It is of critical importance to evaluate the regional carbon cycling and balance more accurately as global warming is the key issue of the climate changes. This research is aimed to reveal the soil respiration(SR) and its responses to soil temperature and water in a waterwind erosion interlaced zone in the Loess Plateau, China. Hourly SR in bareland, cropland, alfalfa land, cartagena korshinkii land and abandoned land were measured automatically from May to October, 2012, using Infra- Red Gas Analysis(IRGA) method. Soil temperature at depth of 5 cm and 15 cm(T_5 and T_(15), respectively) and volumetric water content of 0~10 cm and 10~20 cm soils(VWC_(0-10) and VWC_(10-20), respectively) were also monitored. The SR diurnal dynamics was a single-peak curve, with the peak occurring between 13:00 (UTC+8, similarly hereinafter) and 15:00, and the trough between 05:00 to 07:00. The SR measured at 9:00 and 19:00 was equal to the daily average value. Land use patterns had significantly impacts on SR (P<0.01), with order of alfalfa land>cartagena korshinkii land>abandoned land>cropland>bare land. The SR had significant positive correlation with T_5, T_(15), VWC_(0~10) and VWC_(10~20) for five land use patterns, but the correlationship with soil water was weaker than with soil temperature. The correlation coefficients were greater for T_5 than T_(15) and VWC_(0~10) than VWC_(10-20). The precipitation increased SR during May to late August, but inhibited SR during September to late October. Soil SR increased with increasing soil water when VWC_(0~10) was less than 0.2 cm~3·cm~(-3), but decreased as soil water further increased when VWC_(10~20) was greater than 0.2 cm~3·cm~(-3). The E- P- Q(Exponential- Piecewise- Coefficient) model could rationally explain relationships between SR and soil temperature and/or soil water.