Pre and post fire monitoring in Mediterranean woodlands using the phenology-based method
A fuel-based fire risk map was produced from decomposing time series of MODIS NDVI for Mt Carmel. The fire risk map reproduced fire spread behavior of the largest wildfire in Mt Carmel (December 2010) with great accuracy. Post fire assessments of changes in woody and herbaceous vegetation cover and fire severity were implemented using the decomposition method.
|
The Normalized Difference Vegetation Index (NDVI) is a good indicator of the vegetation dryness status. Thus, it could be used to assess dry woody vegetation matter, which is the main fuel source in Mediterranean forest wildfires. A decline in NDVI attributed to woody vegetation would imply drying trends and increasing dry matter for potential fires. Oppositely, low woody cover would reduce fire spread vulnerability in woodlands because there is less fuel for fire. At the Mt Carmel woodlands, the ephemeral herbaceous vegetation plays only a minor role in wildfires because of its relatively low biomass. Figure 1a shows that the fuel-based fire risk map retrieved from mean annual and trends in NDVI attributed to the woody cover (NDVIW, see in "Monitoring vegetation with NDVI" page or in Helman et al. 2015 for explanation on the phenology-based decomposition method) reproduced fire-spread behavior of the wildfire of 2010 with great accuracy. Most of the pixels within the wildfire area were ranked with the highest risk level being 18 and 21% (for levels 9 and 10, respectively) from the total Mt Carmel cells ranked with the same levels (Figure 1b-c). This is more than the expected in the burnt area, which comprises less than 9% of Mt Carmel area. These results, based only on fuel density and status (both from NDVIW) do not consider topographic or weather conditions, which are important driving forces of fire spread behavior. However, such a fuel-based map might improve fire hazard models that mostly rely upon static fuel maps. Those do not consider the effects of drought years on the woody (and/or herbaceous in other cases) vegetation status, which is shown to be an important factor determining fire spread behavior. We used NDVIW to assess fire severity in the burnt area of Mt Carmel (Figure 2). The difference in NDVIW between the years before (2010) and after (2011) the wildfire (dNDVIW) was correlated to the fire severity classified at field following Neary et al. (2005). The mean dNDVIW was significantly different (p<0.001) between areas classified in field as low, medium and high severity with mean dNDVIW of 0.1, 0.13 and 0.16, respectively. The decomposed NDVI signals (NDVIW and NDVIH) were also used to assess woody and herbaceous vegetation recovery (i.e., changes in cover) in the burnt area (Figure 3). Results agreed well to field estimates when using both, empirical relationships and the classical two-end members equation. All estimates showed that after a woody cover reduction following the wildfire, woody vegetation cover decreased from 2011 to 2012 (from the year after the wildfire to the next year) but increased in subsequent years (2013 and 2014). The Herbaceous vegetation, which generally increased in the burnt area, showed an opposite change with initial increase from 2011 to 2012 (first to second year after the wildfire) but a following decrease in the next two years (in 2013 and 2014). The results mentioned above are presented more extensively in Helman et al. 2015. |
Figure 1. (a) A fuel-based fire risk map for Mt Carmel produced for the year 2009 (the year prior to the wildfire of 2010) from woody vegetation cover (mean NDVIW) and status (NDVIW trends). Superimposed is the burnt area; Histograms of (b) the total number of pixels in Mt Carmel with their respective risk levels and (c) the ratio between number of pixels with a specific risk level in the burnt area to that with the same rank level in the entire Mt Carmel area (in %). The dashed line in (c) indicates the percent of burnt area from the total area of Mt Carmel (i.e., 8.6%).
Figure 2. Maps of (a) low, medium and high severity burnt areas classified in field and extended with high-resolution aerial photograph; and (b) the difference between post- and pre-fire NDVIW (∆NDVIW). (c) Box plot of mean, 1st and 3rd quartiles (with respective standard deviations) of ∆NDVIW in the low, medium and high severity areas mapped in field (shown in a). Different letters indicate statistically significant differences at p<0.001 using a two-tailed Student’s t-test after Bonferroni correction.
|