Click on the >>>PA Downloads Link in the Resources Menu to download these KMZs for free!
Note: In late 2013 we published kmz that extracts the Class B streams based on our analysis of the information described below. If you’re interested in Class B-D streams you will certainly want to download the Class B Streams KMZ as well as the overlays described here.
PFBC publishes detailed lists and vector datasets of wild trout and “Class A” streams. But of Classes B, C, and D there is barely a mention, except in passing. You don’t know how most of the “Trout Natural Reproduction” streams are classified, just the few in Class A. You have to look hard even to find out how the different classes are defined.
Turns out it’s all based on trout “biomass”:
- Class A streams contain more than 40 kg/hecatre* of trout if it’s dominated by brown trout or rainbows, or more than 30 kg/hectare if dominated by brook trout.
- Class B contain less than the 40/30 of a Class A, but more than 20. Class B also becomes the “Class A” in waiting, since it can take a while for a stream to be certified Class A.
- Class C have less than 20 and more than 10.
- Class D have wild trout populations, but somewhere between 0 and 10kg/hectare.
- Class E are cold water streams that have no wild trout, but are suitable for stocking
While PFBC is mysterious about the class designation of specific, listed streams, in the 2010 Strategic Plan it published a small map (a jpeg of only 855 pixels across the entire state of PA and 455 tall) showing these class B-C-D waters. This map had been rendered on a GIS system of some sort, though using a projection not readily compatible with Google Earth’s. Nevertheless, we processed the image in Photoshop to eliminate the background, sliced it into small pieces (often as small as a county or two) so the errors introduced by the different projections would be minimal, and hand registered the images in Google Earth using the county boundaries in the PFBC image.
The results are suprisingly useful.
You can see a state-wide image here. The cyan lines are Class A streams rendered using the detailed vector dataset published by the PFBC. The other streams are from the overlay. Dark blue lines are Class B, green are Class C, and brown (the majority) are Class E. The white fringing is an artifact of the way we eliminated the white background.
It’s really when you zoom in that the overlays become useful. Here we see a detail from Clinton County, where a number of streams run into the West Branch of the Susquehanna River (the large river cutting across the left side of the image). The purple lines are rendered by a KML containing the PFBC’s “Trout Natural Reproduction” vector dataset. The Cyan lines are Class A streams, also rendered from a vector dataset. The brown, green, and blue “halos” are the overlays, with transparency set so you can see both the stream designation and the base aerial imagery of Google Earth. A few observations follow below the picture.
First, given the low resolution of the original image, you wouldn’t expect registration to be “perfect”. There are small streams, for example, that were too small to show up on the original image. However, a surprisingly large number of streams do seem to register pretty accurately. The West Branch, for sure. (interestingly, it’s shown as a Class D stream on the overlay, but not included in the vector dataset of “Trout Natural Reproduction”. You’ll find there are a lot of streams in that situation).
But, for example, most of the Class A streams fall into “gaps” in the overlay, which is exactly how they should fall. And many streams register quite well. The large purple stream to the right of the image is called “Lick Run”. It’s headwaters are pretty clearly Class B, and it turns into a Class C stream as it approaches the West Branch.
Amazing how much information is contained in a 855×455 pixel image.
*For those of you not raised on the Metric system, a kg is 2.2 pounds and a hectare is 2.4 acres. So you can take the biomass numbers, add 10%, and convert it to pounds/acre. Or just use the numbers as they are (10% is no doubt within the margin of error).