Building a Platform to Communicate Long-Term Climate Projections and Climate Analogs

Abstract:

As climate change intensifies, long-term changes in temperature and precipitation risk destabilizing the agriculture sector. For growers, making sense of the inherent uncertainty and opaqueness of future climate projections is imperative for contingency planning and climate adaptation. Using data from the Climate Model Intercomparison Project 6 (CMIP6), we propose answers to the following questions about long-term climate at a specific location: (1) What is the projected change in temperature and precipitation? (2) How will these projected changes affect individual crop types? (3) When will climate change fundamentally alter a location’s climate? (4) Where should I move after this period of this departure? (1) We demonstrate bias-correction and weighting techniques using reanalysis and CMIP6 models that reduce the root mean square error on hold-out simulated historical data, and the uncertainty of future projections. (2) In addition, we identify the climate niche (Xu et. al.) of the crop grown at this location, or the ideal temperature and precipitation range for a crop based on global classification data. We also show the location’s current and future position in this climate niche. (3) Next, we find the future twenty-year period (Challinor et. al.) where the climate change signal is detected for this given location based on departures in climate variability. (4) Finally, for this period of departure, we find climate analogs of the location’s historical climate. For growers seeking to adapt to climate change, these locations are possible locations where they can move their operations.