| Since the project of “grain for green” was put into practice in the Loess Plateau, the vegetation community types and coverage due to vegetation construction has been fundamentally changed, soil physicochemical characteristics has been improved, the positive effect of plant roots on soil construction is gradually strengthened, and vegetation controlled significantly the gully development. In the past, most research use the remote sensing image of basin and area to analyze the gully development change after farmland returning, but these studies have not analyze and reveal the effect of soil and vegetation change on headward erosion through experimental installations. Hence, the paper took the rehabilitated grassland gully headward erosion in Loess Plateau-gully Region as the study object, with the farmland as the contrast group, the plots of gully headward erosion was established in experimental station of Nanxiaohegou in Gansu province. The method of simulated rainfall and runoff scouring experiment was first adopted to research the variation of gully headward erosion characteristic in the rehabilitated grassland with different coverage. The effect of soil physicochemical properties and root system on runoff and sediment yield, the runoff dynamic characteristics, gravity erosion temporal and spatial characteristics and gully head morphology changes were clarified. Research had important scientific significance and the value of actual production. The results can provide the scientific guidance for gully control and ecological environment construction in Loess Plateau Gully Region. The main conclusions are as follows:(1) Comparing with the farmland, soil physicochemical property of table-land and gully head abandoned farmland were improved. Soil anti-scouribility of the table-land and gully head increased 0.78~7.57 times and 7.57~7.12 times, respectively. The soil bulk density, infiltration coefficient, water stable aggregate content, disintegration rate, organic matter content, root biomass and density had significant effects against anti-scourability. Water stable aggregate was the most crucial parameter to affect the anti-scourability of abandoned land.(2) 1~2 mm and <0.5 mm root were the main factors influencing the runoff and sediment yield of headward erosion in rehabilitated grassland, respectively. Runoff rate and sediment yielding rate changed dramatically for the CK, 0%, 20%, 50% and 80% coverage degree of 3°and 6° in gully headward erosion process. Compared with CK, the sediment decreased efficiency of 0%~80% plots of 3° and 6° was 58.19%, 58.71%, 65.07%, 65.31% and 58.75%, 60.34%, 66.85%, 71.08%. 1~2 mm root length density in 0~20 cm soil layer was the optimal root parameter affecting total runoff volume. Sediment yield has a significant correlation with 1~2 mm aggregate and <0.5 mm root length density in 0~20 cm soil layer.(3) 0.02~0.05 mm was the critical particle size deciding the enrichment or decrease of sediment. Compared with original soil, the<0.02 mm particle of CK, 50% and 80% coverage plot decreased, and the>0.02 mm particle increased. However, there was a opposite tendency for 0% and 20% coverage plots. The sediment enrichment rate increased with the increment of particle size for all experiments of slope 6°. The content of<0.02 mm size sediment was less than original soil and increased for>0.02 mm size particle.(4) Runoff regime of farmland and rehabilitated grassland was belong to turbulent flow. The more coverage was, the more weakening turbulent flow was, and flow regime switched jet flow to slow flow when coverage was more than 20%. Runoff shear stress, stream power, unit stream power and section energy decreased with the increase of coverage. Resistance coefficient and roughness coefficient increased as coverage increased. Sediment yield rate and flow velocity, reynolds number, froude number, resistance coefficient, roughness coefficient, shear stress, stream power, unit stream power and section energy were significantly related.(5) Gravity erosion was triggered by rainfall and runoff, and the change of soil structure caused by vegetation coverage was the key factor deciding gravity erosion intensity and formation. Steep slope collapse and gully bank collapse were mainly collapse types in the farmland gully headward erosion. Layered erosion occurred in 20 cm soil layer due to root system existing on the gully head of rehabilitated grassland. Steep slope collapse was generated in 20~120 cm soil layer due to runoff scouring, and unstable suspended soil was formed in 0~20 cm soil and caused gully head collapse.(6) Experimental results revealed firstly that time and space distribution and frequency of gravity erosion of rehabilitated grassland in gully headward erosion process. The farmland collapse frequency of slope 3°and 6°were 11 times and 13 times, respectively. The collapse frequency of rehabilitated grassland coverage 0% of 3° and 6° were 5 times and 0 times due to the protection effect of root network. Compared with the farmland, the collapse frequency of 20%, 50% and 80% coverage of 3° and 6° increased by 1.36~3.0 times and 3.0~1.69 times, respectively. More than half of collapse events taken place in the terminal phase(>30min), and the frequency of collapse happening on the up-slope and mid-slope was more than 64.3%.(7) The study illuminated the different coverage rehabilitated grassland controlled the gully headward rate. For slope 3°, compared with CK, the gully length of plots of 0%~80% coverage decreased about 52.41%, 66.97%, 78.37% and 79.18%, respectively. For slope 6°, compared with CK, the gully length of plots of 0%~80% coverage decreased about 44.34%, 78.94%, 85.38% and 83.42%, respectively. The width of CK plot increased gradually as flow discharge increased. 0% coverage plot had not occurred collapse and the depth of the gully was only about 20 cm. 20%~80% coverage plots formed about 100 cm depth gully after gully head collapse occurring, since then the depth had a not obvious change. Apart from 0% coverage plot, gully depth of 20%, 50% and 80% coverage plot of 3° and 6° were 1.98 times, 1.88 times, 2.07 times and 1.12 times, 1.13 times, 1.07 times of that of CK plot, respectively.(8) The rehabilitated grassland controlled the gully development area, and the gully area of 0%~80% coverage plots of 3° and 6° decreased 66.36%~76.88% and 69.26%~86.92%, respectively. The coverage of rehabilitated grassland influenced directly the gully shape. The gully shape of CK plot and 0% coverage plot can be characterized by leptokurtic. The gully morphology of 20%~80% coverage plots was curved shape. The gully can be characterized by wide and shallow for 0% coverage rehabilitated grassland of slope 3° and 6°, and the breadth depth ratio was 3.08 and 3.75, respectively. The breadth depth ratio for the rehabilitated grassland 20%~80% coverage plots of slope 3° was less than the farmland. The breadth depth ratio for the rehabilitated grassland of slope 6° was close to CK.(9) Research revealed the mechanism that root controlled headward erosion and morphological development. The>2 mm root density in 0~20 cm soil layer was the optimal parameter determining gully morphology and depth. The gully split degree was depended on 0.5~1 mm root biomass in 20~120 cm soil layer. The length of gully head erosion had a closed correlation with>2 mm root biomass in 0~120 cm soil layer. Soil physical and chemical properties had no significant effect on the gully width depth ratio and ditch depth. The split degree had the most closely associated with>0.25 mm size aggregate. Soil disintegration rate had the most intense impact on the gully shape and the gully length. |
PDF Full Text Request