RESTORING PRODUCTIVITY TO AN ARTIFICIALLY ERODED DARK BROWN CHERNOZEMIC SOIL UNDER DRYLAND CONDITIONS
A field was artificially eroded by levelling in 1957 and then continuously cropped to barley for 7 yr. Subsequently, a wheat-fallow experiment was conducted from 1965 to 1979 to determine the effects of four fertilizer treatments and green manure (yellow sweet clover) on restoring the productivity to soil that had been "eroded" to various depths. After 22 yr and 14 crops, the productivity of the land from which soil was removed has been improved but not fully restored. Although green manuring with yellow sweet clover improved soil structure, wheat yields were not improved because of competition for soil moisture and poorer in-crop weed control in this part of the rotation. The addition of 45 kg N plus 90 kg P2O5 per hectare in each crop year to sites from which 8–10, 10–20, or 46 + cm of soil had been removed resulted in yield increases of 18, 46, and 70%, respectively, over the unfertilized check of each treatment; the average yields were 104, 91, and 70%, respectively, of the undisturbed, unfertilized (check) treatment. On "erosion" treatments where only 8–10 cm of soil were removed, 45 kg N plus 22 kg P2O5 per hectare were sufficient to restore the productivity. Precipitation apparently had a greater effect than fertilizer application on wheat yields. The loss of organic matter and associated soil structure characteristics seemed to be critical factors contributing to yield losses associated with soil erosion. These results show that it is more practical to use management practices that prevent soil erosion than to adopt the practices required to restore eroded soil. Key words: Soil erosion, topsoil loss, water-stable aggregates, soil organic matter, green manure, precipitation