**2.1 Data**

In this study, four CMIP6 general circulation models (**Table 1**; [25]) and three greenhouse gas emission scenarios known as Shared Socioeconomic Pathways (SSPs; [26, 27]) (SSP1-2.6, SSP2-4.5, and SSP5-8.5,) were used. The CMIP6 GCM runs were developed in support of the Intergovernmental Panel on Climate Change Sixth Assessment Report (IPCC AR6). Each of the climate projections includes daily precipitation and temperature variables for the periods 1950–2014 ("hindcast") and 2015–2100 ("forwardcast"). A bias reduction and correction procedure, described in the following paragraph, was applied to the historical and future climate projections to provide a set of climate change projections at high resolution (0.25°) relative to their native resolution that can be used to assess the impacts of climate change on

*Evolution of Agroclimatic Indicators in Senegal Using CMIP6 Simulations DOI: http://dx.doi.org/10.5772/intechopen.109895*


#### **Table 1.**

*List of CMIP6 models used in this study, modeling centers, horizontal resolution and references.*

processes that are sensitive to smaller-scale climate gradients and the effects of local topography on climate conditions.

## **2.2 Bias correction**

A bias correction procedure with the CDF-t method was applied on each grid point with the GCM data via the historical Global Meteorological Forcing Dataset (GMFD; [28]). For each daily parameter, the algorithm generates the cumulative distribution function (CDF) for the GMFD data and for the retrospective GCM simulations. The corresponding source values (day of year +/− 15 days) are clustered and sorted over the period from 1960 to 2014 at various probability thresholds to produce a quantile map between models and observations. Based on this map, the model values in the different quantiles (e.g., p = 90%) can be translated into corresponding GMFD values for the same quantile. Assuming that the CDF of the simulations is stable during the retrospective and prospective periods, the algorithm simply searches for the probability quantile associated with the predicted climate variations from the estimated CDF of the historical data and then accepts these as the adjusted climate projections. The climate projections adjusted in this way have the same CDF as the GMFD data; therefore, possible biases in the statistical structure (variance, in particular) of the original GCM outputs are eliminated by this procedure. At the end of the bias correction step, the previously extracted parameter climate trends (precipitation and temperature) are added to the fitted model climate fields.
