PubAg

Main content area

Modelling and quantifying the spatial distribution of post-wildfire ash loads

Author:
Chafer, Chris J., Santín, Cristina, Doerr, Stefan H.
Source:
The International journal of wildland fire 2016 v.25 no.2 pp. 249-255
ISSN:
1049-8001
Subject:
Landsat, fire severity, geographic information systems, land management, landscapes, methodology, models, remote sensing, subwatersheds, water supply, wildfires, Australia
Abstract:
Ash is generated in every wildfire, but its eco-hydro-geomorphic effects remain poorly understood and quantified, especially at large spatial scales. Here we present a new method that allows modelling the spatial distribution of ash loads in the post-fire landscape, based on a severe wildfire that burnt ~13600ha of a forested water supply catchment in October 2013 (2013 Hall Road Fire, 100km south-west of Sydney, Australia). Employing an existing spectral ratio-based index, we developed a new spectral index using Landsat 8 satellite imagery: the normalised wildfire ash index (NWAI). Before- and after-fire images were normalised and a differenced wildfire ash image (dNWAI) computed. The relationship between dNWAI and ash loads (t ha-1) quantified in situ at nine sampling locations burnt under a range of fire severities was determined using a polynomial regression (R2=0.98). A spatially applied model was computed within a geographic information system (GIS) to illustrate the spatial distribution of ash across the area burnt and to estimate ash loads in the five subcatchments affected by the wildfire. Approximately 181000 tonnes of ash was produced by the wildfire, with specific loads increasing with fire severity. This new tool to model wildfire ash distribution can inform decisions about post-fire land management in future wildfires in the region. It can also be adapted for its application in other fire-prone environments.
Agid:
4813378