Distribution characteristics and influencing factors of gully density on cropland in the Songnen typical black soil region

Abstract: Black soil region in the Northeast can be one of the most important bases for commercial grain production in China. However, the ever-increasing gully erosion has been a serious threat to agricultural development and ecological environment. It is still unclear on the spatial variation of gully density across this region. In this study, 998 sampling units ranging from 0.17-3.3 km2 were selected to investigate the density of classical gully on the cropland and the influencing factors in the Songneng black soil region, including Phaezoems areas of 10.1×104 km2 and Chernozems area of 11.1×104 km2. The systematic sampling method was also employed in the national soil erosion survey in 2011. Gullies in each sampling unit were extracted by manual visual interpretation using Google images with a resolution of 0.41m. 11 influence factors were determined in the spatial variation of gully density using Geodetector, including two climatic, three topographical, five pedological, and one hydrological factor. The results were as follows. 1) 738 units dominated by cropland were selected to investigate the gully spatial variation among the 998 sampling units. Gully was found in 35.64% of the 738 sampling units on the cropland with gullies in 79 units, in which the gully density cannot be interpreted exactly. Gully density was then analyzed using the other 659 units. 1 229 gullies were found in 184 sampling units. The average gully linear density was 3.91 km/km2 varying from 0.18 to 17.48 km/km2, while, the average gully areal density was 16 484.93 m2/km2 varying from 425.95 to 91 921.34 m2/km2. Most gullies were long and thin size with the average gully length and width of 516.93 and 3.83m, respectively. 2) In the gully linear and areal density using Geodetector, the slope gradient was the most important environmental factor, followed by annual average precipitation, altitude, and aspect, which were all statistically significant at the level of 0.05. Concerning the interaction of factors, the joint effect of slope and soil bulk density posed the greatest explanatory power on the spatial distribution of gully linear density on the cropland, whereas, the joint effect of slope and field capacity shared the greatest explanatory power on the spatial distribution of gully areal density on cropland. In general, the gully density on the cropland in Songnen typical black soil region was dependent mainly on the topography and precipitation. 3) In gully linear density, serious gully erosion was mainly distributed in the Phaezoems area, with the area above severe intensity (gully linear density>2 km/km2) accounting for 33.33%, and the area of extremely severe intensity (gully linear density>4 km/km2) accounting for 27.94%. Sampling units with high gully density were distributed almost evenly across the Phaezoems area, and then interspersed with the sampling units of low gully densities. This distribution was attributed to the landform of the low hills with varying slope gradients and long slope lengths. Less than 10% of the Chernozems area suffered gully erosion, while the units of extremely severe intensity were accounted for 6.65%, and mainly located near the border of the Chernozems area. The reliable performance was achieved in the systematic sampling combined with high-resolution remote sensing. An efficient reference can be offered to assess the spatial variation of gully density in the croplands of Songnen typical black soil regions. The finding can be expected to require more effective measures for gully prevention and control, in order to protect the valuable black soil in northeast China.