Rajan, K. , Natarajan, A. , Kumar, K. S. Anil , Badrinath, M. S. , Gowda, R. C.
2010-09-25 CURRENT SCIENCE 2010 99(卷), 6(期), (823-827页)
Degradation due to erosion, salinity, waterlogging, contamination, etc. reduces the quality of land. Soil and water resources are affected initially then the bio-resources. Uncontrolled degradation processes lead to desertification of land and the ecosystem functions are disturbed. In India, soil erosion has produced maximum degraded lands due to mismanagement and indiscriminate use of land resources. Monitoring of land degradation status is essential to take appropriate and timely soil-conservation measures. Numerous soil and land-quality indicators are used to assess the degradation status. In this study, an attempt has been made to select a few reliable indicators using multivariate statistical tools. Eroded land of southern Karnataka was selected for the study using remotely sensed satellite imageries. Soil samples were collected depth-wise in profiles at different landforms and land uses. There were 24 soil quality indicators estimated from eroded lands to select few indicators for monitoring degradation status. Principal component analysis was employed for the reduction of 24 indicators. Out of the 24 principal components formed, only seven accounted for maximum variance. Twelve soil quality indicators were selected from the seven components based on the highest loadings. Linear discriminant analysis was employed to obtain the most discriminating soil quality indicators from among the 12 selected. Soil organic carbon (SOC) emerged as the most discriminating soil quality indicator with maximum loadings (-2.43). Apart from SOC, there were four other indicators with loadings >1.0. Electrical conductivity (1.95), available soil water (1.45), micro-aggregates (-1.18) and dehydrogenase activity (1.08) have also gained considerable importance along with SOC in monitoring degradation status of land by soil erosion.