Dear Mudavath;

In the following link, you will find a guide to a Choropleth map classification.

http://gisgeography.com/choropleth-maps-data-classification/

Additionally, see the another following link for buffering method in GIS

http://pro.arcgis.com/en/pro-app/tool-reference/analysis/how-buffer-analysis-works.htm.

regards

Hi Mudavath ,

you can follow several concepts:

(1) you can easily use Pearson/Spearman correlation (not of the maps but the data aggregated into the same districts) - but be sure you used paired correlation and have the corresponding area and spatial resolution,

(2) even better you can use local or gwr correlation as defined by Kalogirou in R package called lctools (http://cran.r-project.org/web/packages/lctools/index.html), which gives you local estimates of correlations between pairs of variables in defined neighbourhood,

(3) or you can use e.g. methods for the computing of (local) spatial autocorrelation/association such as Bivariate Moran's I, which can be easily computed in GeoDa or R.

Regards

Lukas

Hi Mudavath,

You may use ArcMap' Spatial Analyst tool for this task.

1). Give each district a unique ID

2). ArcToolbox -> Spatial Analyst Tool -> Zonal -> Zonal Statistics as Table

3) Give appropriate layers and use that district ID as Zone field

4). Select any appropriate statistics, may be mean or maximum elevation

5) The output table will contain mean or maximum elevation value against each district id

6) Now join that table with district shapefile uisng district id, you will get pairs of incidence of disease and elevation for each of the district.

7) Next export attribute table of district shapefile and apply Pearson's correlation on incidence of disease and elevation.

(Optional)

8) You may again join that final table with district shapefile by using district id field to create district level choropleth map based on correlation values.

Cheers!

I found zonal statistics plugin for QGIS. It is also useful in identifying the details of particular polygon in the given map.

Dear Mudavath

When you said elevation map I think you are saying DEM, a grid based layer. My respond is based on that assumption and I will to give you a simple analysis model I think you can perform with QGIS or ArcGIS.

The problem you have here is complex because at first you must find a way to create comparable maps or layers. In one hand you have a

choropleth map in a form of vector data and in the other hand an elevation map in a form of grid data.

Because of different structure and agregation criteria of your layers you must find a way to create comparable maps.

I think the easy way is to make the elevation data comparable to the district divisions and not in the oposite way.

You also must take in account that elevation maps can give you a different kind of information, not only elevation. That could be:

Steepness of relief, orientation of relief, rugosity, diference with maximun an minimun altitude, etc.

If you intersect districts with those previous maps you will have many values of elevation parameters for each district and this is something that it will difficult to correlate. I think you need only one value for each district in order to make the correlation feasible.

In this case you can generate one value using common statistics like average, mode, standard desviation of the parameters you chossed, and then asign this value to each district and look if there are any correlation.

Summarizing

1) create derived models from your elevation map

2) For each district select the area in you elevation models

3) Calculate the statistics of each district area in order to obtain only one number

4) See if there are a correlation between number of disceases and elevation paramteres

5) conclusion

I hope this will help

Cheers Gabriel

Thank you Gabriel Asato.

Spatial Efficiency metric (SPAEF) is proven to be robust when comparing two raster maps. Python and Matlab codes are available at: http://space.geus.dk/tools_products/index.html

Two papers on the metric:

https://www.hydrol-earth-syst-sci.net/22/1299/2018/hess-22-1299-2018-assets.html

https://www.geosci-model-dev-discuss.net/gmd-2017-238/

In a slight different direction, you can Create Random Points and extracts all values (elevation and disease prevalence data) to the points and use the Geostatistical Tool, all in ArcGIS or you can even export the data from the random points created to excel and use different statistical software to analyze your data, e.g. (R, Statistica, or SPSS).

Dear also you can select points and create table using these points from your maps and its values then do regression in excel

regards