Soil evaporation reduction can improve rainfed crop production in semi-arid climates. The purpose of this study was to develop empirical models to estimate cumulative soil evaporation (∑Es) from a maize field across the basin and runoff area (BA and RA) of the in-field rainwater harvesting (IRWH) technique as influenced by different levels of stover mulch ("dry-mulch") and canopy shading ("green-mulch"). Soil water evaporation (Es) was measured with micro-lysimeters 300 mm deep and 71 mm in diameter in the BA and RA sections for three consecutive drying cycles. Four different runoff strip lengths, RSL (RSL-1, 1.5, 2 and 3m) and three mulching levels (ML) of bare, 39%, and 96% were used to estimate the Es. The coefficient crop cover (Kcover) values for full and partial canopy shade (FC and PC) on the RA were in the range of 0.46–0.97 and 0.43–0.94, respectively, while in the BA the Kcover values ranged between 0.33 and 0.92. Under IRWH, the ∑Es values from a basin area beneath a maize canopy averaged 42% of the seasonal rainfall. ∑Es proportion of seasonal rainfall is about 62, 64, and 66 percent for bare soil, but as low as 28, 30 and 32 percent for FC, PC, and unshaded canopy (UC), respectively. In conclusion, the microclimate of the cropping system changes according to surface treatments, which greatly affect the Es losses for the IRWH technique and contributes to ET partitioning. Hence, reducing unproductive water losses in dryland crop production improves water use efficiency by storing more water in the root zone for transpiration.