Sensors and Systems
Breaking News
BARSC Welcomes ICEYE as New Member
Rating12345The British Association of Remote Sensing Companies (BARSC), the industry...
Citadel Defense Delivers on Urgent Contract for Counter Drone Systems
Rating12345SAN DIEGO – Citadel Defense, the manufacturer of innovative, high-performance...
HERE and Cerence bring more personalized, contextualized in-car experiences to drivers worldwide
Rating12345Amsterdam and Burlington, Mass. – Cerence Inc., the global...

April 3rd, 2009
Soil Data Goes Global

image
  • Rating12345

hempel_jon_thumb.jpgThe global soil mapping effort aims to update and catalog the world’s soil information accurately within one clearinghouse called GlobalSoilMap.net. V1 Editor Matt Ball spoke with the North American coordinator of this effort, Jon Hempel, Co-Director-National Geospatial Development Center, Soil Business Area Analysis Chair at the Natural Resources Conservation Service (NRCS) in Morgantown, West Virginia.

hempel_jon.jpgThe global soil mapping effort aims to update and catalog the world’s soil information accurately within one clearinghouse called GlobalSoilMap.net. V1 Editor Matt Ball spoke with the North American coordinator of this effort, Jon Hempel, Co-Director-National Geospatial Development Center, Soil Business Area Analysis Chair at the Natural Resources Conservation Service (NRCS) in Morgantown, West Virginia.

V1: I was very interested to learn that your soil mapping research group has been around since just 2004. Had there been any other effort on national geospatial soil analysis before then?

Hempel:
This is a new center that was developed with funding from an earmark by Senator Robert Byrd. It’s a partnership between the Natural Resources Conversation Service and West Virginia University. One of the primary goals of the Center is to research and develop technologies that will enable the National Cooperative Soil Survey (NCSS) to take full advantage of the potentials of geospatial technology to provide the most advanced and relevant soil information.  Prior to 2004, research in this area was not well coordinated between the NCSS partners.  We now have a coordinated agenda that allows us to move forward in lock-step to understand how best to produce and provide data.  We have research projects with University of Arizona, Oregon State, Washington State University, Penn State University,  Utah State University, Purdue University University of Wisconsin and the University of Florida.

V1:
What’s the status of soil mapping in the United States? Given our agricultural background, I imagine that there’s been work done for quite some time.

Hempel:
I believe about 90% of the nation has detailed soil-mapping completed. There are a number of different websites you can go to see this. The most popular one is called the Web Soil Survey (http://websoilsurvey.nrcs.usda.gov) and this is where you can locate and zoom into areas around the country and get data and interpretations for the soil information of that particular area.

The existing information is built on what is  called a vector-data model, and you’re probably familiar with that, so it’s points, lines, and polygons. It’s displayed as such, and the information attached to each of the polygons is an aggregated data set, so there’s one data set that represents the entire polygon. Now, we know that there’s more soil information than just that one data set in the polygon, but, in the vector model, that’s how  the information is presented.

Global-soil map.net will be displayed as a 90-meter raster grid, with each grid cell having its own specific data. With each cell having its own specific data, the raster presentation allows for data to be displayed on a natural continuum, as small nuances in the landscape changes, the soil data adjusts to those variations.  Vector data only changes at the edges of polygons, polygons can be many hectacres in size.  The other big difference is that the global soil project will display soil properties – rather than  a mapunit aggregated value, we have individual soil properties attached to each raster.

V1: With the raster data, can you do more granular analysis?

Hempel: Most  natural resource data that’s out there — climate, geology, relief or elevation information — is all in a raster format, so this raster soil-property information set blends in from a GIS standpoint really well which will make analysis more congruent.

We can rasterize the vector information, but it there is no more  gain  in the detail of the information. Rasterizing vector information at a certain resolution, say 30 meters, each cell within a given polygon will still display only the  aggregated data value for the polygon., The  continuous nature of the landscape is not displayed within a polygon. That’s what we’re trying to get out with the raster soil information, to show how properties vary as the land surface changes. If you have differences in slope or differences in convexity or concavity or landscape position, the soil properties will adjust as land surface changes on a much higher resolution than what our current displays allows.

V1: Do you need to re-visit locations to get back the data that was aggregated previously in vector format?

Hempel:
There won’t be funding to go out and do any new sampling for the global soil map, for that reason, we plan  to use legacy information for producing the grid data  across North America. That means we’ll assemble  all of our point data  across the continent.  Canda, Mexicon and the US have a wealth of point data available for this project. We’ll use disaggregation techniques to better understand the detail of our polygon maps and refine this data with extrapolation techniques that ulitize the point data.  The analysis  of the point-data and the existing legacy soil-maps along with the co-variate data will enable us to provide cell by cell soil-property information.

V1: A 90-meter raster sounds rather small, given the amount of land in North America.

Hempel: This is really a high-resolution data set because most of the existing information in other natural resource fields is at a kilometer per pixel, and you can fit many of 90 meter cells in a  square kilometer. Across the globe’s land surface, there are around  18.5 billion cells at 90 meters, so there’s a lot of information.

V1:
Do you have the infrastructure there in terms of servers and data storage?

Hempel: We won’t store the entire worlds’ data-set, we’ll only store North America. We have the capacity to store some of that right now, but as we start building this information, we’ll probably have to upgrade our system to store it all. There are currently six nodes that will be working in different parts of the globe, and each node will house their own data. There will also be one centralized location that will build an application that will reach out to each of the nodes and pull the information as it’s requested. So there won’t have to be one source of all 18.5 billion pixels.

V1: It’s interesting that the global soil map will start with the Africa soils with funding from the Bill Gates foundation.

Hempel: It kind of makes sense because Africa has the greatest need for this information, and they also have the least amount of information of any of other place on the globe, plus the foundation centers their work in that part of the world.

V1:
Does soil information correlate closely with agricultural history?

Hempel:
What has traditionally driven the need for soil survey information is agriculture, but that’s changing , and we’ve got many other users that are really interested in soils information. The climate change people are real curious about this project, because soils have a big influence on climate change with either the ability to sequester carbon or emit carbon  depending on how the soil is used.

There are other environmental groups that have  interests in soil information as well. One of our biggest requests for soil data right now are people that are looking at urban development as it relates to smart growth and trying to gauge where they should move their development. They are fairly new users — probably in the last twenty years they’ve beomce interested in soils and how the information  can influence growth.

V1: That’s curious.  Do they want to place development where the soil isn’t good for agriculture?

Hempel: That’s kind of the idea, to leave the best agricultural grounds as agriculture or green space. There are a lot of communities across the U. S. that are looking at smart growth initiatives. They’re trying to steer away from converting the best farmland and try to go to areas that aren’t as valuable from an agricultural standpoint.

People have caught on, and the ease of access of soil survey data on the web makes that data even more usable and valuable for urban planners. It used to be that we would have environmental consulting firms from all across the country that would be contracted to do urban planning work in different parts of the country. They’d have to mail around hard copy information, and I can remember requests where large firms would ask for a copy of each soil  survey in the whole country.” There are on the order of 3,000 different soil survey books in the US. Now with the web soil survey there’s no need for that at all. Clients can now navigate to the web site and zoom  to any place where we have current information, and view or download the information.

V1: How many different data providers are there? Are the counties the primary source of information?

Hempel:
The current structure of the soil survey is that they’re produced in a county by county fashion. For the web soil survey, they’ve stitched all those county data sets together into one large data set. But, if you’re interested in soil information to use in a GIS, you would download one county at a time.

There’s several different websites that host the  county soil survey information — one is what they call the soil datamart (http://soildatamart.nrcs.usda.gov/) and then the other is the resource data gateway (http://datagateway.nrcs.usda.gov/). The data gateway has vast collectin  of resource information beyond soil information,  elevation information, road information, infrastructure information, land cover information are all available on this site

V1:
Is soil conservation a big push as well of this effort?

Hempel: There’s a whole side of  NRCS that deals with soil conservation, and it’s a huge program. NRCS is the agency that provides conservation assistance to private landowners, so we work with private landowners across the country in helping them with their conservation needs.

We’re structured so that nearly every county  across the country has a NRCS office.  The office is typically in a cooperative partnership with the local soil and water conservation district. They work with landowners and agricultural producers on soil conversation issues. The office is staffed by soil conservationists, and typically both the state and federal agencies partner with us. It’s not unusual to have federal, state and county employees in these offices.

We have a nice grassroots network to work on conservation. In fact, it is said that NRCS is the largest conservation agency in the world, and that’s probably true. It’s really pretty unique how we are structured to support the grassroots’ initiatives, so we work closely with the local counties or local government to provide support to solve their conservation issues.

V1:
So you’ve likely solved a lot of local, regional and federal data issues that still occur, such as data redundancy and incompatibility? I would imagine because you’re working so closely with one another that these issues could be avoided.

Hempel: I think so. There are a lot of checks and balances in the system to see how we’re doing. It’s a huge program with billions of dollars a year put into the conservation efforts that NRCS does across the nation. NRCS works in all areas of natural resource conservation — soil conservation, wildlife conservation, forestry, water quality.

We have expertise in all those areas for private land owners, and in contrast to that there are federal agencies that do the same thing on federal land — Forest Service, Bureau of Land Management, Park Service. They manage their federal portion of the land area, but NRCS is responsible for the privately held land in the country. area.

V1: So, do you interface with the Bureau of Land Management and others to share data?

Hempel: We have what’s called a National Cooperative Soil Survey, so we try to work together with the other federal entities that are doing soils work. We also include private consultants in the partnership, and we work with our research arm as well as with land grant universities.

V1: You mentioned that climate scientists are very interested in the kind of data that you have. Do you have much interface with them in terms of the kinds of information that they’re gleaning from your data?

Hempel:
I’m not the best source because I’m not a climatologist, but  our data could be used in  model projections to look at how climate change affects cropping systems, for example. Under projected climate change, soil information is needed to optimize location of crops, irrigation and conservation systems from areas that are drying to areas that are becoming wetter. 

As  areas trend toward more precipitation this will have to be a consideration for management of the soil and  what that means for conservation efforts as we try to mitigate increases in  flooding, nutrient depletion, land degradation due to increased erosion. If there’s more rain than there used to be, we have to look at how to offset that.

The global product is so interesting because we’re providing a dataset that’s contiguous across the whole globe. Right now, the information is very disjointed and it’s hard for the modelers and other users to use information consistently from one country to the next.

V1: It seems like it may be one of the first standardized global geodata sets.

Hempel: I don’t know that for sure, but I think it might be one of the first attempts to provide a global dataset. There’s climatic information that spans the globe, but, as far as a terrestrial data-set, I think this is first attempt to do this and to make it contiguous. There’s geological information that is available across the globe, but not in a harmonized fashion, so what we’re trying to do is provide the same information, the same resolution across the globe and it’s no small undertaking.

One of the big challenges of this is that in many countries, the standards of how  information is produced is a little different. We have different taxonomic classifications systems. For example, in Mexico, the U.S. and Canada, how we describe the soil is different. So somehow you’ve got to understand those three systems and be able to harmonize them together to make data consistent.

It takes some thought and research to understand how to bring the systems together, but I think that’s why people are so excited about this  because we  will  a dataset that is produced from the same standard. We’ve been working on data-harmonization for some time in many different places. Europe has been doing a lot of this work because their thirty or forty countries have produced soils information at a variety scales, at a variety of vintages. They’ve been working hard for a number of years to try to understand how to bring all that information into one contiguous dataset.

V1: I’m really excited about the level work that’s being done and the amount of global effort on this. Are there symposia and conferences and meetings that are taking place to kind of pull together things globally?

Hempel: There are. We just had big meetings on the whole concept. One was in Nairobi where they did the launch for the African portion of the project, and then just a few weeks ago, the Earth Institute at Columbia University hosted a  launch for the rest of the globe in New York City.  In addition, the International Union of Soil Science host a bi-yearly meeting on digital soil mapping technology.  There have been meeting in France, Brazil and the US.  The global consortium which is leading the GlobalSoilMap.net project meets bi-yearly to discuss goals and progress of the project.

V1: I appreciate the focus on sustainability as our coverage area is how geospatial technology can aid the stewardship of our planet.

Hempel: Soils have become a key component of any discussion on sustainability. It makes a lot of sense, because soils are key to just about everything else that happens from climate change — our ability to produce food, our ability to have good quality water. Soils are a key component in the sustainability of all those things.

Leave a Reply

Your email address will not be published. Required fields are marked *