Seasonal and interannual variation of radiation and energy fluxes over a rain-fed cropland in the semi-arid area of Loess Plateau, northwestern China

Understanding the land-atmosphere interactions over the semi-arid area of Loess Plateau is important due to its special climate and unique underlying surface. In this study, two years' micrometeorological and energy flux observations from the Pingliang Land Surface Process & Severe Weather Research Station, CAS were used to investigate the seasonal and interannual variations of radiation budget and energy fluxes over a rain-fed cropland in the semi-arid area of Loess Plateau, with an emphasis on the influence of rain, soil moisture and agricultural production activities (such as crop type and harvest time) on the energy partitioning as well as the surface energy balance. The results revealed large annual variations in the seasonal distribution of precipitation, which gave rise to significant seasonal and interannual variations in soil moisture. Soil moisture was the main factor affecting radiation budget and energy partitioning. There was a negatively linear relationship between the albedo and the soil moisture. The main consumer of available energy varied among months and years with an apparent water stress threshold value of ca. 0.12 m(3) m(-3), and the evapotranspiration was suppressed especially during the growing season. On an annual scale, the largest consumer of midday net radiation was sensible heat flux in 2010-2011, while it was latent heat flux in 2011-2012, which accounted for about 35% and 40% of the net radiation, respectively. The agricultural activity altered the sensitivity and variability of albedo to soil moisture, as well as energy partitioning patterns. The surface energy budget closures during Dec. 2010-Nov. 2011 and Dec. 2011-Nov.2012 were 77.6% and 733%, respectively, after considering the soil heat storage. The closure was comparable to other sites in ChinaFLUX (49% to 81% of 8 sites). The patterns of energy partitioning and the water stress threshold found in the semi-arid cropland could be used to evaluate and improve land surface models. (C) 2016 Elsevier B.V. All rights reserved.