Yesterday I was running quality control on a series of conservation maps produced by the Litchfield Hills Greenprint. I found something that on the landscape that conflicted with our natural resource data, something a landowner had done in the last half dozen years that converted a wetland to a farm field. I though I'd share how I made that discovery, although I'm not yet sure if there is anything to be done about it.
We maintain data for three overarching resources of regional conservation interest at the Greenprint - Prime Agricultural Land, Large Interior Forests, and Drinking Water Resources - and have developed data layers from a variety of sources to identify and analyze in a Geographic Information System or GIS. Our forest resource layer includes data which were derived from satellite images, in which land cover is classified by which of 13 possible types is best represented in each 100' x 100' pixel.
Our farmland layer includes the use of data derived from aerial photographs of actual farm fields which were then recorded digitally. I found a place where both of these layers - farm fields and forests - overlapped, which should not have been the case as we excluded farmland pixels identified by satellite from the forest resource layer.
Even more intriguingly, the farm field in the forest did not show up on our prime farmland layer, which intersects the digital farm fields from aerial photographs with the State of Connecticut's Prime Farm Soils and Farm Soils of Statewide Importance data. There is a high correlation between the good soils and existing agricultural fields in our region, yet when we turned on the layer that showed just the farm fields, there it was, jutting out into the forest and almost entirely outside of the area with agricultural soils.
Now the forest layer does not only include forest. It includes 4 types of land cover - Deciduous Forest, Coniferous Forest, Forested Wetland, and Non-forested Wetland. It does not consider areas of open water, and any heavily modified land cover types the satellites identified nearby - barren land, agricultural fields, turf and grass, developed land and power lines - got buffered by 300 feet and excluded from the forest analysis. We kept the non-forested wetland in the mix because it is usually a small component of the forest and we were looking for areas of forest interior greater than 200 acres. If we buffered the non-forested wetlands, we might have excluded the forest habitat we were seeking by forcing it below the 200-acre threshold. I'm now wondering whether we should treat it as we did open water, and leave it out of the analysis altogether.
It was clear that what a satellite told us was forest or wetland was in conflict with what an aerial photograph said was a farm field. So we did what everybody does and went to Google Earth. And there it was: a farm field on wetland soils next to a forested wetland that used to be bigger. This mowed field was so dark and muddy it was clearly still wet. We never would have looked for it if our two data layers from two different sources had not been in conflict.
As for what to do about it, that is harder. Farmers get a good deal of dispensation to manage their land "as of right", but even they need to go through a formal approval process to convert a wetland to agriculture. There are also state rare species records in the associated forested wetlands, but very few people at the State to monitor them. Connecticut identifies jurisdictional wetlands by soils, not by what grows there, and on the surface, at least, this area seems to qualify. A satellite in 2002 thought it was in natural cover, and a plane that flew over in 2006 shows the field there instead.
So maybe the landowner went before the Inland Wetland Commission and got approval to convert the land to fields. I can just as easily imagine that no permit process was followed, or if it was that it was pro forma only. Aerial monitoring of conservation land is becoming more widely used, though, and anyone with a computer can use Google Earth. A word to the wise: I am not the only one who may be watching.