Risk Analysis of Planting Date Under Current and Projected Climate Scenarios in Chile Using Substor-Potato.

Planting date is a key risk management factor for crop production, yet potato producers in southern Chile lack site-specific data on optimal planting dates and how these might be impacted by climate change.  The objective of this study was to assess the risk of yield losses due to planting date under current and projected climate scenarios using the SUBSTOR-Potato model in DSSAT (Decision Support System for Agrotechnology Transfer).  Experimental data collected in 2014 and 2015 at two locations in Chile were used to calibrate the genetic coefficients for the cultivar Patagonia.  Data from 2005-2015 and 12 locations were used to evaluate model performance.  Crop yield under rainfed conditions was simulated for eight planting dates between mid-August and late November for five environments ranging in latitude from 36.907 to 41.433^o S and average annual precipitation from 1128 to 1871 m.  Simulations were conducted under recent weather conditions (1983-2015) and under projected weather scenarios generated for 2050 and 2080 by DSSAT Perturb using the HADCM3 global circulation model (GCM) and an ensemble of 40 GCMs, and two Representative Circulation Pathways (RCP6.0 and RCP8.5).  SUBSTOR-Potato predicted potato yield well for the Patagonia cultivar over multiple locations (0.91 Willmott index, 0.71 R2, 27% nRMSE, and 0.56 modelling efficiency).  Yield potential under rainfed conditions differed among the five environments, ranging from 2 to 19 Mg ha^-1 dry matter, which was attributed primarily to differences in water supply.  Delaying planting reduced yields at all sites, but the sensitivity of tuber yield to planting date varied greatly among environments and was positively associated with average annual rainfall.  Results from model simulations under future climate scenarios will be presented.