I needed a soil map of an area of 6500 sq km. I only have the topographical maps of the area. What methods are available to predict or derive soil properties of such an area and how do I evaluate the degree of accuracy?
It depend on the properties you are interested in. There are a few basic conceptual ways to get where you need to be. First you need a digital elevation model. Form a DEM you can classify landforms and make some general statement from that. You could do a wetness index from the DEM to infer wetness. After that, you have two basic choices, dig holes and relate soils to landforms or apply someones model to the spatial data (DEM) you have. Search Geoderma, SSSA, and others for soil landscape modeling papers until you find something in a similar climate - region that is useable based on the technology at your disposal. Start by checking out the global soil mapping project. http://www.globalsoilmap.net/
Agree with the above. But please do not forget land use and pollution maps. Also, if you are in US, search maps by US Geological Survey.
You cannot create soil map from DEM.It just as saying that you want to create soil map from contour maps.DEM are elevation data . It is geological/geomorphological maps that you need in this case, to derive soil map from the geological/geomorphological properties and how DEM/DTM comes in is when you need the height of the soil classes, this can be done by Triangular Irregular Network(TIN) which can be obtained in ARCGIS by converting contour lines to TIN and but can be overlain on .the geological/geomorphological maps
I have to disagree with that post to some extent. I can think of several ways to map without geological/geomorphological maps. The basis in digital soil mapping is the DEM or DTM. You could convert the contours to points and interpolate the points to a surface with the z values being elevation. That surface raster would be a dem. I would look for free elevation data. The scale you need is related to the scale you want to map. Back to my original comments, what you want the map for and what resources you have to generate the map as well as your accuracy standards will determine how you make it. So there is no canned you do it like this answer.
Remote Sensing (RS) and Geographic Information Systems (GIS) are powerful tools in predicting soil properties and climatological modelling. An empirical methodology to prepare digital data set of land use/land cover, soil and digital elevation model (DEM) using RS and GIS techniques is possible. Standard false color composite (Std FCC) bands (green, red and near Infrared bands) of LANDSAT-7, ETM+ sensor can used to produce land use/land cover dataset. DEM can be built using the contours and spot heights which are collected from Topographical maps of the region. With the help of soil region map of an area the soil texture dataset can be built. All the data sets are converted to the spatial resolution of 1km and 5km, which can be used as the independent variables for climatological modelling (specifically air temperature modelling) of the study area.
1. Please refere websites: www.tropecol.com/pdf/open/PDF_43_1/43106.pdf
1. www.indjst.org/archive/vol.4.issue.6/june11sailesh-8.pdf
2. www.indjst.org/archive/vol.4.issue.6/june11sailesh-8.pdf
3. soils.usda.gov/survey/geography/statsgo/description.html
Thanks for your contributions so far. I wish to agree with Michael here. Few research works have demonstrated the possibilty of creating a soil map form DEM. A typical example is the work by Feras M. Ziadat in Geoderma 140 (2007) 73–80. My worry about using DEM for creating soil map lies on the level of accuracy. What methods are availble to ensure high degree of accuracy of predicted soil properties or map from DEM. Which softare or programme would be appropriate for an area of 6500 sq km.
Thanks Selvam for this information. @Riazuddin Ahmed , the links are very useful. thanks. I als appreciate the contribution of Yemi and Michael Harman
- Badges
- Science topic
- Similar topics
- Agricultural Science
- Soil Science
- Soil Analysis