A Synergetic Approach to Burned Area Mapping Using Maximum Entropy Modeling Trained with Hyperspectral Data and VIIRS Hotspots

Abstract

Southern European countries, particularly Spain, are greatly affected by forest fires each year. Quantification of burned area is essential to assess wildfire consequences (both ecological and socioeconomic) and to support decision making in land management. Our study proposed a new synergetic approach based on hotspots and reflectance data to map burned areas from remote sensing data in Mediterranean countries. It was based on a widely used species distribution modeling algorithm, in particular the Maximum Entropy (MaxEnt) one-class classifier. Additionally, MaxEnt identifies variables with the highest contribution to the final model. MaxEnt was trained with hyperspectral indexes (from Earth-Observing One (EO-1) Hyperion data) and hotspot information (from Visible Infrared Imaging Radiometer Suite Near Real-Time 375 m active fire product). Official fire perimeter measurements by Global Positioning System acted as a ground reference. A highly accurate burned area estimation (overall accuracy = 0.99%) was obtained, and the indexes which most contributed to identifying burned areas included Simple Ratio (SR), Red Edge Normalized Difference Vegetation Index (NDVI 750 ), Normalized Difference Water Index (NDWI), Plant Senescence Reflectance Index (PSRI), and Normalized Burn Ratio (NBR). We concluded that the presented methodology enables accurate burned area mapping in Mediterranean ecosystems and may easily be automated and generalized to other ecosystems and satellite sensors.

Maximum entropy method-based forest fire prediction mapping of Sikkim Himalaya.

“…Being fundamentally distinct from other machine learning methods, MaxEnt uses presence-only dataset to train itself (Elith et al 2011). However, like many studies have shown earlier, the present study also indicated that this distinction of MaxEnt does not limit its capability in generating reliable hazard prediction maps (Arpaci et al 2014;Massada et al 2013;Peters et al 2013;Fonseca et al 2016;Kim et al 2015;Fernández-Manso and Quintano 2020;Lim et al 2017). In this study a limited set of features have been considered for forest re prediction.…”

“…Studies indicate that MaxEnt has performed equally well in comparison to other machine learning methods in predicting a forest re. (Arpaci et al 2014;Massada et al 2013;Peters et al 2013;Fonseca et al 2016;Kim et al 2015;Fernández-Manso and Quintano 2020;Lim et al 2017) In this study, MaxEnt has been applied to prepare a forest re prediction map of Sikkim Himalaya using MODIS and Ground data-based forest re inventory. As features, meteorological, topological, ecological and human-induced data have been used to train the MaxEnt model.…”

Drivers and distribution of forest fires in Sikkim Himalaya: A maximum entropy-based approach to spatial modelling.

“…Being fundamentally distinct from other machine learning methods, MaxEnt uses presence-only dataset to train itself (Elith et al, 2011). However, like many studies have shown earlier, the present study also indicated that this distinction of MaxEnt does not limit its capability in generating reliable hazard prediction maps (Arpaci et al, 2014;Fernández-Manso & Quintano, 2020;Fonseca et al, 2016;Kim et al, 2015;Lim et al, 2017;Massada et al, 2013;Peters et al, 2013). In this study a limited set of features have been considered for forest re prediction.…”

“…Studies indicate that MaxEnt has performed equally well in comparison to other machine learning methods in predicting a forest re. (Arpaci et al, 2014;Fernández-Manso & Quintano, 2020;Fonseca et al, 2016;Kim et al, 2015;Lim et al, 2017;Massada et al, 2013;Peters et al, 2013) In this study, MaxEnt has been applied to prepare a forest re prediction map of Sikkim Himalaya using MODIS and Ground data-based forest re inventory. As features, meteorological, topological, ecological, in-situ and human-induced data have been used to train the MaxEnt model.…”

MODIS Sensor Capability to Burned Area Mapping—Assessment of Performance and Improvements Provided by the Latest Standard Products in Boreal Regions

“…To assess the impact of the PB on the accuracy of a BA classified map, the area enclosed by it (the Area Under the Pareto Boundary, AUPB) was calculated, by analogy with the area of the ROC (Receiver Operating Characteristic) curve used in other burned area studies using machine learning or data mining [ 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 ]. The ROC curve is a probability curve constructed from sensibility and 1-specificity pairs {(S i ,1-Sp i )}, obtained using a procedure similar to that of the PB, and the area enclosed by it, the Area Under the ROC Curve (AUC), is interpreted as a measure of the separability between the two classes considered from the selected p parameter.…”