Predicting water retention is difficult for highly aggregated volcanic soils (Andisols) with heterogeneous pore size distribution. This paper aimed to examine how andic materials influence soil water retention based on… Click to show full abstract
Predicting water retention is difficult for highly aggregated volcanic soils (Andisols) with heterogeneous pore size distribution. This paper aimed to examine how andic materials influence soil water retention based on soil composition. We compiled the data of water retention in the soils from literature with physicochemical properties and soil constituents (organic matter, sand, silt, crystalline clay, and short‐range‐order mineral fractions) to obtain the parameters of water retention curve models (van Genuchten, Durner, and Assouline). The pedo‐transfer functions were successfully constructed in the van Genuchten and Assouline models with bulk density, pH, soil C, cation exchange capacity, sand content, and andic property, but the direct estimation from soil composition (clay content, sand content, soil C, oxalate extractable Al, Fe, and Si) achieved the better prediction of soil water retention curves. Organic matter increases plant‐available water retention (pF 1.5–4.2), while short‐range‐order minerals in volcanic soils enhance water retention at pF 4.2–7.0. This is caused by the larger volume of finer micropore (av. 0.012 µm in diameter) in volcanic soils, compared to non‐volcanic soils (av. 0.020 µm). Our study highlights the importance of including andic properties (oxalate‐extractable aluminum and iron) to improve prediction of soil water retention in volcanic regions, although soil water retention could roughly be predicted by sand content at the global scale.
               
Click one of the above tabs to view related content.