Differences in Infrared NAIP Imagery 2017-2019

[Henry]

Hi all,

I've been looking at available sources of high-resolution aerial imagery that includes Infrared bands to better identify a specific plant species (Scotch Broom, Cytisus scoparius) in riparian areas. Typically, we use NAIP as provided by the USDA. It shows up really great in infrared in 2017 and prior, but not nearly as well in the 2019 imagery. See included image, left is 2017, right is 2019.

Left, 2017 NAIP - Right, 2019 NAIP

The band settings as set through the Image symbology are the same, R-4, G-2, B-3. However the resulting colors are very different.

For anyone else that uses the NAIP dataset, is this a result of a different date for image acquisition? Or is it something that could further be manipulated through the image tools to get the same results?

Thanks!

[PeterBecker]

Presume you are looking at the NAIP Services in the Living Atlas.  These provide access to the source NAIP imagery for multiple years and different renderings. There will be differences in the R,G,B,IR values returned for different years not only due to change in dates, but the fact that the images may not be flows at exactly the same time of the year, the fact that different sensors were used and the data values are not calibrated say to surface reflectance. One way to normalize small differences in an areas is to use the DRA (Dynamic Range Adjustment) stretch. Also look to change the renderer to NDVI, you should find that the NDVI values (or colorized NDVI) provides a better distinction of vegetation change.

[Henry]

Thanks for the insight, @PeterBecker! I was looking at a few image services provided by a state open geospatial site, but they appear the same as the NAIP services provided by the Living Atlas.

I will be sure to try the DRA stretch and NDVI renderer.