Resources for ecological niche modeling: Environmental variables
Fig: BIO1 which represents annual mean temperature, obtained from the MERRAclim repository.
In 2020, together with Koffi Mensah Ahadji-Dabla, A. Townsend Peterson, and others, we published the manuscript: Potential roles of environmental and socio-economic factors in the distribution of insecticide resistance in Anopheles gambiae sensu lato (Culicidae: Diptera) across
Togo, West Africa. Here, we explored if variables from the climatic, landscape, or socio-economic realms were able to explain the presence of insecticide resistance patterns in Anopheles mosquitoes in Togo. Although we didn’t find any particular association, we had the opportunity to explore many variables that were related with the landscape and socio-demographic features of the studied locales, including population density, gross domestic product, vegetation continuous fields, among others. I invite the readers to explore Table 1 of that publication for details on the rasters files used. This table is an interesting example on the amount of information that should be shared when using raster predictors:
Description (e.g., Gridded population of the World version 4 (GPW v4): Population density)
The source of the variables (e.g., NASA-SEDAC )
Spatial resolution (e.g., 2.5 minutes )
Temporal resolution (e.g., 2015)
In the following description you will find an expanded version of that table in terms of sources but not in details since those should be individualized for the particularities of your own study.
Other features of the environment
Altitude. From the plethora of options, I find the following easy to use and to implement, go into the folder RESAMPLE, and select the more appropriate resolution for your study.
Human footprint. A combination of eight variables transformed in an index measuring the impact of humans on Earth.
Animal production and wildlife:
Remote sensing data:
Land-cover and Land-use data. Although they seem similar they refer to a different categorization of the land. Land-cover refers to the intrinsic characteristic of the studied area, a forest, a grassland, a mountain. Land-use refers to the ways the studied area is being studied. For example, an artificial lake, a rice field, a crop land. There is some overlap in this categorization but products are build either as land-use or land-cover data, for example:
The Cropland Data Layer from the NASS CropScape geospatial portal from the USDA shows how different areas of the United States are used for different crops like cotton, sorghum, rye, etc.
On the contrary, we can review the how crop lands, in despite of its use, are distributed across Earth using the Croplands of the world product.
Other products include:
Remote sensing products related with socio-demographic predictors:
There are multiple satellite images available for download and process, with a lot of potential applications. This process has became more achievable thanks to open access tools such as those from the Modis Reprojection Tool and Google Earth Engine (GEE). Using GEE interface will allow you to work with a lot of imagery addressing the climate, surface temperature, vegetation indexes (e.g., NDVI, EVI), forest cover (e.g., vegetation continuous fields), etc.
There are databases already summarizing plenty of other raster products such as the one from NIMBios that you also might want to explore.
Finally, there is a database in Spanish with a lot of variables that are not included in this English post but you might want to check it for more raster options such as cloud covers, oceanic and freshwater variables, vegetation disturbance, among others.