Automated OSD lookup and display

 
UPDATED 2013-04-08


This functionality it now available in the soilDB and sharpshootR packages. All code on this page is now superseded by the fetchOSD() and SoilTaxonomyDendrogram() functions.

 
UPDATED 2012-11-07

I have been thinking about a URL-based interface to basic Official Soil Series Description (OSD) data for a while now... something that when fed a URL, would return CSV-formatted records to the calling process. These type of interfaces can later be used to support more complicated systems, such as our smartphone interface to SoilWeb. URLs can be accessed like files in R, making it possible to do something like this:

R provides several frameworks for composing figures. Base graphics is the simplest, grid is more advanced, and the lattice/ggplot packages provide convenient abstractions of the grid graphics system. Multi-element figures can be readily created in base graphics using either par() or layout(), with analogous functions available in grid. Mixing the two systems is a little more complicated, somewhat fragile, but entirely possible.

I recently needed to combine base+grid graphics on a single page; two sets of base graphics on the left, and the output from xyplot() (grid) on the right. Following some tips from Paul Murrell posted on r-help, the solution was fairly simple. A truncated example of the processes is listed below, corresponding to the attached figure. While the result was "good enough" for a quick summary, there are clearly some improvements that would make figure more useful.

Mehrten Soil Summary

A map depicting the spatial distribution of a given soil series can be very useful when working on a new soil survey, updating an old one, or searching for specific soil characteristics. We have recently added a soil series query facility to SoilWeb, where results are returned in the form of a KML file. Two modes are currently supported:

1. map unit based querying- only map units named for the given soil series are returned
2. component based querying- map units containing components named for the given series are returned

For example, if someone was interested in the spatial distribution of the Amador soil series, they could use the Series Extent Mapping tool to get a quick description of which survey areas contain (and how many corresponding acres of) this series. For an even more detailed description of where the Amador series is mapped, one could use our new soil series query like this:

http://casoilresource.lawr.ucdavis.edu/soil_web/reflector_api/soils.php?what=soil_series_extent&q_string=amador

 
This is a preliminary version, and a subsequent post will contain links to a Google Earth file that can be used to simplify the query process. In most cases queries take about 1-5 seconds, which is quite fast considering: 1) either 730k component names or 275k map unit names are searched, 2) 35 million map unit polygons are filtered for the series in question, and, 3) bounding boxes for matching polygons are merged together-- all on-the-fly. Full text searches for map unit/component names are very fast thanks to advanced text indexing and searching algorithms implemented in PostgreSQL and spatial processing functions implemented in PostGIS. In the final version, the location of the official series description (OSD) will be included in query results.

Attached at the bottom of the page is a KMZ demo showing sample output from the two query modes. Screen shots from the demo are posted below.

Soil Series Query Results: Amador Series: blue regions: map units dominated by the Amador series; red regions: map units that contain at least one component of the Amador series.

It turns out that you can generate a quasi-numerical distance between soil profiles classified according to Soil Taxonomy (or any other hierarchical system) using Gower's generalized dissimilarity metric. For example, taxonomic distances computed from subgroup membership are based on the number of matches at the order, suborder, greatgroup, and subgroup level. This approach allows for the derivation of a quasi-numerical classification system from Soil Taxonomy, but it is severly limited by the fact that each split in the hierarchy is given equal weight. In other words, the quasi-numerical dissimilarity associated with divergence at the soil order level is identical to that associated with divergence at the subgroup level. Clearly this is not ideal.

Gower's generalized dissimilarity metric is conveniently implemented in the cluster package for R. I have posted some related material in the past, but left out some of the details regarding which clustering algorithms produce the most useful dendrograms. Divisive clustering best represents the step-wise splits within the hierarchy of Soil Taxonomy, as expressed in terms of pair-wise dissimilarities. Code examples are below, along with the data used to generate the figure of California subgroups. Discontinuities in figure below are caused by errors in the underlying data, e.g. mis-matches in soil order vs. suborder membership.

Subgroups from California

Western Fresno Soil Hierarchy: partial view of the hierarchy within the US Soil Taxonomic system

Maybe this is just craziness, but wouldn't be neat to have an XML formatted version of the Keys to Soil Taxonomy? The format might look something like the following code snippet, although there may be more efficient uses of XML... The only problem I can see is that it would take a hell of a long time to type in the entire 300+ page document. A complete document of this nature would support all kinds of new and creative uses for the 'keys-- electronic look-up, automated generation of a PDA-ready version, an awesome teaching tool, or just something that could be used to generate cool figures. Anyone know of a quick way to get this put together, or of any similar document that has already been published? Anyone want to help type-in the data?