Validation of soil organic carbon dynamics model in the semi-arid tropics in Niger,West Africa |
| |
Authors: | Satoshi Nakamura Keiichi Hayashi Hide Omae Tabo Ramadjita Fatondji Dougbedji Hitoshi Shinjo Addam Kiari Saidou Satoshi Tobita |
| |
Affiliation: | (1) Japan International Research Center for Agricultural Sciences (JIRCAS), 1-1 Ohwashi Tsukuba, Ibaraki 305-8686, Japan;(2) International Crops Research Institute for Semi-Arid Tropics, West and Central Africa (ICRISAT-WCA), Niamey, B.P. 12404, Niger;(3) Graduate School of Agriculture, Kyoto University, Kyoto 606-8502, Japan;(4) Institut National De La Recherche Agronomique du Niger (INRAN), Niamey, B.P. 429, Niger |
| |
Abstract: | The fertility of sandy soils in the Sahelian zone (SZ) is extremely low. This poor soil fertility is one of the limiting factors
of crop production in the SZ. Therefore, it is imperative to improve or to maintain soil fertility through various agricultural
management methods. Further, it is well known that soil organic matter plays an important role in improving the physico-chemical
properties of these sandy infertile soils. Therefore, it is essential to develop a suitable tool for the appropriate evaluation
of soil organic carbon (SOC) dynamics in the SZ. Therefore, the Rothamsted carbon model (Roth-C) was verified in 32 treatments
of two long-term field experiments with and without crop residue application. These experiments were performed by ICRISAT.
The performance of the model was evaluated by statistical methods using four indices (RMSE: root mean square error, LOFIT: lack of fit, r: correlation coefficient, and M: mean difference). As a result, the predicted SOC values in the case without crop residue management decreased with time
in approximately 10 cultivated years. In contrast, in the case with crop residue application, the predicted SOC remained roughly
equal to the initial SOC value during the term observed. Mostly, the Roth-C-modelled values agreed well with the actual value.
RMSE and LOFIT, the statistical indicators of agreement between predicted and observed values, showed a significant conformity between the
predicted and observed SOC values in all the 32 treatments. This fact means that Roth-C can estimate long-term SOC dynamics
of several technical options that developed with short-term trials. Moreover the annual carbon requirement for SOC maintaining
can be calculate if enough number of cases was estimated. And also analysis of regional carbon dynamics was made possible
with using Roth-C model. It will contribute to show the sustainable development in SZ against global warming and other climatic
changes. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|