Indicators for nitrogen mineralization in paddy rice soils in Bangladesh

The N requirement for paddy rice cultivated in Bangladesh amounts to approximately 50-120 kg N ha-1. Lack of knowledge on N mineralization from soil organic matter (SOM) leads farmers to fill up this N requirement exclusively by costly mineral fertilizers, which have typically an efficiency of less than 50%. After decades of searching for a practical tool to estimate the N mineralization capacity of soil, there is still no consistent methodology. We assessed different physical and chemical fractions of soil organic matter (SOM) as predictors for the net N mineralization measured from aerobic (field capacity) and anaerobic (saturated) soil cores in 100 days laboratory incubations. First, N mineralization and C and N were assessed from long-term experimental plots of a highly weathered ‘terrace soil’ and a younger fluvisol, involving combinations of mineral fertilizer dressings and OM application. A positive correlation between N mineralization and soil N content, which was clearly related to management, was found for the terrace soil while no correlation was found for the fluvisol. Instead a strong negative relationship with soil OC content suggests a simultaneous accumulation of SOC and immobilization of N in this soil. Hence the relation between OM management for SOM accumulation and availability of N through mineralization is not straightforward in different soils in Bangladesh. Therefore other predictors than soil N content or C:N ratio were investigated . A soil set consisting of 29 paddy rice fields were physically fractionated into particulate OM (POM) and silt and clay associated OM following mild ultrasonic dispersion and wet sieving. The silt and clay sized OM was further chemically fractionated by oxidation with 6% NaOCl to isolate an oxidation-resistant OM fraction, followed by extraction of mineral bound OM with 10% HF thereby isolating the HF-resistant OM (NaOCl-res. HF-res.OM). A parallel fractionation scheme involved a ‘thermal fractionation’ by sequential water extraction at 100, 150 and 200°C via accelerated solvent extraction. Several physical and chemical fractions correlated positively with the linear fitted aerobic N mineralization rate, while no correlations were observed with anaerobic mineralization. The R2 values of linear regressions between N content of the physicochemical fractions and aerobic N mineralization rate were, however, lower than the R2 found for bulk soil N content. ‘Thermal fractionation’ yielded two fractions which correlated better with aerobic N mineralization, namely the 100+150°C extractable N and 200°C extractable N, but the R2 were close to bulk soil N. In conclusion, while some useful predictors for aerobic N mineralization were found, insight in the anaerobic N mineralization is lacking. Further research into importance of soil mineralogy, surface area, soil texture, etc. will be needed to increase our understanding of anaerobic mineralization