Select to view content in your preferred language

Performing Upslope Analysis for a large number of points

226
4
a month ago
Labels (2)
AlecWalker1
New Contributor

I am attempting to perform a landslide analysis of a large area of the UK (the north york moors) and am struggling to collect a parameter for each point I am considering (a 10mx10m resolution).

I require the proportion of the upslope area with slope angle >23 degrees for each point. I have completed the required slope analysis with a raster output but am struggling with finding how to complete the next stage of analysis interms of upslope area. Any recommendations of a procedure I could use to undertake this?

Many thanks for any suggestions.

0 Kudos
4 Replies
BobBooth1
Esri Contributor

You could use the Watershed tool to identify the upslope cells for each point.

https://pro.arcgis.com/en/pro-app/latest/tool-reference/spatial-analyst/watershed.htm

Use it in loop and make a separate output for each point (use Python or ModelBuilder to automate this, naming each output using a unique value associated with the given study point).

Use Con to determine, for all cells, which ones are greater than 23 degrees slope and code them as 1.

https://pro.arcgis.com/en/pro-app/latest/tool-reference/spatial-analyst/con-.htm

Then use the output upslope watershed for each point to summarize the number of cells in that watershed that have have a value of 1.

Could probably use Zonal Statistics for that:

https://pro.arcgis.com/en/pro-app/latest/help/analysis/raster-functions/zonal-statistics-global-func...

https://gis.stackexchange.com/questions/142412/counting-number-of-raster-cells-satisfying-a-criteria...

 

 

0 Kudos
AlecWalker1
New Contributor

Amazing, thank you for the advice I'll see what I can do!

0 Kudos
AlecWalker1
New Contributor

Do you think this would be viable for a 32 million point dataset?

0 Kudos
BobBooth1
Esri Contributor

Well, as I said, you will need to do some automation.

It may also be useful to break up the input points into chunks in different feature classes to process.

0 Kudos