Normalized Difference Flow Index (NDFI) Mapping

The Chesapeake Conservancy has already produced two novel, precise, and accurate datasets for conservation purposes; concentrated flow path mapping and high resolution land use/land cover analysis. It was only natural for the Conservancy to seek to combine these datasets to create a unique dataset with unlimited potential for land and water conservation purposes. The Conservancy gave this analysis a new term, the Normalized Difference Flow Index (NDFI), taking its roots from the well known Normalized Difference Vegetative Index (NDVI) used for identifying plant life in satellite imagery.

Although all water flows across a surface before reaching a stream or river, not all surfaces are created equal. Land types such as impervious surfaces (roads, parking lots, buildings) and barren ground more heavily contribute pollutants to waterways due to their inability to filter out contaminants. If you have ever heard of the term, riparian buffer, then you may know that they way plant life interacts with the soil forms a natural pollution filtration system for water runoff. The larger and more proliferic the plant life the better the filtration system, which is why most streams and rivers should have forested areas along their banks.

The high resolution land classification analysis completed by the Conservancy was used to account for the unequal contribution of pollutants to the precise stream channels described in the concentrated flow path mapping analysis. From zero to ten all of the land types depicted in the land use dataset were given a value depending on their contribution of polluted water runoff. Land categories such as impervious surfaces and tilled fields were given high values (10 & 9) while open water, wetlands, and forests were given the lower values (0, 1, & 3).

Subsequently, the detailed flow accumulation channels mapped out using elevation data were then overlaid onto the weighted land classification data to determine which stream channels would be responsible for the highest flow of low quality water. Just because a stream channel had a high amount of flow accumulation wouldn’t necessarily mean it was of more concern than a smaller channel if the larger channel flows through a heavily wooded area while the smaller channel is running off of a tilled field.

The ultimate use of these datasets is to allow landowners and governments the ability to precisely target and guide their resources towards implementing best management practices most efficiently. Instead of putting in an expensive and time consuming buffer alongside an entire corn field, a farmer could analyze his or her parcel and determine an exact location to implement a singular more robust buffer which filters all of the water flowing off of that field. The result is a more effective environmental impact while simultaneously saving valuable time, money, and even viable cropland. The image to the right provides an example of this analysis. There is a stark difference between the management priorities of channels flowing through a forest area versus those flowing off of a tilled field.