Grass canopy interception of Hulun watershed under different grazing systems

Abstract

Different grazing systems have different influences on the biomass, leaf area index (LAI) and canopy morphology of grassland vegetation. Canopy interception is an important part of both water allocation and the balance of water across the whole basin during precipitation. In this paper, water soakage and rainfall simulation methods were used to study the change in rainfall interception by the grass vegetation canopy under three different grazing systems: no grazing, rotation grazing and free grazing (over grazing), in the Hulun watershed, Inner Mongolia, China. We discuss the relationships between trends in individual plants and the vegetation interception as well as the trends between the overall plant height and weight measurements and vegetation interception. Satellite image data from Landsat TM5 were used to extract the Normalized Difference Vegetation Index (NDVI) of grassland in the Hulun watershed and the actual coverage of the no grazing, rotation grazing and free grazing grasslands. The relationship of actual coverage and NDVI was used to measure the grassland areas in the three grazing systems. Estimating the amount of rainfall interception of three different grazing grasslands in the Hulun watershed, even without considering the spatial and temporal variation of rainfall, can provide basic data for future basin water resources planning studies. The vegetation interceptions estimated by the water soakage method for no grazing, rotation grazing and free grazing grasslands were 0.468 mm, 0.320 mm and 0.271 mm, respectively. Those estimated by the rainfall simulation method were 0.957 mm, 0.613 mm and 0.431 mm, respectively. The predominant vegetation types having an interception advantage from three different grazing systems were; Stipa krylovii, Leymus chinensis, Cleistogenes squarrosa and Artemisia firigida. In terms of leaf area proportion, the main interception species were Stipa krylovii and Cleistogenes squarrosa in free grazing, Stipa krylovii and Leymus chinensis in rotation grazing and Stipa krylovii with no grazing. Thevalues of coverage, soil bulk density, biomass and LAI were significantly (P <0.05) different among the three grazing systems. The LAI values were 4.36 m2 / m2 with no grazing, 2.72 m2 / m2 in rotation grazing and 1.37 m2 / m2 in free grazing. The height and weight of individual plants were also linearly correlated with interception in the three grazing systems with vegetation interception increasing with the height and weight of individual plants. The height and weight of the plants overall was linearly correlated with interception in the no grazing and rotation grazing grasslands, where interception increased with a height and weight increase. There was no such trend in the free grazing grassland. The overall plant cover was linearly correlated with interception in all three grazing grasslands. The vegetation interception of the plants overall and the individual plants had obvious phenomenon grading in all three grasslands. Through field measurements in the Hulun watershed, the cover ranges of no grazing, rotation grazing and free grazing grasslands were 79%-98%, 57%-84% and 35%-62%, respectively. Thus, the total canopy interception in the Hulun watershed will be 6.462×106 m3 and the canopy interception of the different grazing systems will be 4.995×105 m3 with no grazing, 3.283×106 m3 with rotation grazing and 2.679×106 m3 with free grazing if the whole basin rainfall reaches 30 mm.