What do you mean by creating a raster of Erosivity from precipitation?

I am using USLE for predict the soil loss in a area, but in my country (Peru)  no data of erosivity (R) of precipitation is available. I want to estimate the R factor from data available like, climate data (Precipitation). I know that some equations can use for that propose. I want the more accuracy equation for my area. Or a method for make a equation for my area. Thank you Very Much

I think that it can help you:

http://www.hindawi.com/journals/ija/2010/365249/

+

http://wiki.dpi.inpe.br/lib/exe/fetch.php?media=trabalho_completo_hugo.pdf (Methodology).

In the first article you will find a methodology to estimate the erosivity using precipitation values. In the second article, the methodology to build raster using TRMM precipitation images linked to models. The example is aplied to plant disease prediction models, but you can adapted to other aplications.

In this one (http://www.geoinfo.info/proceedings_geoinfo2014.split/Paper10-S-p32.pdf), you will find an idea about how to use this models linked to enviromental data, in real time, using the TerraMA2. That is a new solution developed by INPE.

All this papers are on my research gate page.

If you need more details, feel free to contact me ([email protected]),

Best regards,

Hugo

The publication propossing several aproaches to do what you wanta, estimating rainfall erosivity, the R factor for USLE/RUSLE, is the following:

Renard & Freimund (1994) [J. of Hydrology 157, 1994, pgs.: 287-306]

Here you have several eqs., even one using only annual precipitation, but the most used one is eq. 14, that needs annual and monthly precipitation.

After having estimated values in different places with their coordinates, there are GIS procedures as the indicated by Hugo.

Nevertheless, you must be awere that in mountain environments, erosion is caused in an important proportion by processes not taken in acount by USLE/RUSLE like mass movements and extensive gullies. Unless aplieded to small areas (plots) where these processes are not important, the estimates of USLE/RUSLE are going to be less than the real ones.

IF you have high temporal resolution data (10 minutes, 30 minutes, hourly), then you can use our proposal:

http://www.sciencedirect.com/science/article/pii/S004896971500011X

Otherwise you have to apply a function for transforming precipitation data into R-factor.

My answer to the colleague from Ecuador holds for the present question. In that answer I referred to the classic paper of Renard and Freymund in the Journal of Hydrology, indicating the most used and recommended  equation to transform precipitation data into R factor of USLE and RUSLE.

By searching for a suitable mechanism

You can download global R data here: https://esdac.jrc.ec.europa.eu/content/global-rainfall-erosivity