AGGREGATE STABILITY AND RAIN-IMPACTED SHEET EROSION OF AIR-DRIED AND PREWETTED CLAYEY SURFACE SOILS UNDER INTENSE RAIN

Stability of aggregates at the immediate soil surface affects the deta chment, transport, and deposition of soil particles. The aggregate sta bility (expressed as mean weight diameter, MWD) and erosion rate were tested for clayey surface soils at two different water contents. The s oils consisted of six Oxisols, one Ultisol, three Vertisols, one Alfis ol, and one Aridisol. The MWD was determined by wet sieving of 4.76 to 8.00 aggregates, which were either air-dried (MWD(d)) or capillary pr ewetted at -0.5 kPa for 24 h (MWD(w)). Air-dried and prewetted (satura ted from the bottom up for 2 h) soils were subjected to a 110-mm h-1 r ain until a steady-state infiltration was reached. Runoff, sediment lo ss, and infiltration (for prewetted soil only) were monitored througho ut the rain. The MWD(d) ranged from 0.18 to 4.80 mm, and the MWD(w) ra nged from 0.44 to 5.66 mm. Prewetting increased the MWD for the oxidic or kaolinitic Oxisols and Ultisols and for one smectitc/kaolinitic Ve rtisol. The MWD(w) was positively related to CEC for highly weathered soils and negatively related for swelling soils, whereas base saturati on was negatively related for both groups of soils. The DELTAMWD, whic h expresses the increase in stability caused primarily by a decrease i n slaking, was greatest for soils rich in kaolinite and Fe and Al oxyh ydroxides and low in organic carbon. The steady-state infiltration rat e (I(s)) ranged from 1.8 (illitic soil) to 54.8 mm h-1 (sandy loam kao linitic soil). The I(s) was consistently less for swelling than for hi ghly weathered soils, and it was related to the sand/clay ratio, sand, and water retained between -33 and -1500 kPa. The increased aggregate stability with prewetting was generally not reflected in smaller amou nts of erosion and runoff, possibly because of the faster rate of wett ing than for the stability test, a higher water table for the prewette d soil than for the air-dried soil, and the inclusion of smaller aggre gates in the erosion tests. Prewetting decreased the time to ponding f or three highly weathered Oxisols and the illitic Alfisol, increased s teady-state runoff rate (Q(w)) for five highly weathered soils, increa sed sediment loss rate (Q(s)) for the illitic and for one highly weath ered soil, and increased Q(s) on a stable Oxisol. Soil textural proper ties (usually clay and sand/clay) and ''binding'' agents (organic carb on and Fe and Al oxyhydroxides) were the significant independent varia bles included in the stepwise multiple linear regressions for I(s), Q( s), and Q(w).