In September 2019, I posted a blog about identifying vegetation moisture stress through spectral analysis, and how that has life saving potential with electric transmission lines in the Wildland-Urban Interface (WUI). In a nutshell, if areas of high moisture stress are mapped with transmission lines, we can quickly identify areas of high fire risk and take preventative or mitigative measures before a disaster happens. The feedback I’ve received from that blog is encouraging because it means people are thinking about the real-world applications of spectroscopy. I love the fact that the wheels are turning in peoples’ minds, but I also want to emphasize the biggest potential reward of this analysis: to not hear anything on the news, to not have to hear about another tragedy that didn’t have to happen.

Now I’d like to share some of the work I did in my career as an undergraduate and graduate research assistant to provide some background on why I think so much about this stuff. In the summer of 2009, I worked as a field data collector to record species of vegetation in areas affected by wildfires in southern Arizona several years previously. The main purpose was to study what kinds of plants survive fires, what plants replace vegetation that was lost to fire, and what kinds of vegetation actually need fire to reproduce, Ponderosa Pine for example.  It is important to understand that wildland fire is a natural and necessary process, but forest management practices in the previous century have complicated fires of today. But that digression is for another blog. So – with a couple of grad students, I would hike to fairly remote areas in the Santa Catalina Mountains and locate plots that were originally surveyed by ornithologists over a decade before. They recorded the vegetation at these plots, so we had a baseline to compare post-fire vegetation. Aside from getting in the best shape of my life, I fell in love with the idea of using remote sensing and GIS to identify areas of moisture stress at large spatial scales. I had the opportunity to use MODIS imagery, processed by the Arizona Remote Sensing Center at my alma mater, which has a 250 m nominal spatial resolution for the NIR bands.

Calculating fuel moisture stress using vegetation indices is only part of identifying areas of high fire danger. There is another value called the Energy Release Component (ERC) that accounts for live and dead fuels as an indicator for fire season (Heinsch et al. 2009). I also used weather data from Remote Automated Weather Stations (RAWS; Warren and Vance, 1981) that was processed using Kansas City Fire Access Software (KCFAST). Fire season was established at any time when the ERC reached the 90^th percentile, but as far as utilities are concerned, fire season is year-round.

Ok – as Tom Petty said, “Don’t bore us, get to the chorus!” So here it is. The below images are samples of fire seasons in 2011 from the Santa Catalina – Rincon, Santa Rita, Pinaleño, Huachuca and Chiricahua mountains of southern Arizona.  I chose 2011 because that was one of the more severe years for fire in Arizona.

The original intent for these images was to identify areas of vegetation moisture stress to investigate a relationship between winter rainfall and summer fire season. If we can identify vegetation live and dead fuel moisture stress for this purpose, we can certainly identify moisture stressed vegetation at the Wildland Urban Interface. It is not perfect or glamorous, but spectral analysis in conjunction with field data and other mitigative efforts provides a real opportunity to reduce the devastating effects of fires that may be caused by electric transmission lines.