Köppen Climate Classification Maps for Brazil Across Four Consecutive Climatological Normals (1961–1990, 1971–2000, 1981–2010, and 1991–2020)
Description
Dataset of Köppen–Geiger Climate Zone Maps for Brazil (1961–2020) Based on BR-DWGD Gridded Data at 0.1° Spatial Resolution This dataset contains the Köppen–Geiger climate classification maps for the entire Brazilian territory, derived from the Brazilian Daily Weather Gridded Data (BR-DWGD; Xavier et al., 2022) at 0.1° spatial resolution (~11 km). Four consecutive World Meteorological Organization (WMO) Climatological Normals are covered: 1961–1990, 1971–2000, 1981–2010, and 1991–2020. Files included: Koppen_Class_BR_1961-1990.tif — Raster grid of Köppen climate classes for the 1961–1990 normal Koppen_Class_BR_1971-2000.tif — Raster grid for the 1971–2000 normal Koppen_Class_BR_1981-2010.tif — Raster grid for the 1981–2010 normal Koppen_Class_BR_1991-2020.tif — Raster grid for the 1991–2020 normal Koppen_Climate_BR.gpkg — GeoPackage with vectorized climate zones, including a color scheme for cartographic visualization Technical specifications: All rasters share the same grid (391 × 393 pixels), coordinate reference system (EPSG:4326, WGS84), and spatial extent (73.90°W–34.80°W; 33.90°S–5.40°N). Pixel values are integer codes representing Köppen classes: 11 (Af), 12 (Am), 13 (As), 14 (Aw), 21 (BSh), 23 (BWh), 31 (Cfa), 32 (Cfb), 34 (Cwa), 35 (Cwb). The BWh class (code 23) is absent in the three earlier normals and emerges exclusively in the 1991–2020 period, representing the unprecedented onset of hot desert conditions in northeastern Brazil. Source data: Monthly climatologies of mean air temperature and total precipitation were computed from daily BR-DWGD gridded data and used to apply the hierarchical Köppen algorithm following Peel et al. (2007) and Alvares et al. (2013).
Files
Steps to reproduce
The Köppen–Geiger climate classification maps were generated from the Brazilian Daily Weather Gridded Data (BR-DWGD; Xavier et al., 2022), a gridded meteorological dataset covering Brazil at 0.1° spatial resolution (~11 km), built from the interpolation of data from 11,473 rain gauges and 1,252 meteorological stations. Daily data on maximum (Tmax) and minimum (Tmin) air temperature and total precipitation (pr) were obtained for the full period 1961–2020. 1. Data preprocessing Daily mean air temperature (Tmean) was computed as the arithmetic mean of Tmax and Tmin. From daily values, monthly climatologies were calculated for each of the four WMO Climatological Normals: 1961–1990, 1971–2000, 1981–2010, and 1991–2020. Precipitation was aggregated as monthly totals; temperature variables were aggregated as monthly means. 2. Derived variables For each normal and each grid pixel, the following variables were extracted: mean annual temperature (Tma); mean annual precipitation (Pma); mean temperature of the warmest month (Tmmq) and coldest month (Tmmf); precipitation of the driest month (Pmms); number of months with Tmean > 10°C (Tm10); and seasonal precipitation extremes (driest and wettest months in summer and winter hemispheric semesters), following Alvares et al. (2013). 3. Aridity threshold The criterion was established as follows: if 70% or more of the mean annual precipitation (Pma) occurs during the winter semester, the threshold is calculated by the equation Plimite = 2∙Tma; if 70% of the annual precipitation is concentrated in the summer semester, the adjusted equation Plimite = 2∙Tma + 28 is applied; and Plimite = 2·Tma + 14 in all other cases (Peel et al. 2007). 4. Köppen classification algorithm The hierarchical Köppen algorithm was applied pixel by pixel in Python 3.13, assigning each grid cell to its major group (A, B, C) and subsequent subtype based on the thresholds defined in Peel et al. (2007) and Alvares et al. (2013). The resulting class codes are: 11 (Af), 12 (Am), 13 (As), 14 (Aw), 21 (BSh), 23 (BWh), 31 (Cfa), 32 (Cfb), 34 (Cwa), 35 (Cwb). 5. Output generation Classified grids were exported as GeoTIFF (EPSG:4326, float32, NaN as NoData) for each climatological normal. Climate zones were additionally vectorized and exported as a GeoPackage (Koppen_Climate_BR_color.gpkg) with an associated color scheme for cartographic use.
Institutions
- Universidade Federal do Espírito SantoEspírito Santo, Vitória