## Water Tool

Download the current version here

### Summary:

Calculates distance-decayed effects of upstream land cover.

### Description:

Extracts NLCD 2006 and SSURGO Soil Hydrologic Group data for a selected catchment in North Carolina to construct a SCS Curve Number raster. Higher curve number values indicated a larger proportion of rainfall going to runoff rather than being infiltrated, stored or otherwise prevented from flowing down slope.

The inverse of the curve number [(100-CN)/10], or "flow cost" is used as a proxy for amount of runoff lost by passing across that cell. Higher "flow cost" values suggest that more runoff stays in the cell meaning less pollution dissolved or suspended in the runoff will depart the cell into its downstream neighbor.

Starting at the stream and moving upwards along various flow paths, accumulated flow costs are tallied (using a weighted flow length tool) to produce a raster dataset representing the summed runoff loss along the flow path. For example two pixels, "A" and "B" each reside 5 cell lengths along a flow path to a stream. Pixel "A"'s flow path crosses cells with low flow costs (i.e. high curve numbers), while "B"'s flow path crosses cells with high values. "A"'s value in the above calculation would be lower than "B"s, since the value is determined by summing up all the cost values between the stream and the pixel along the flow path. We therefore interpret that "A" is effectively closer to the stream than "B" (even though they are the same physical distance from the stream) since more runoff from "A" reaches the stream than from "B".

The result from the weighted flow length tool is then applied to a distance decay function. First, a decay coefficient (`k`) is calculated based on a user provided distance threshold (`d`), such that only 1% of the original runoff amount remains at distance (`d`). This is determined by dividing the log of 0.01 by (`d`):

` k = log(0.01) / d`

The decay coefficient is then multiplied to the result of the weighted flow length tool (above), and the exponent of the result thus represents the decayed distance from the stream, weighted by the summed runoff cost:

`decayed distance = exp(weighted flow length * k)`

This decayed distance result estimates, for each pixel location, the relative contribution of runoff/pollution to the stream, with locations "near" the stream (in terms of curve number generated runoff contribution, not just flow distance) contributing exponentially more runoff/pollution than pixels further away.

The decayed distance result can be multiplied by binary rasters of developed land (developed = 1; everything else = 0) to produce a map of development impacts *in terms of stream impact*.

Additionally, the above map (the product of binary development and decayed distance) can be used as a weight in a flow accumulation analysis to tally the cumulative, weighted development impact for any downstream location.