Background

I am interested in exploring post-fire landscape evolution of western landscapes 1-20 years following major wildfires. Specifically, I am looking for trends in remotely sensed imagery associated with significant mass movements (i.e. debris flows, landslides, and scour of hillslopes). I have limited my area of interest (AOI) to within the recorded burned perimeter of recent wildfires reported by the Monitoring Trends in Burn Severity (MTBS) service (Eidenshink et al. 2007, Finco et al. 2012). Barlow et al. (2003) used Landsat NDVI and slope (< 15°) to detect translational landslide scars. Landsat bands 3 and 7 are suggested for a normalized difference index (NDI) by Mwaniki et al. (2015). Landsat 8 has the Operational Land Imager (OLI) which takes 15 m panchromatic images. Other techniques include satellite InSAR, lidar, and GIS modelling approaches have yielded promising results. 

I plan to apply the normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), and surface reflectance data from Landsat time series in Google Earth Engine as potential predictors of mass movement. The Earth Engine already has a time lapse derived from Landsat and the National Agricultural Imagery Program (NAIP) which may reveal past landslide events globally. 

Landsat has been used by others (1, 2, 3, 4) to identify mass movements. The AGU Landslide Blog reports on major mass movement events.

Method

The proposed methodology uses a Google Earth Engine API framework of delineating past fires from MTBS (and potentially the Canadian Wildland Fire Information System and the US Wildfire Decision Support System) and comparing their respective landslide probabilities. Other useful independent parameters include the NASA Shuttle Radar Topography Mission, USGS NED DEM 1/3 arc second.

MTBS

I downloaded the MTBS burned area shape file and converted it to keyhole markup language (.kml) using QGIS 2.16. (Note: older versions of QGIS have GDAL python errors where the extended fields of the shapefile are not converted to .kml)

The full MTBS burned area .shp file had to be reduced in size from 550 mb to <250 mb so that it could be uploaded as a Google fusion table. I selected a small subset of the data for Arizona and New Mexico for my test set.

Google Earth Engine uses fusion tables for importing vector file types. I uploaded the resulting 16 mb .kml file for the western USA to my Google.Drive account.

MTBS fire perimeter subset - Google Fusion Tables

Earth Engine

In EE Code editor I created a new script and gave it an arbitrary name (MTBS fires AZ NM); in the scripting window I entered

// Create a feature collection from a Fusion Table.
//
// Select features from the MTBS Arizona and New Mexico subset.
var fc = ee.FeatureCollection('ft:1Yo6yjtw_mJ7tEIm5hDc0H6Fzyf3InCBTOHbRbqxy');
// Paint into a blank image.  Cast to byte() so we can use more than 1 color.
var image1 = ee.Image()
               .byte()
               .paint(fc, 'Id');
// Display the image using random colors for each value.
Map.addLayer(image1.randomVisualizer());
Map.setCenter(-110, 32, 6);

I plan to use zonal statistics of each fire area using the Grouped Reductions and Zonal Statistics tool.

References

Barlow, J., Martin, Y., & Franklin, S. E. (2003). Detecting translational landslide scars using segmentation of Landsat ETM+ and DEM data in the northern Cascade Mountains, British Columbia. Canadian journal of remote sensing, 29(4), 510-517.

Eidenshink, J. C., Schwind, B., Brewer, K., Zhu, Z. L., Quayle, B., & Howard, S. M. (2007). A project for monitoring trends in burn severity. Fire ecology,3(1), 3-21.

Finco, M., Quayle, B., Zhang, Y., Lecker, J., Megown, K. A., & Brewer, C. K. (2012). Monitoring trends and burn severity (MTBS): monitoring wildfire activity for the past quarter century using Landsat data. In: Morin, Randall S.; Liknes, Greg C., comps. Moving from status to trends: Forest Inventory and Analysis (FIA) symposium 2012; 2012 December 4-6; Baltimore, MD. Gen. Tech. Rep. NRS-P-105. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. [CD-ROM]: 222-228.

Kane, V. R., North, M. P., Lutz, J. A., Churchill, D. J., Roberts, S. L., Smith, D. F., ... & Brooks, M. L. (2014). Assessing fire effects on forest spatial structure using a fusion of Landsat and airborne LiDAR data in Yosemite National Park. Remote Sensing of Environment, 151, 89-101.

Mwaniki, M. W., Agutu, N. O., Mbaka, J. G., Ngigi, T. G., & Waithaka, E. H. (2015). Landslide scar/soil erodibility mapping using Landsat TM/ETM+ bands 7 and 3 Normalised Difference Index: A case study of central region of Kenya. Applied Geography, 64, 108-120.