Modeling risk analysis for rice production due to agro-climate change in Taiwan

摘要

This study proposes a risk analysis model for the rice production due to climate change in terms of agro-climate indices (i.e., cumulative temperature anomaly, cumulative precipitation anomaly, cumulative sunlight anomaly, cumulative radiation anomaly, and E1 Nio). This risk analysis model is developed by incorporating the multivariate Monte Carlo simulation method, multivariate regression equation, and uncertainty analysis method (advanced first-order second-moment, AFOSM). The study area is composed of 15 counties/cities in Taiwan, East Asia. The data set for the model development and applicability contains 27 years of annual rice productions and agro-climate indices in addition to cultivation areas. Through the proposed risk analysis model, it can be seen that the rice production in Taiwan is especially sensitive to temperature, precipitation, and sunlight. Also, on average, improving performance by reducing insufficient rice risk can rise by 80 % when the rice production increases from 3 x 10(4) to 3 x 10(5) tons.