Evaluation of Roth C model using four Long Term Fertilizer Experiments in black soils, India

Abstract

Carbon content in soils changes depending on the land use system, type of management practice and time. There is an increasing concern about the soil quality vis-à-vis organic carbon content in soils due to global warming and enhanced CO 2 concentration in the atmosphere. This has led to estimate carbon stock in soils at global and regional levels. The objective of the present study was to evaluate Roth C model to estimate total organic carbon (TOC) changes under four long term fertilizer experimental sites representing sub-humid moist (Sarol and Nabibagh), sub-humid dry (Panjri) and semi-arid (Teligi) climate in India. The plant carbon input rate was calibrated using organic carbon and other soil parameters using Roth C. The results showed that Roth C could simulate changes in TOC in two contrasting eco-sites for surface soil layers. The root mean square error (RMSE) considered as modelling error ranged from 11.50 to 15.01, 4.70 to 11.60, 2.14 to 6.52 and 1.45 to 13.74 in the surface layers of Sarol, Nabibagh, Panjri, and Teligi sites, respectively. The simulation biases expressed by M (relative error) by Student‘ t’ value for all the treatments at these sites were non-significant with two exceptions. Observed trends in TOC consist of an increase for all the four treatments in the sub-humid site of Sarol and Nabibagh; while manures alone or in combination increase TOC appreciably in Teligi and Panjri. TOC remained, however, almost similar over years for the control (no fertilizer or manure) and NPK treatments in all the four sites. Analysis of Roth C output data showed that this model could be used as a tool to arrive at different threshold values of rainfall to influence decomposition rate modifier and thus to find out rate of organic carbon sequestration in various bioclimatic systems.