Before my GIS career path led me to pipeline and electric utilities, I analyzed moisture stressed vegetation before and during fire seasons in southern Arizona, to possibly identify where the highest fire danger areas may occur. I used near infrared (NIR) spectral index data, from a space-based sensor called MODIS, to investigate the relationship between antecedent winter rainfall and moisture stressed vegetation the following summer. Ok, that was a mouthful, and you may be asking – Why bother using near infrared? Can’t you just see that plants are dried and withered? The short answer is that plants reflect more infrared light than visible light. Indeed, they do not like near infrared light, which lies in wavelengths just beyond the human eye’s capability to see, because it causes the plant to heat up excessively. For the sake of this article, however, a slightly deeper explanation is necessary to understand how spectral analysis is useful in the context of protecting forests and utilities.

Here is a quick rehash of some high-level botany. Stomata are pores on a leaf’s surface. They take in carbon dioxide (CO₂) and release oxygen (O₂). During photosynthesis, blue and red light are absorbed by chlorophyll, and green light is reflected (Now you can explain to a small child why the grass is green). A defense mechanism of moisture stressed vegetation is stomatal closure, which minimizes evapotranspiration (how plants sweat) and, in turn, reduces CO₂ intake and O₂ release. The diminished CO₂ intake causes less absorption of blue and red light and increased absorption of infrared light, which heats the plant, causing additional moisture stress. Since plants reflect more infrared light than visible light, we can detect moisture stress in the near infrared part of the electromagnetic spectrum before we see a plant start to wither or change color. How much sooner is dependent on the species of plant. Mixed conifers, for example, do not wilt like leafy vegetation and may not appear dried out to the human eye in the short term. This is where spectral analysis has lifesaving potential.

Transmission lines carry electricity over long distances, often in rural and forested areas. An accurate spatial representation in GIS of the Right of Way (ROW), Easement and actual location of the power line is crucial to the safety of field crews and people who live near the wildland-urban interface. To supplement asset, easement and ROW location accuracy, the opportunity exists to create spectral layers in GIS that identify the location of moisture stressed vegetation and their proximity to transmission lines and ROWs. To create such a layer, spectral data from an Earth Observing System (EOS) would be required to cover a larger spatial scale, but aerial spectral products could also be used at smaller scales. A widely used product known as the Normalized Difference Vegetation Index (NDVI) can show snapshots of current moisture stress. There are, however, temporal limitations to NDVI, and moisture stress may not be captured as well in mixed conifer type vegetation as compared to leafy vegetation. Another index that can model temporal dynamics of moisture stress is called the Fuel Moisture Stress Index (FMSI). This index takes the inverse z score of the NDVI to observe overall moisture stress over yearly or even decadal timespans.

Most of the general public, and nearly all of those who work in the electric utility industry, are well aware of the disastrous outcomes that can happen when power lines come into contact with trees or plants. Vegetation that is moisture stressed will almost certainly amplify those outcomes. The odds of these disasters occurring can be reduced significantly by integrating spectral analyses of vegetation with utility footprints. The earlier we are aware of high-risk areas, the better chance we have to mitigate or prevent loss of life, property and natural resources.