@Dr. Behnam,

Please see the following RG thread.

Not sure but according to the Rene Gommes`s suggestion, you may find a r package in the below link: https://r-forge.r-project.org/R/?group_id=723

https://www.researchgate.net/post/How_to_interpolate_spatial_climate_data_according_to_DEM#view=537e0439d685cc981f8b4571

If you know exactly the geographical location of the meteo stations, you can create an excel file that include all type of climatic data (precipitation, temperature...etc). Next you can use any interpolation algorithm to create a grid "as a raster format" (in ESRI ARCGIS software or QGIS )

May be Geographically Weighted Regression (GWR) could work.

Actually the outcome of the interpolation and regression methods depends on the number of weather stations. So, any method other than the previously mentioned ones will not give better results unless enough weather stations data exists. In addition to the number of weather stations uncertainty arises from uneven distribution of the weather stations (e.g. concentrated in one place while another is empty). There is one way to model sea surface and land temperatures (create temperature rater data) using thermal bands of satellite and weather data collections and again the reliability of the model depends on the number of weather stations and their distributions. Sometime ANN can reduce uncertainty but this is not guaranteed.

Dear M. Awad

Thanks indeed. Would you please send a paper which help me to use satellite thermal bands to create and validate climatic data?

Dear Ron Mahabir

Thanks for your reply.

Is there a special tool for GWR? Or general regression methods with climatic data as dependent and independent variables should be applied?

I know with versions of ArcGIS 10.0 and upwards there is a GWR tool. There should also be a SAGA extension for QGIS which is free and should have this capability. R is definitely a powerful package as well.

Thanks Ron. Could you send the R package to me or Help me to find it?

I have a paper about predicting sea surface temperature from Landsat ETM+ images you can download it from the list of my papers. I am not the first author but you can correspond with the author for more information. There are many other papers which predict other climate variables

Author(s)

Tao Chen and Takagi, M., Rainfall prediction of geostationary meteorological satellite images using artificial neural network, International Geoscience and Remote Sensing Symposium, 1993. IGARSS '93, vol. 3, pp.1247 - 1249.

Dear friend,

Thank you for your kind help. I will use introduced papers .

@Dr. Behnam,

Please see the following RG thread.

Not sure but according to the Rene Gommes`s suggestion, you may find a r package in the below link: https://r-forge.r-project.org/R/?group_id=723

https://www.researchgate.net/post/How_to_interpolate_spatial_climate_data_according_to_DEM#view=537e0439d685cc981f8b4571