Table 1.

Climate data (rainfall and temperature) for use in malaria impact assessment—advantages and limitations

	Climate information	Source	Advantages	Limitations
Ground observations
Rainfall	Ground observations globally available	GTS*	Freely available. Actual measurements of rainfall. Information highly valuable for impact assessment at the location of the station and for testing other climate products	Represent a very small percentage of the station data available at the national level. May not be quality controlled and data may be missing. Rainfall highly localized but representativeness for a larger area can be improved if data aggregated over space and time. Large regions with very poor or no data.
	Climatology gridded (interpolated) ground observations	WorldClim†	Freely available. Pan African coverage. Information relevant for large-scale impact analysis where climatology is sufficient.	Quality entirely dependent on number, spatial distribution, and quality of ground observations, which varies a lot in space and time.
	Time series gridded (interpolated) ground observations	UEA-CRU‡	Freely available. Pan African coverage. Information relevant for large-scale impact analysis.	Quality entirely dependent on number, spatial distribution, and quality of ground observations, which varies a lot in space and time. Significant decline in data in recent decades. Best used for large-scale national and regional analysis.
		GPCC§
	Ground point observations locally available—hourly to monthly	National archives and monitoring observations of the national meteorological agency.	Relatively high national coverage. The main source of meteorological data obtained using stations established and maintained using WMO criteria. Local knowledge of the data can help with quality control.	Data often restricted by national meteorological agencies—need data access agreement. Data may not be quality controlled and some may be missing. Ask for meta data associated with data files when daily data is aggregated to weekly or monthly data.
Air temperature (minimum and maximum)	Ground observations globally available	GTS	Freely available – information. highly valuable for impact assessment at the location of the station and for calibrating and verifying other climate products	Represent a very small percentage of the station data available at the national level. May not be quality controlled and data may be missing. Temperature varies with orography. Representativeness for a larger area can be improved if elevation and lat/long are incorporated into data aggregated over space and time. Large regions with very poor or no data
	Climatology gridded (interpolated) ground observations for period 1960–1999	WorldClim	Freely available. Pan African coverage. Information relevant for large scale impact analysis where climatology is sufficient.	Quality entirely dependent on number, spatial distribution, and quality of ground observations, which will vary in space and time.
	Time Series ridded (interpolated) ground observations	UEA-CRU	Freely available. Pan African coverage. Information relevant for large scale impact analysis.	Quality entirely dependent on number, spatial distribution, and quality of ground observations, which will vary in space and time. Significant decline in data in recent decades. Best used for large-scale subnational, national, and regional analysis.
	Ground point observations locally available from hourly to monthly	National archives and monitoring observations of the national meteorological agency.	Relatively high national coverage. The main source of meteorological data obtained using stations established and maintained using WMO criteria. Local knowledge of the data can help with quality control.	Data often restricted by national meteorological agencies—need data access agreement. Data may not be quality controlled and some may be missing. Ask for meta data associated with data files when daily data is aggregated to weekly or monthly data.
Reanalysis
Rainfall	Created via a “frozen” data assimilation scheme and model(s), which ingest all available observations every 6–12 hours	ERA-40¶	Freely available, Pan African coverage with high temporal resolution (6–12 hours).	Large spatial scale. Quality of data varies by space and time due to varying data inputs over the years. Very poor representation of rainfall at any scale.
		ERA-Interim‖
		NCEP-DOE**
		MEERA, †† and many more
Temperature (minimum and maximum)	Created via a “frozen” data assimilation scheme and model(s), which ingest all available observations every 6–12 hours	ERA-40	Freely available, Pan African coverage with high temporal resolution (6–12 hours). Good representation of temporal and spatial changes in air temperature, which can be improved by temporal aggregation and combining with ground station data.	Large spatial scale. Quality of data varies in space and time due to varying inputs.
		ERA-Interim
		NCEP-DOE
		MEERA, and many more
Satellite climate data
Rainfall	Weather monitoring satellites with global coverage	Satellite-based rainfall estimates (some combine satellite and limited station data from global archives/GTS)	Freely available. Pan African coverage. Provides a very good approximation of the spatial distribution of rainfall at the country or sub national level. Has high spatial and temporal resolution—for example, 4 km and daily.	Relationship to observed rainfall may vary according to orography and other local characteristics. May not capture rainfall extremes well. Quality dependent on calibration and integration of observed station data.
		RFE2,‡‡ ARC§§ TAMSAT¶¶	Representation of actual rainfall at local scale is best over areas where convective rainfall is dominant
		CMAP‖‖ CHIRPS,*** etc.
LSTs and estimated minimum air temperatures	LSTs derived from thermal sensors	MODIS LST††† during the night	Freely available, Pan African, high spatial resolution (1 km).	Relationship to observed temperature may vary according to land cover and other local characteristics. May not capture air temperature well. Quality dependent on calibration and integration of observed station data.
ENACTS‡‡‡	Blended product rainfall	Combines all quality controlled national station data with best globally available satellite product	High spatial and temporal resolution (4–5 km and daily) for over 30 years with much higher accuracy than other products as it incorporates data from the national observations archive and monitoring data. Suitable for analysis at national, sub-national, and local level. Derived climate products available on national meteorological agency websites	Quality varies according to the number and quality of observations used to calibrate and integrate into data set. ENACTS climate product data may be restricted by national meteorological agencies—need data access agreement.
	Blended product temperature	Combines all quality controlled national station data with best globally available elevation and reanalysis products	High spatial and temporal resolution (4–5 km and 10 daily) for over 30 years with much higher accuracy than other products as it contains the national observations archive and monitoring data. Suitable for analysis at national, sub-national, and local level. Climate products available on national meteorological agency website	Quality varies according to the number and quality of observations used to calibrate and integrate into data set. ENACTS climate product data may be restricted by national meteorological agencies—need data access agreement.

ARC = Africa Rainfall Climatology; CHIRPS = Climate Hazards Group InfraRed Precipitation with Station Data; CMAP = CPC Merged Analysis of Prediction; CRU = Climate Research Unit; DOE = Department of Energy; ENACTS = Enhanced National Climate Services products; GPCC = Global Precipitation Climatology Center; GTS = Global Telecommunications System; LST = Land Surface Temperature; MODIS = Moderate Resolution Imaging Spectroradiometer; NCEP = National Centers for Environmental Prediction; RFE = Rainfall Estimates; UEA= University East Anglia;

^*Available at: www.wmo.int/pages/prog/www/TEM/GTS/index_en.html.

^†Available at: www.worldclim.org.

^‡Available at: www.cru.uea.ac.uk/cru/data/hrg/.

^§Available at: climatedataguide.ucar.edu/climate-data/gpcc-global-precipitation-climatology-center.

^¶Available at: www.cgd.ucar.edu/cas/catalog/reanalysis/ecmwf/era40/sfc_mmeans.html.

^‖Available at: climatedataguide.ucar.edu/climate-data/era-interim.

^**Available at: www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis2.gaussian.html.

^††Available at: http://iridl.ldeo.columbia.edu/expert/SOURCES/.NASA/.GSFC/.MERRA/.

^‡‡Available at: http://www.cpc.ncep.noaa.gov/products/fews/rfe.shtml.

^§§Available at: http://www.cpc.ncep.noaa.gov/products/fews/AFR_CLIM/afr_clim.shtml.

^¶¶Available at: http://www.met.reading.ac.uk/∼tamsat/data/rfe_anom.html.

^‖‖Available at: http://www.cpc.ncep.noaa.gov/products/global_precip/html/wpage.cmap.html.

^***Available at: http://chg.geog.ucsb.edu/data/chirps/.

^†††Available at: http://modis-land.gsfc.nasa.gov/temp.html.

^‡‡‡Available at: http://iri.columbia.edu/resources/enacts/.