| Importance of remote sensing variables for risk assessment of codling moth (Lepidoptera: Tortricidae) in East Azarbaijan province of Iran |
| Paper ID : 1177-3IICE (R1) |
| Authors: |
|
Hakimeh Shayestehmehr *1, Roghaiyeh Karimzadeh1, Bakhtiar Feizizadeh2, Shahzad Iranipour1 1Department of Plant Protection, Faculty of Agriculture, University of Tabriz, Tabriz, Iran 2Department of Remote Sensing & GIS, Faculty of Planning and Environmental Sciences, University of Tabriz, Tabriz, Iran |
| Abstract: |
| Codling moth, Cydia pomonella L. (Lepidoptera: Tortricidae) is the most destructive insect pest of apple across the world. In this study, maximum entropy modeling (MaxEnt) was used to generate habitat suitability maps of codling moth in order to predict the regions with high risk of this pest infestation in East Azarbaijan province of Iran. Moreover, the importance of remote sensing variables was evaluated in improving model performance and accuracy. Field surveys for the occurrence of codling moth were conducted during two growing seasons (2017 and 2018) in four Counties of East Azarbaijan province including Ahar, Maragheh, Marand, and Tabriz. The activity of codling moth adult males was monitored using yellow delta-shaped traps baited with sex pheromone (PH-227-1RR, Russell IPM, UK) of the insect. During two-years survey, codling moth was sampled from 44 occurrence points. Climatic and topographic variables and normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), percent tree cover, percent non-tree cover and percent non-vegetated (bare) area derived from MODIS (Moderate Resolution Imaging Spectroradiometer) imagery were used as environmental variables in modeling. Primary models were implemented using only layers from each category of environmental variables to determine the layers contributed significantly to the distribution of codling moth. Then, two final models were implemented, one with remote sensing variables and the other without remote sensing variables. The performance of the models was evaluated using the area under the receiver operating characteristics curve (AUC) metric. The AUC values of the models with and without remote sensing variables were 0.938 and 0.991, respectively. Also, percent tree cover of 2017-2018, NDVI for the months of September and April, precipitation of the warmest quarter and standard deviation of mean monthly temperatures had the most percent contribution in MaxEnt model with remote sensing variables. The results indicated that the remote sensing data have potential to improve model performance and to obtain more accurate habitat suitability maps. These maps can assist managers in forecasting and planning control measures and, therefore, effective management of current infestations of codling moth. |
| Keywords: |
| Codling moth, Remote sensing, MaxEnt modeling, Risk map |
| Status : Paper Accepted (Oral Presentation) |
