Ho Hum

It's been nice not having to map any active fires since October, 2009, when I was asked to map the Sheep Fire in California for the BAER team. After looking over my data for 2009, I noticed that I mapped my first fire (in CONUS) in March and my last in October. In reality, 2009 was "Driving Miss Daisy" when compared to 2006 - 2008's "Fast and Furious."

During the off season, I focus most my time on MTBS (Monitoring Trends in Burn Severity). I wish MTBS wasn't so huge because I rarely get time to pursue some research and applications ideas pertinent to BAER. For example, I've been trying to find ways to help BAER teams and Forests fulfil their responsibility to monitor BAER treatments and burn scars for at least 3 years. There is some great work out there by people like Pete Robichaud relating to how certain treatments work in terms of tons of runoff / acre in control (untreated) sites as well as those with various treatments. Pete and his crew do great work for the BAER community and that work affects what happens in terms of treatments all the time.

I'm more interested in landscape-scale treatment monitoring. This is apropos since I don't get to do the same field work Pete does and my expertise is in remote sensing. One way to leverage the technology is to capture imagery of burn scars during their growing seasons following the burn. I did a project a few years ago comparing vegetation response on the Cerro Grande Fire burn scar annually for 5 years. I compared areas heavily treated by BAER teams versus the natural greenup and die off of land that wasn't burned. Incidentally, we found that in areas treated by BAER teams, the vegetation greened up faster than those areas without any treatment but after 5 years, you couldn't really see any difference. In that case, the treatments did their job: slow immediate runoff and debris while letting nature take it's course.

Recently, a request came in to help track greenup during a single growing season on the Basin/Indians Complex. The local land manager was interested in not only when things greened up after the fire, but whether any greenup occurred during the year. If you capture a snapshot in time, you get a picture as to what is happening at that very moment. However, if you capture an image in August, for example, you may have some shrubs that are green and providing cover, but all the spring grasses are long dead. That's not a bad thing, though. Grasses that grow but then die off still provide ground cover and that information was desired by the land managers.

I captured 11 Landsat (30m) images during 2009 over this 2008 fire and created an NDVI (Normalized Differenced Vegetation Index) for each image. This provided a look at the vegetation vigor for each image. Then I did an overlay and captured the maximum NDVI value for each image and made a new layer showing where the vegetation was green sometime -- anytime -- between March and September, 2009.

It was a pretty cool layer and helps the locals quickly determine where there is no canopy or ground cover, even after a full growing season. The next step will be to overlay BAER treatments on these GIS layers to see if treatments really helped or hindered any growth.