Skip to main content
ARS Home » Southeast Area » Stoneville, Mississippi » Crop Production Systems Research » Research » Publications at this Location » Publication #331540
Title: Integrating remote sensing, GIS and dynamic models for landscape-level simulation of forest insect disturbance

Author
item LU, LIANG - University Of Arkansas
item LI, XUECAO - Iowa State University
item Huang, Yanbo
item QIN, YUCHU - Us Geological Survey (USGS)
item HUANG, HUABING - Chinese Academy Of Sciences

Submitted to: Ecological Modeling
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/9/2017
Publication Date: 3/20/2017
Citation: Lu, L., Li, X., Huang, Y., Qin, Y., Huang, H. 2017. Integrating remote sensing, GIS and dynamic models for landscape-level simulation of forest insect disturbance. Ecological Modeling. 354:1-10.

Interpretive Summary: Insect migration, density, and distribution patterns are of primary concerns in forest industry. To deal with the challenge a systematic approach is needed. Scientists at University of Arkansas, Iowa State University, USDA-ARS Crop Production Systems Research Unit, Stoneville, Mississippi, US Geological Survey, and Chinese Academy of Sciences have worked collaboratively using spatio-temporal modeling and imagery to simulate forest insect disturbance at landscape level. This work innovatively integrated methods and data of remote sensing, geographic information system and dynamic models. The high accuracy and performance of the integration/modeling work illustrates that the developed framework is useful for forest insect dynamic analyses. The results of this study provides new information for further landscape insect movement analyses.

Technical Abstract: Cellular automata (CA) is a powerful tool in modeling the evolution of macroscopic scale phenomena as it couples time, space, and variable together while remaining in a simplified form. However, such application has remained challenging in landscape-level chronic forest insect epidemics due to the highly dynamic nature of insect behavior and interactions with surrounding environments. Recent advances in trajectory-based multi-temporal image analysis offer an alternative way to obtain high-frequency model calibration data. In this study, we propose an insect-CA modeling framework that integrates cellular automata, trajectory-based classification, and geographic information system to understand the temporal changes of insect ecological processes, and tested it with mountain pine beetle epidemics in the Rocky Mountains. The 88%~94% overall accuracies illuminate that such modeling framework is useful for forest insect dynamic analyses. To gain a deeper understanding of the model uncertainties, we further conducted sensitivity analysis to examine responses of model performance to various parameter settings. The simulation accuracy is sensitive to the construction methods of transition rules and ensemble random forest algorithm outperforms the traditional linear regression in our MPB case. Small neighborhood size is more effective in simulating the MPB movement behavior, indicating that short-distance is the dominating dispersal mode of MPB. The introduction of a stochastic perturbation component did not improve the model performance after testing a wide range of randomness degree, reflecting a relative compact dispersal pattern rather than isolated outbreaks. We conclude that CA with remote sensing observation is useful for landscape insect movement analyses, however, consideration of several key parameters are critical in the modeling process and should be more thoroughly investigated in future work.