Dias, Henrique Boriolo, Sentelhas, Paulo Cesar, Inman-Bamber, Geoff, and Everingham, Yvette (2021) Sugarcane yield future scenarios in Brazil as projected by the APSIM-Sugar model. Industrial Crops and Products, 171. 113918.

PDF (Published Version) - Published Version Restricted to Repository staff only

 

8

1

Abstract

Crop models like APSIM-Sugar have been used to assess the impacts of climate variability and change on sugarcane. APSIM-Sugar was recently upgraded to simulate sugarcane in Brazil and to cater for climate impacts under water-limited environments. In this context, our first objective was to evaluate the recent upgrade on determined traits for the most cultivated Brazilian variety under water-limited conditions. As a second objective, we applied these model settings to project the sugarcane crop performance under likely changing climates in important producing regions. The stalk mass estimates for four published experiments across Brazil were satisfactory, with a R2 of 0.96, the agreement d index of 0.99, and root mean square error (RMSE) of 3.8 t/ha. Future climate projections (until 2099), considered two emission scenarios (RCP4.5 and RCP8.5), were derived based on a recently proposed subset approach of five climate models (GCMs) from CMIP5. The subset aimed to capture the full ensemble of temperature and rainfall changes by considering basic classes of climate changes (relatively cool/wet, cool/dry, middle, hot/wet, and hot/dry). A typical system with 12-month old crops was simulated for rainfed and irrigated sites. The ensemble of future projections indicated that cane yields would decline compared to the present average simulations (1980−2009) for rainfed sites, mainly due to increased water stress. Yields at fully irrigated sites are also projected to decline slightly. However, the expected changes were highly variable between GCMs with the End-of-Century period presenting large uncertainty. In short, while providing an up-to-date projection of future climate change impacts on sugarcane in Brazil, such results should be interpreted with caution.

Actions (Repository Staff Only)

Item Control Page