SEARCH

SEARCH BY CITATION

Keywords:

  • multiple regression analysis;
  • physicochemical properties;
  • principal component analysis;
  • potentially mineralizable organic matter;
  • recalcitrant organic matter

Abstract

To understand the dynamics of soil organic matter in humid regions in Asia, several soil and meteorological properties were tested to estimate the amounts of potentially mineralizable organic carbon (PMC) and nitrogen (PMN) as well as recalcitrant organic carbon (ROC) and nitrogen (RON). Eighty-nine surface soil samples were collected in Thailand, Indonesia and Japan from cropland and forest ecosystems. The fresh soils were incubated for 133 days under constant temperature and moisture conditions, and CO2 emissions and mineral N from the soils were monitored. The PMC and PMN were determined by fitting models to the cumulative curves of the CO2 and mineral N. The ROC and RON were determined by subtracting PMC and PMN from total C and total N, respectively. The soil properties tested included soil pH, sand, silt and clay contents, amounts of the acid oxalate soluble Si, Al and Fe, carbon and nitrogen contents of the light fraction ([LF] < 1.6 g cm−3) and heavy fraction (HF), and the C/N ratio of the LF and HF. The meteorological properties considered were mean annual temperature and precipitation. Using multiple regression analysis and a stepwise method, PMC and PMN was well estimated by the carbon and nitrogen content of LF (LFC and LFN) and the clay content. This suggests the partially labile nature of clay-associating organic matter as well as LFC and LFN. The PMC and PMN were well estimated by LFC and LFN in forest soils and clay contents in cropland soils. Major factors controlling ROC and RON were light fraction-organic matter, amorphous materials and clay content, suggesting the importance of parent materials in estimating the amount of recalcitrant organic matter in humid Asia.