, selectively decreases the soil nutrients and organic matter content, and exposes soil layers having unsuitable characteristics for crop growth. Crop yield is thus a function of root growth, which in turn is affected by soil environment. Though loss of any amount of soil by erosion is generally not considered beneficial, years of field experience and scientific research indicate that some loss can be tolerated without affecting crop production significantly (Schertz 1983). This acceptable rate of erosion is known as Soil Loss Tolerance (SLT) (Wischmeier and

PROOF * NOT FOR DISTRIBUTION *

Smith 1978; McCormack et al. 1982; Soil Conservation Society of America 1982; Lal 1988; Beach and Gersmehl 1993; ISSS 1996) or permissible soil loss (Kok et al. 1995). In this study, two different approaches (models) have been adopted to determine soil loss tolerances. These modeling approaches involve estimation of SLT by relating erosion-induced change in soil quality. The first approach is a widely known index that relates total soil depth and soil quality with the productivity index (PI). The PI model has been widely used under diverse soils and agroecological regions. The basic assumption in the PI model is that crop yield is a function of root development, which in turn is governed by soil quality. It is, in fact, a depth-weighted soil quality index. The PI model can be used to characterize the soil erosion/productivity relationship for a specific site (Lal 1998; Liu et al. 2009). Soil loss tolerance could be based on explicit losses in productivity judged over a specific planning horizon. This has already been advocated by Delgado (2003) using a 10% allowable decline in yield over 100 years. The second approach is based on current functional state and structural integrity of a soil resource. A quantitative, weighted additive model was used to define the current state of soil resource. Soils were categorized as soil group 1, 2, or 3, depending upon their overall aggregated score. A two-way matrix, developed by USDA Natural Resources Conservation Service, was then followed with specific soil group and soil depth to determine the T value of a given soil. In both methods, two sets of indicators were used as minimum data set (MDS). The main objective of this study was to examine the relative variation in the two methods generally employed to determine Soil Loss Tolerance Limit (SLTL) for Doon Valley soils of India. The region experiences soil erosion on a moderate to severe scale (Dhruva Narayana and Rambabu 1983), and it may be as high as 53 Mg ha–1 y–1 (23.66 tn ac–1 yr–1) (Khola and Sastry 2005). In order to assess future risks of soil erosion, more precise and quantitative information

Debashis Mandal is a senior scientist, Vishwa Nath Sharda is the director, and Krishan Prakash Tripathi is a principal scientist at the Central Soil and Water Conservation Research and Training Institute, Dehradun, India.

journal of soil and water conservation

Figure 1 Study area with sampling locations in two identified soil series: Dhulkot and Bainkhala.

is needed on the soil attributes and soil loss tolerance limit.


UTTARAKHAND

PAKISTAN

CHINA

NE

PA BHUTAN L

Legend ARABIAN SEA

Materials and Methods The study was carried out in fine loamy Typic Eutrochrepts and loamy sand Dystric Udifluvent soils located in the Doon Valley region of India. The Doon Valley lies between the great Himalayas running on its northeastern side and the Shiwalik Range of hills running on its southern side. It is approximately situated between 77°35' and 78°19'E longitude and from 29°57'30" to 30°30'N latitude with an elevation ranging between 315 m (1,033.5 ft) and 2,500 m (8,202.1 ft) above mean sea level. It has a subtropical climate with average annual rainfall varying from 1,600 mm (63 in) (hills and piedmont plain) to 2,200 mm (98.42 in) (mountainous area) and a mean annual temperature of 67.3°F (19.61°C). The two most important rivers of north India, the Ganges and the Yamuna, demarcate Doon valley’s southeastern and northwestern boundaries, respectively. The “fragility” of the Doon Valley is accentuated by the presence of a major boundary fault passing through the northern part of the valley and the unusually heavy rainfall. The average width of the valley is about 20 km (12.42 mi), and the length is nearly 70 km (43.50 mi). Sampling locations are shown in figure 1. Representative sample sites at different locations were spatially separated by 40 to 50 m (131.23 to 164.04 ft) within the same physiographic unit with similar slope and aspect. There are two prominent soil series namely, Dhulkot and Bainkhala (Bhardwaj and Singh 1981). The Dhulkot soil series is derived from heavy textured, very deep alluvium, is yellowish brown to dark yellowish brown in colour with few gravel and coarse rock fragments and is taxonomically classified as Inceptisols. The texture of the Dhulkot soil series varied from silty loam to silty clay loam with increase in soil profile depth. On the other hand, the Bainkhala soil series has originated from the recent alluvium of stream bed material and is taxonomically classified as Entisols. The sedimentary layers of surface and subsurface soil differ in thickness and have variable texture with weak granular, very friable structure. Cobbles forming 15% to 60% are present throughout the soil profile. Two different models were used to compute SLTL. These modeling approaches

N

BAY OF BENGAL

Unidentified Dhulkot Bainkhala Sampling location

involve estimation of SLT by relating erosion-induced changes in soil quality or soil degradation. Productivity Index–based Model. A widely known index that combines the use of total soil depth and soil quality is the productivity index (PI).The SLT values were estimated by computing PI. The PI model is an algorithm based on the assumption that crop yield is a function of root development, which in turn is governed by the soil environment. The PI is calculated by using the following multifunctional model (Delgado and Lopez 1998): n PI = ∑ (Ai Bi Ci Ki) . i = 1

(1)

All these parameters (Ai, Bi, Ci, and Ki ) are evaluated separately for each horizon of the soil up to a depth of 100 cm (39.4 in) and on a scale of 0 and 1, with 1 corresponding to the condition of the parameters that is most favorable to the radical growth of the crop.


Each parameter is calculated from the following equations: Parameter Ai is plant available water holding capacity (water retained in a section between 33 to 1,500 kPa [4.78 to 217.55 psi]), Ai = 0.05W if Ai = 1.00 if

0 ≤ W ≤ 20 W > 20 ,

where Ai is the value of the parameter “available water capacity” of the soil and W is available water (%) of the soil on gravimetric basis. Parameter Bi is bulk density of the soil as related to textural type. For medium texture (loamy, coarse silty loamy), Bi = 1.87 – 0.67BD for 1.30 ≤ BD ≤ 1.55 Bi = 6.00 – 3.33BD for 1.55 < BD ≤ 1.80, where Bi is the value of the parameter “bulk density” and BD = bulk density of the soil (Mg m–3).


405

Table 1 Categorical ranking of soil attributes used to convert soil properties into 0 to 1 scale. Soil attribute

Range /score

Categorical ranking 1 2

3

4

5

Model

Infiltration Range 5.0 Scored using “more is better” curve (cm h–1) (Model 2)*. Score 0.2 0.3 0.5 0.8 1.0 Bulk density Range 1.63 Scored using “less is better” curve (Mg m–3) (Model 1)†. Score 1.0 0.8 0.5 0.3 0.2 Water stable Range 150 10.0 12.5 Note: Q = state of soil and was found using equation 3.

Soil units were grouped into three soil groups: viz I (Q < 0.33), II (Q = 0.33 to 0.66) and III (Q > 0.66) based on the aggregated state of the soil score (Q) as obtained in equation 3. The Q indicates that a soil site falling in soil group III performs all soil functions in an optimum manner to resist erosive forces and thus may allow higher erosion rates than other soil groups. A general guide developed at the Iowa State University Statistical Laboratory (USDA NRCS 1999) was used to arrive at the soil loss tolerance values for each soil unit (table 2) based on the soil group of each soil mapping unit and the soil depth. In both the methods, two sets of indicators were used as minimum data sets (MDS). Appropriate scoring methods were used to

Group III (Q > 0.66) 5.0 7.5 10.0 10.0 12.5

transform indicators into unitless scores. A sensitivity index was calculated using the concept of a relative comparison for each of the methods (Gregorich et al. 1994; Biederbeek et al. 1998). The sensitivity index (SI) for a given site was calculated as SI = (T)PI / (T)WA ,

(4)

where (T)PI represents the T-value index based on the PI approach, and (T)WA is the T value index based on the weighted-additive approach. A parametric paired t-test (Steel and Torrie 1980; Montgomery 1984) was used to determine the overall significance of the sensitivity index for a given T value (H0: [T]PI ÷ [T]C = 1.0). Using the sensitivity index, the different T-values at each of the

Table 3 Soil productivity determinants and productivity index (PI) of different study sites. Location

Depth (cm)

Water holding capacity (%)

Bulk density (Mg m–3)

pH

Depth-wise PI

Dhulkot cropland Overall PI

0 to 20 20 to 40 40 to 70 70 to 100

10.4 to 12.4 (11.4 ± 0.35) 11.5 to 13.8 (12.6 ± 0.46) 12.5 to 14.2 (13.4 ± 0.27) 13.6 to 15.5 (14.6 ± 0.34)

1.37 to 1.44 (1.40 ± 0.013) 1.40 to 1.46 (1.43 ± 0.010) 1.46 to 1.52 (1.49 ± 0.010) 1.50 to 1.62 (1.56 ± 0.013)

5.8 to 6.1 (5.92 ± 0.06) 5.8 to 6.2 (6.0 ± 0.09) 5.9 to 6.2 (6.10 ± 0.07) 5.9 to 6.3 (6.06 ± 0.07)

0.13 0.15 0.21 0.21 0.70

Dhulkot agroforestry Overall PI

0 to 20 20 to 40 40 to 70 70 to 100

10.6 to 13.5 (12.0 ± 0.50) 11.8 to 14.1 (13.0 ± 0.44) 13.4 to 15.5 (14.2 ± 0.36) 14.6 to 15.5 (15.2 ± 0.25)

1.35 to 1.40 (1.37 ± 0.010) 1.40 to 1.47 (1.43 ± 0.010) 1.46 to 1.54 (1.50 ± 0.014) 1.50 to 1.62 (1.56 ± 0.023)

5.6 to 6.2 (5.88 ± 0.10) 6.2 to 6.6 (6.42 ± 0.07) 5.8 to 6.4 (6.10 ± 0.10) 5.9 to 6.3 (6.10 ± 0.07)

0.14 0.15 0.22 0.22 0.73

Bainkhala-I Overall PI

0 to 20 20 to 40 40 to 50

5.4 to 6.3 (5.8 ± 0.16) 5.4 to 7.2 (6.2 ± 0.30) 5.5 to 7.6 (6.4 ± 0.40)

1.38 to 1.42 (1.40 ± 0.006) 1.39 to 1.46 (1.43 ± 0.013) 1.46 to 1.56 (1.50 ± 0.020)

6.4 to 6.7 (6.54 ± 0.05) 6.1 to 6.4 (6.20 ± 0.05) 6.0 to 6.3 (6.14 ± 0.05)

0.073 0.074 0.040 0.187 (=0.19)

Bainkhala-II Overall PI

0 to 20 20 to 40 40 to 60

5.3 to 6.8 (5.8 ± 0.20) 5.4 to 7.2 (6.2 ± 0.31) 5.5 to 7.6 (6.4 ± 0.41)

1.36 to 1.42 (1.40 ± 0.011) 1.39 to 1.46 (1.43 ± 0.010) 1.44 to 1.5 7(1.50 ± 0.024)

6.3 to 6.8 (6.54 ± 0.08) 6.0 to 6.4 (6.18 ± 0.07) 6.0 to 6.4 (6.14 ± 0.07)

0.073 0.074 0.072 0.219 (=0.22)




407

Table 4

Productivity index (PI) and soil loss tolerances for productivity permissible loss rate δ (δ = 0.1) and planning horizon H (H = 100 years). Soil removal Vulnerability Location (cm) PI equation PIf (δ = 0.1)

Soil loss tolerance (Mg ha–1 y–1)

Dhulkot cropland

0 4 8 12 16

0.70 0.67 0.65 0.63 0.60

y = 0.69 – 0.006x

0.63

10.00

14.0

Dhulkot agroforestry

0 4 8 12 16

0.73 0.70 0.68 0.65 0.62

y = 0.73 – 0.007x

0.657

10.42

14.4

Bainkhala-I

0 4 8 12 16

0.19 0.17 0.16 0.15 0.13

y = 0.19 – 0.004x

0.171

4.75

6.6

0.22 0.20 0.19 0.18 0.16

y = 0.22 – 0.004x

0.198

5.50

7.7

Bainkhala-II 0 4 8 12 16 Notes: y = PI values. x = soil removal (cm).

study sites were separately compared to their conventional counterparts and thus can be considered as paired observations. Results and Discussion Description of Minimum Datasets for Productivity Index–based Approach. The soil parameters required under this approach are (1) water holding capacity, (2) bulk density, and (3) pH. The depth-wise soil information on water holding capacity, bulk density, and pH for the study sites is presented in table 3. Soil depth is found to be medium up to 50 cm (19.68 in) and 60 cm (23.62 in) for Bainkhala I and Bainkhala II series, respectively. Water holding capacity is relatively lower in Bainkhala sites (5.3% to 7.6%) than Dhulkot sites (10.4% to 15.5%). Average water holding capacities were 11.4%, 12.6%, 13.4%, and 14.6% in different layers of the Dhulkot cropland site and were 12.0%, 13.0%, 14.2%, and 15.2% in the Dhulkot agroforestry site at 0 to 20, 20 to 40, 40 to 70, and 70 to 100 cm (0 to 7.87, 7.87 to 15.74, 15.75 to 27.56, and 27.56 to 39.37 in) soil depths, respectively. Bainkhala I and

408

Equivalent soil depth (cm) corresponding to PIf


II sites had relatively lower water holding capacity at 5.8%, 6.2%, and 6.4%, at 0 to 20, 20 to 40 and 40 to 50/40 to 60 cm (0 to 7.87, 7.87 to 15.74, and 15.75 to 19.68/ 15.75 to 23.62 in) soil depths, respectively. Generally, the moisture content of the soils was very low, and the reason may be low clay and organic matter content. Such a soil may have little inherent fertility and will not be able to retain high moisture content and plant nutrients to a large extent. Bulk density increased with depth, probably due to decreased organic matter content. Bulk density varied between 1.35 Mg m–3 and 1.62 Mg m–3 with mean values of 1.37 Mg m–3 to 1.56 Mg m–3. With increasing soil depth, from surface to lower horizons, the bulk density and water holding capacity increased. The bulk density did not differ significantly at the 0.05 probability level among the sites in different soil layers. The pH varied between 5.6 and 6.8 for the entire region.The soil pH of the Dhulkot sites was significantly lower than those of the Bainkhala sites, particularly for the surface soils. This indicates that crop cultivation and


soil management have caused soil acidification. The productivity index of 0.70 and 0.73 was recorded for Dhulkot cropland and Dhulkot agroforestry sites, respectively, while Bainkhala I and Bainkhala II registered the lowest PI of 0.19 and 0.22, respectively (table 3). The results indicate that the initial productivity of Dhulkot soils are very high and that as soil is removed, productivity decreases even though it remains between high and very high levels. The higher PI for Dhulkot soils is attributed to favorable soil properties such as water holding capacity, pH, bulk density, and deeper rooting depth as compared to Bainkhala soils. Assigning T Values Based on the Productivity Index–based Approach. The soil-loss tolerance values based on the PIbased approach are presented in table 4 for all four selected sites of the Doon Valley region. The corresponding changes in PI with removal of soils indicated that the PI value decreased with the removal of soils. According to Delgado (2003), a PI value of more than 0.50 is considered high. An initial productivity index of 0.70 and 0.73 has


Table 5 Original soil characteristics, aggregated scores (Q), soil groups, and tolerance limits of different study sites. Soil characteristics

Dhulkot cropland

Dhulkot agroforestry

Bainkhala I

Bainkhala II

Soil depth (cm) 150 120 40 60 Infiltration rate (cm h–1) 0.9 to 1.5 (1.12 ± 0.24)a 0.91 to 1.73 (1.23 ± 0.36)a 10.0 to 12.6 (11.0 ± 1.0)b 10.5 to 13.4 (11.3 ± 1.1)b Bulk density (Mg m–3) 1.37 to 1.44 (1.40 ± 0.013)a 1.35 to 1.40 (1.37 ± 0.010)a 1.35 to 1.42 (1.40 ± 0.011)a 1.36 to 1.42 (1.40 ± 0.011)a Soil aggregates WSA (%) 66 to 78 (71.2 ± 4.60)a 66 to 82 (72.4 ± 6.06)a 44.0 to 58 (48.2 ± 6.56)b 46 to 62 (49.4 ± 7.55)b Organic carbon (%) 0.82 to 1.0 (0.88 ± 0.07)a 0.55 to 0.60 (0.58 ± 0.02)b 0.42 to 0.50 (0.45 ± 0.02)c 0.45 to 0.52 (0.47 ± 0.02)c Total N (%) 0.10 to 0.14 (0.11 ± 0.016)a 0.06 to 0.08 (0.07± 0.003)b 0.06 to 0.07 (0.06 ± 0.001)b 0.06 to 0.07 (0.06 ± 0.001)b Available P kg ha–1) 52 to 56 (53.2 ± 1.7)a 25 to 36 (31 ± 4.0)b 12 to 18 (14.4 ± 2.6)c 14 to 19 (15.4 ± 2.5)c Available K (kg ha–1) 290 to 296 (292 ± 2.4)a 220 to 228 (223.6 ± 3.6)b 72 to 96 (86.4 ± 10.0)c 78 to 96 (87.2 ± 9.8)c Aggregated score (Q) 0.63 0.58 0.70 0.70 Soil group II II III III T value (Mg ha–1 y–1) 12.5 12.5 7.5 10.0 Notes: WSA = water stable aggregates. N = nitrogen. P = phosphorus. K = potassium. Values in parenthesis are mean ± standard error. Different letters within the same row are significantly different at p < 0.05.

been computed for Dhulkot cropland and Dhulkot agroforestry sites, respectively. With the removal of 16 cm (6.28 in) of surface soil at the Dhulkot sites, the PI dropped to 0.60 (cropland) and 0.62 (agroforestry), a loss of 14.3% and 15.0%, respectively. Hence, soils in this group would continue to produce at high levels for extended periods of time. It has been observed that if 16 cm of surface soil was removed, the PI would drop from 0.19 to 0.13 and from 0.22 to 0.16, a loss of 31.6% and 27.3%, for Bainkhala I and Bainkhala II sites, respectively. The PIf values for all the soils were calculated by assuming δ = 0.1. The result showed that PIf values ranged from 0.17 to 0.66. Among the four selected sites, Dhulkot soils had higher PIf values than Bainkhala soils, with a highest value of 0.66 at the Dhulkot agroforestry soil. The higher values at Dhulkot soils are due to their higher initial PI values and favorable subsurface properties. The equivalent soil depth and soil loss tolerance limits were calculated based on corresponding PIf values. According to this estimation, a higher loss of equivalent soil depth ranging from 10.0 to 10.42 cm (3.93 to 4.10 in) and 4.75 cm to 5.50 cm (1.87 to 2.16 in) was obtained for Dhulkot and Bainkhala soils, respectively. After converting these soil depths, tolerance limits of 14.0 Mg ha–1 y–1 (6.25 tn ac–1 yr–1) and 14.4 Mg ha–1 y–1 (6.42 tn ac–1 yr–1) were obtained for Dhulkot cropland and Dhulkot agroforestry sites, respectively. For Bainkhala I and Bainkhala II sites, the values were 6.6 Mg ha–1 y–1 (2.94 tn ac–1 yr–1) and 7.7 Mg ha–1 y–1 (3.43 tn ac–1 yr–1), respectively. Many researchers used this approach to estimate SLT on the basis of the vulnerability equation. Results in the present study were in


conformity with those of Delgado (2003) and Lobo et al. (2005). Description of Minimum Dataset for Weighted Additive Model. The minimum dataset of soil, required for this model, was decided based on primary functions of soil with regard to resistance to water erosion. These functions include (1) infiltration rate, (2) bulk density, (3) organic carbon, (4) water stable aggregates, and (5) fertility parameters. The ranges of original soil properties of the representative sites are presented in table 5. Infiltration values of three sites ranged between 0.9 and 12.6 cm h–1 (0.35 and 4.96 in hr–1). Among the sites, Dhulkot cropland had significantly lower infiltration values ranging from 0.9 to 1.5 cm h–1 (0.35 to 0.59 in hr–1), thus strongly suggesting that runoff from this site is maximum. The highest values of infiltration (10.0 to 12.6 cm h–1 [3.93 to 4.96 in hr–1]) of the Bainkhala series was due to its skeletal textural nature. Bulk densities ranged between 1.35 and 1.44 Mg m–3. Contents of water stable aggregates ranged between 44% and 82%. Dhulkot agroforestry had the significantly largest water stable aggregate–content (66% to 82%) due to higher organic carbon content. Content of organic carbon ranged from 0.45% to 1.0%. Low organic carbon content was generally observed in riverbed deposition of Bainkhala series. The Dhulkot cropland site contained more total nitrogen (0.1% to 0.14%) than the other sites because of its higher organic carbon status. The Dhulkot cropland site also contained significantly higher nutrients in terms of available phosphorus (P) and potassium (K) than other sites due to frequent application of inorganic fertilizers.


Soil Grouping and Assigning T Values Based on Weighted Additive Model. Soil properties were converted into standard scores using appropriate scoring curves presented in table 1. Measure of state of soil (Q) was calculated quantitatively by aggregating the score for each soil parameter obtained by multiplying the rating with respective weightage (equation 3). According to this model, sum of weighted indicators determined the level of resistance to erosion offered by a particular soil. T values were then assigned to each soil on the basis of a T matrix (table 2). Out of the three representative locations, the soil of lowest aggregated score in terms of erosion resistance was found at the Dhulkot agroforestry site. Bainkhala sites had the highest aggregated score (Q) of 0.70 followed by Dhulkot cropland (0.63). There were significant differences among the observed mean values of organic carbon, total nitrogen, available P, and available K (table 5). The study showed that infiltration rate is reflected by fundamental edaphic characteristics and that it is the most sensitive parameter for such soil assessment. None of the soils have aggregated Q values of less than 0.33, indicating that the entire region qualifies for soil group II and III. Mandal et al. (2006) has explained variations in soil loss tolerance values for different sites in the western Himalayas, including the study region. Soils under Dhulkot series are capable of affording higher soil loss tolerance (12.5 Mg ha–1 y–1) (5.5 tn ac–1 yr–1) due to their better resiliency. For favorable soil, higher T values have also been reported by Renard et al. (1997), Ringo (1999), and Lakaria et al. (2008). Beach and Gersmehl (1993) assumed higher tolerance limits for soils with greater depth, stone-free topsoil material, and/or not


409

Table 6 Sensitivity index for values at different sites of Doon Valley.

previously eroded than for soils that are shallow or previously eroded. Sensitivity Index for Tolerance Values. At all the locations, the T values based on the PI-based approach were higher than the values estimated by the weighted additive approach. A sensitivity index (TPI ÷ TWA) was calculated using the concept of a relative comparison for each of the two methods employed in this study. According to this index, the maximum variability of 1.12 to 1.15 was observed in Dhulkot soils (table 6). In contrast, Bainkhala soils showed very small variation (0.77 to 0.88) of this index value. In contrast, Dhulkot soils showed very small variation (1.12 to 1.15) of this index value. The parametric paired t-test showed that the overall mean of the sensitivity index was statistically insignificant at p < 0.05 for each location. Hence, there is no reason to reject the hypothesis that the T value estimation by two methods do not differ significantly. So, in other words, it can be concluded that the estimation of soil loss tolerance by weighted additive approach was generally in good agreement with the results of the PI–based approach. This validation test suggests that the weighted additive approach could well be used by soil managers and policy planners for assigning T values for Doon Valley conditions and identical soils. Summary and Conclusions Soil conservation programs need to be justified in terms of productivity and environmental sustainability. Soil loss tolerance values can be used as a guide to decide the maximum soil loss value that can be permitted for a given soil without causing degradation of the soil. One of the major objectives of the present study was to compare the findings of the recently proposed approach to estimate soil loss tolerance employed earlier in the Doon Valley region of India. The efficacy of two multivariate techniques was tested by comparing the T values. The result suggests that fewer numbers of carefully chosen indicators, when scored nonlinearly and used in a simple index, can adequately provide the information needed for estimating SLT. Assessment tools that reliably reflect the permissible limit of erosion, such as the ones evaluated here, may significantly improve the sustainability of agricultural management practices. Comparison with the comprehensive multivariate technique yielded results similar to all the indexing combinations.

410


Location

(T)PI

(T)C

(T)PI ÷ (T)C

T statistic, i.e., the calcuDhulkot cropland 14.0 12.5 1.12 lated value of paired t-test Dhulkot agroforestry 14.4 12.5 1.15 (=0.21) being less than the Bainkhala-I 6.6 7.5 0.88 critical value of 3.18 (table Bainkhala-II 7.7 10.0 0.77 value). So, the null hypothOverall mean 0.98 esis (H0: [T]PI ÷ [T]C = 1.0)] is accepted. Standard deviation 0.18 Note: p value [H0: (T)PI ÷ (T)C = 1.0] using a paired t-test with the sites as replicates, where (T)PI represents the T-value index based on the PI approach,

The overall mean of the sensitivity index was statistically insignificant at p < 0.05 for each location. Both methods resulted in determining T values are equally representative of variability for different locations in the study area. However, the PI–based approach requires a complicated depth-wise dataset including available water capacity, BD, and pH, which at present is not available for most of the agroecological regions. Generating such data sets would require a long time and a huge investment. On the other hand, the weighted-additive model requires an MDS of six soil attributes, which are readily available for all locations.

Daily, G., P. Dasgupta, B. Bolin, P. Crosson, J. Du Guerny,

Acknowledgements

Gregorich, E.G., M.R. Carter, D.A. Angers, C.M. Monreal,

P. Ehrlich, C. Folke, A.M. Jansson, B-O. Jansson, N. Kautsky, A. Kinzig, S. Levin, K.-G Mäler, P. PinstrupAndersen, D. Siniscalo, and B. Walker. 1998. Food production, population growth, and the environment. Science 281:1291-1292. Delgado, F. 2003. Soil physical properties on Venezuelan steeplands: Applications to conservation planning. The Abdus Salam International Centre for Theoretical Physics. College on Soil Physics. Delgado, F., and R. Lopez. 1998. Evaluation of soil degradation impact on the productivity of Venezuelan soils. Advances in GeoEcology 31:133-142. Dhruvanarayana, V.V., and Rambabu. 1983. Estimation of soil erosion in India. Journal of Irrigation and Drainage Engineering 109(4):419-434.

The authors express their gratitude to all the scientists of

and B.H. Ellert. 1994. Towards a minimum dataset

Central Soil and Water Conservation Research and Training

to assess soil organic matter quality in agricultural.

Institute (ICAR) who have directly or indirectly contrib-

Canadian Journal of Soil Science 74:367–385.

uted to ensure timely completion of the study. Financial

ISSS (International Soil Science Society). 1996. Terminology

support provided by the institute to undertake this study is

for soil erosion and conservation. Grafisch Service

thankfully acknowledged. We thank Mr. Ashok Kumar and Mr. M.P. Juyal for their help in preparing the document. We also thank two anonymous reviewers for their insights.

Centrum, Wageningen. Karlen, D.L., and D.E. Stott. 1994.A framework for evaluating physical and chemical indicators of soil quality. In Defining Soil Quality for a Sustainable Environment, ed.

References Bakker, M.M., G. Govers, and M.D.A. Rounsevell. 2004.The crop productivity-erosion relationship: An analysis based on experimental work. Catena 57(1):55-76.

J.W. Doran et al., 53-72. SSSA Special Publication. No. 35. Madison, WI: Soil Science Society of America. Khola, O.P.S., and G. Sastry. 2005. Conservation efficiency of mechanical measures and productivity of maize-

Beach, T., and P. Gersmehl. 1993. Soil erosion, T values,

wheat cropping sequence under different land slopes

and sustainability: A review and exercise. Journal of

in foothills of western Himalaya. Indian Journal of Soil

Geography 92:16-22.

Conservation 33(3):221-224.

Bhardwaj, S.P. and P.N. Singh. 1981. Soil, land capability

Kok, K., M.B.W. Clavaux, W.M. Heerebout, and

land use characteristics of soil conservation research

K. Bronveld. 1995. Land degradation and land cover

farm, Dehradun, 4-8. Technical Bulletin No. T-14/D-

change detection using low-resolution satellite images

11, Central Soil and Water Conservation Research and

and the CORINE data base: A case study in Spain. ITC

Training Institute, Dehradun, India.

Journal 3(7):217-227.

Biederbeck,V.O.,V. Rasiah, C.A. Campbell, R.P. Zentner, and

Lakaria, B.L., H. Biswas, and D. Mandal. 2008. Soil loss

W. Guang. 1998. Soil quality attributes as influenced by

tolerance values for different physiographic regions of

annual legumes used as green manure. Soil Biology and Biochemistry 30(8-9):1177-1185. Clark II, E.H., J.A. Haverkamp, and W. Chapman. 1985. Eroding Soils: The Off-farm Impacts. Washington DC:

Central India. Soil Use and Management 24:192-198. Lal, R. 1988. Assessing soil erosion – crop yield relationships. In Soil Erosion Research Methods, ed. R. Lal. Ankeny, IA: Soil and Water Conservation Society.

The Conservation Foundation.



Lal, R. 1995. Erosion-crop productivity relationships for soils of Africa. Soil Science Society of America Journal 59(3):661-667.

Revised Universal Soil Loss Equation (RUSLE). USDA Agriculture Handbook No. 703. Rijsberman, F., and M. Wolman. 1985. Effect of erosion on

Lal, R. 1998. Agronomic consequences of soil erosion. In Soil Erosion at Multiple Scale: Principals and Methods

soil productivity an international comparison. Journal of Soil and Water Conservation 40(4):349-354.

for Assessing Causes and Impacts, ed. P. Devries,

Ringo, D.E. 1999. Assessment of Erosion in the Turasha

F. Augus, and J. Kerr, 149-160. International Board for

Catchment in the Lake Naivasha Area, Kenya. MSc

soil research and management (IBSRAM). Wallingford,

thesis, International Institute for Geo-Information

UK: CABI Publishers.

Science

Li, L., Z.H. Zhou, and G.C. Liu. 2005. The present situation and conceive of soil loss tolerance study. Advances in Earth Sciences 20:65-72.

and

Earth

Observation,

Enschede,

The Netherlands. Schertz, D.L. 1983. The basis for soil loss tolerances. Journal of Soil and Water Conservation 38(1):10–14.

Liu, G.C., L. Li, L. Wu, G. Wang, Z.H. Zhou, and S. Du. 2009. Determination of soil loss tolerance of an Entisol in Southwest China. Soil Science Society of America Journal 73(2):412-417.

Soil Conservation Society of America. 1982. Resource Conservation Glossary, Third edition. Steel, R.G.D., and J.H.Torrie. 1980. Principal and procedures of statistics: A biometrical approach, 2nd ed. New York,

Lobo, D., Z. Lozanoand, and F. Delgado. 2005. Water erosion risk assessment and impact on productivity of a Venezuelan soil. CATENA 6(2-3):297-306.

NY: McGraw-Hill Book Company. Tiwari, D.N. 1991. Ecological economic approach to reosurce use impact analysis at the local level. A case

Mandal, D., K.S. Dadhwal, O.P.S. Khola, and B.L. Dhayni. 2006. Adjusted T values for conservation planning in Northwest Himalayas of India. Journal of Soil and Water Conservaton 61(6):391-397.

study of Nakkhu Khola watershed, report submitted to ICIMOD, Kathmandu, Nepal. USDA NRCS (Natural Resources Conservation Service). 1999. National Soil Survey Handbook - Title 430-VI.

McCormack, D.E., K.K. Young, and L.W. Kimberlin. 1982.

http://soils.usda.gov/technical/handbook/.

Current criteria for determining soil loss tolerance.

Wischmeier, W.H., and D.D. Smith. 1978. Predicting

In Determinants of Erosion Tolerance, ed. D.M. Karl,

rainfall erosion losses - a guide to conservation planning.

95-111. Madison: American Society of Agronomy.

USDA Agricultural Handbook 537, Washington DC.

Matson, P.A., W.J. Parton, A.G. Power, and M.J. Swift. 1997. Agricultural

intensification

and

ecosystem

properties. Science 277:504-509. Montgomery, D.C. 1984. Design and Analysis of Experiments, 2nd ed. New York, NY: Jon Wiley and Sons. Nearing, M.A., L. Deer-Ascough, and J.M. Laflen. 1990. Sensitivity analysis of the WEPP hillslope profile erosion model. Transactions of American Society of Agricultural Engineering 33:839-849. Pierce, F.C., W.E. Larson, R.H. Dowdy, and W.A. Graham. 1983. Productivity of soils: Assessing long-term changes due to erosion. Journal of Soil and Water Conservation 38(1):39–44. Pimentel, D., C. Harvey, P. Resosudarmo, K. Sinclair, D. Kurz, M. Mcnair, S. Crist, L. Shpritz, L. Fitton, R. Saffouri, and R. Blair. 1995. Environmental and economic costs of soil erosion and conservation benefits. Science 267:1117-1123. Ponzi, D. 1993. Soil erosion and productivity: A brief

review.

Desertification

Control

36-44. No.22. Nairobi, Kenya: United

Bulletin, Nations

Environment Programme. Renard, K.G., G.R. Foster, G.A. Weesies, D.K. McCool, and D.C. Yoder. 1997. Predicting Soil Erosion by Water: A Guide to Conservation Planning With the




411