Quantifying and predicting soil water evaporation as influenced by runoff strip lengths and mulch cover

Abstract

Soil water evaporation from the cropping surface is a wasteful loss of potentially productive rainwater, thus efficient use of rainwater can help to sustain dryland production. The purpose of this study was to quantify the effect of canopy shading (CS) and mulch levels (ML) on soil water evaporation (Es) from each 1m section of in-field rainwater harvesting (IRWH) and to evaluate the Ritchie (α′) and Stroosnijder (β′) soil evaporation models on the effect of surface treatments. A microlysimetric method was used to measure Es from beneath maize (Zea mays L.) canopy for three consecutive drying cycles across the basin and runoff sections of IRWH on fine sandy loam soil of Bainsvlei Kenilworth ecotope. First, main effects of four runoff strip lengths (RSL) and three ML treatments were statistically analysed on the weighted Es values. Second, the ML treatments were allocated to the main plots and four levels of CS allocated according to lengths of the runoff sections. Third, cumulative Es (∑Es) measurements were used to evaluate empirical equations related to time (α′) and potential evaporation (β′). The two models for Es were compared by considering the effects of surface treatments. A significantly higher Es was observed from a bare (ML0%) treatment compared with either of two mulched treatments viz. mulch level 39% and 96% cover (ML39% and ML96%); no significant differences were found between the mulched treatments. The insignificant effect of RSL treatments on Es implied the dynamics of spatial distribution of soil water and energy that influenced evaporation were as a result of green mulch or shading cover (CS) on Es beneath the canopy. Less suppressive Es properties were developed from bare surface and efficient Es restriction was found under high mulch and shading cover treatments. The α′ and β′ values ranged from 2.34 to 4.26mmd−0.5 and from 1.38 to 2.06mmd−0.5, respectively. In all the treatments the simulated ∑Es was underestimated by the Ritchie model and overestimated by the Stroosnijder model. The main effect of shading was due to the dominant effect of energy limited evaporation (stage-1), while the mulched treatments were mainly driven by soil limited stage (stage-2) of evaporation. The Ritchie model performed well to estimate ∑Es from the basin area and the potential Stroosnijder model from the unshaded runoff strips. The microclimate of the cropping system changed according to surface treatments that highly influenced the Es losses in IRWH of dryland production.