Detecting beetle infestations in pine forests using MODIS NDVI time-series data

Abstract

The paper considers the detection of beetle infestations in North American pine forests using high temporal resolution, coarse spatial resolution MODIS remotely sensed satellite images. Two methods are proposed to detect beetle infestation, both applying a triply modulated cosine model. The first method uses an Extended Kalman Filter (EKF) for estimating model parameters, and the second a Least Squares estimator. When beetles infest a forest, the changes in the affect large geographical area. Therefore, the change detection metrics are based on the time series of each pixel, and do not utilize information from neighboring pixels. Using data from the Rocky Mountain region of the United States and of British Columbia in Canada, we show that our methods are highly effective at detecting beetle infestations.

Near-Real Time Detection of Beetle Infestation in Pine Forests Using MODIS Data

“…The solution of (5) at every position of the window produces parameter time series of (1), , for . Our study [52] about changes introduced by beetle infestation in vegetation index time series, and findings in [22] suggest that beetle infestation affects trend component of the signal significantly and the change can be detected in the trend component alone. Therefore, we apply the change detection strategy on time series.…”

“…This method performs well when the change is a spatially rare event, i.e., when the changed pixels in the window are a minority, and loses its accuracy for the case of changes caused by beetles in pine forests. This is because these changes are normally widespread and involve the majority of the pixels in the window [52].…”

“…In [52], we proposed modifications in the EKF-based method [16] for the detection of beetle infestation, and introduced an nonlinear least squares (NLS) based method for further improvements in the results. We showed that the beetle infestation can be detected well by modeling a single time series at a time using the triply modulated cosine function (1) and the trend parameter ( ) of the function plays a key role in detecting beetle infestation.…”

Quantifying massive outbreaks of the defoliator mothOrmiscodes amphimonein deciduousNothofagus‐dominated southern forests using remote sensing time series analysis

“…For cloudy areas, such as south Chile, chances of obtaining cloud‐free satellite data are limited. For this reason, we used a higher temporal resolution satellite product: the 16‐day composition of MODIS Vegetation Indices (version 5), specifically, the Enhanced Vegetation Index (EVI), which is sensitive to vegetation green biomass (Huete et al, ; White, Pontius, & Schaberg, ; Zhang et al, ) and has been used for insect outbreak detection in the Northern Hemisphere (Anees, Olivier, O'Rielly, & Aryal, ; de Beurs & Townsend, ; Spruce et al, ; Verbesselt, Robinson, Stone, & Culvenor, ). Although with lower spatial resolution (250 m) than Landsat (30 m), these 16‐day EVI composites provide a more complete spatio‐temporal coverage of the vegetated land surface and are especially suitable for cloudy areas like the tropics or high‐latitude regions.…”

“…Hemisphere (Anees, Olivier, O'Rielly, & Aryal, 2013;de Beurs & Townsend, 2008;Spruce et al, 2011;Verbesselt, Robinson, Stone, & Culvenor, 2009). Although with lower spatial resolution (250 m) than Landsat (30 m), these 16-day EVI composites provide a more complete spatio-temporal coverage of the vegetated land surface and are especially suitable for cloudy areas like the tropics or highlatitude regions.…”