Camargo, Köppen and Thornthwaite climate classification systems in defining climatical regions of the state of São Paulo, Brazil

Climate classification systems ( CCS ) are important tools to support a sustainable agriculture providing information about basic and general characteristics of a regional climate. The Köppen CCS ( KGT ) is considered useful mainly in macroscale and Thornthwaite CCS ( TH ), with a more complex symbology, in mesoscale for climatic determinations. Camargo (1991) ( CAM ) proposed a new CCS trying to be as simple as Köppen and robust as TH nevertheless with a simple symbology. In the state of São Paulo, it is known that regions with distinct availability of seasonal and annual rainfall are classified similarly by KGT , bringing erroneous information to the agricultural credit system, sowing times and yield forecasts. A comparison of different CCS raises a better understanding of a regional climate in different spatial scales and of CCS themselves. Aiming to characterize the climate and the performance of the CAM the KGT , CAM and TH were applied in 334 locations in the state of São Paulo, Brazil. The most common CAM climates in the state were subtropical humid with dry winter ( ST‐UMi ) in the Centre and East region of the state, Tropical humid with dry winter ( TR‐UMi ) and Tropical subhumid with dry winter ( TR‐SBi ) in the north and west region, respectively. The total numbers of climates were 7, 19 and 26 from KGT , CAM and TH , respectively. The KGT had the ability in separating climates in macroscale, and CAM and TH in mesoscale. There was a tendency of the three CCS in separating warm and dry from cold and wet regions. The TH and CAM had the same performance (number of climates, NOC ) in the extremes regions, in the hotter and dry (northern plateau) and the colder (mountainous regions, altitude > 900 m) corresponding to 52% of the state (178 locations). In the other areas TH had the double NOC than CAM .