Evaluating the impact of biological control on green oak leaf-roller (Tortrix viridana L.) populations and canopy recovery in Zagros Forests using remote sensing

Background and Objective: Defoliating insects are among the major factors threatening the health and sustainability of forest ecosystems. Climate change and rising global temperatures have contributed to increased frequency and geographical expansion of these pests in Iranian forests. The oak leaf-roller moth (Tortrix viridana L.) is one of the most important pests of oak forests, causing extensive defoliation, reduced photosynthetic capacity, and tree weakening through feeding on the leaves of coppice oak trees. Controlling this pest in the inaccessible areas of the Zagros Mountains requires environmentally friendly approaches. In this regard, biological agents such as Bacillus thuringiensis (Bt) can serve as suitable alternatives to chemical pesticides. This study aimed to evaluate the effect of Bt spraying on the population density of the oak leaf-roller moth and to assess the potential of the NDVI vegetation index derived from Sentinel-2 imagery for monitoring pest-induced vegetation responses.
Material and Methods: This study was conducted over approximately 926 ha of oak forests in the Ghabrehosein area of Piranshahr County, northwestern Iran, which represents one of the major infestation hotspots of the oak leaf-roller moth. To control the pest, ground-based Bt spraying was carried out in May 2023 at a concentration of 1 kg per 100 L of water. The treatment effect was evaluated using a Before–After Control–Impact (BACI) design. In both control and treated areas, three 300-m sampling transects were established using a random-systematic approach, and 30 trees were selected along each transect at 30-m intervals. Larval density was assessed before spraying and one month after treatment. The effect of Bt on larval density was analyzed using a BACI linear mixed-effects model, with treatment, time, and their interaction considered as fixed effects and tree identity as a random effect. The percentage reduction in larval density was also calculated using the ΔDxt index. To evaluate vegetation responses, NDVI was derived from Sentinel-2 Level-2A imagery during April–September 2023. Image processing and generation of NDVI time series at a 10-m spatial resolution were performed using the Google Earth Engine platform. The relationship between NDVI and larval density was evaluated using simple linear regression.
Results: The results showed that the mean larval density of T. viridana in the treated area decreased from 53.37 ± 15.74 larvae per tree before Bt application to 18.73 ± 7.26 larvae per tree after treatment. In contrast, only minor changes were observed in the control area, where mean larval density changed from 48.74 ± 12.25 to 47.03 ± 10.87 larvae per tree. The BACI linear mixed-effects model indicated a highly significant treatment × time interaction (p < 0.001), demonstrating the effectiveness of Bt application in reducing pest population density. Post-hoc comparisons confirmed a significant reduction in larval density in the treated area, whereas no significant change was observed in the control area. The ΔDxt index indicated that Bt spraying reduced larval density by approximately 65%, while changes in the control area remained within the range of natural population fluctuations. NDVI time-series analysis showed similar seasonal patterns in both areas; however, following treatment, NDVI values remained consistently higher in the treated area compared with the control. The smaller decline in NDVI in the treated area indicated better vegetation preservation and reduced feeding damage caused by larvae. Simple linear regression analysis revealed a significant positive relationship between NDVI and larval density in both areas before spraying. After treatment, this relationship became statistically non-significant in the treated area, indicating a reduced association between pest density and vegetation changes following Bt application.
Conclusion: The results demonstrated that biological spraying with Bt significantly reduced the population density of the oak leaf-roller moth, and based on the BACI design and linear mixed-effects model, this effect could be distinguished from natural pest population fluctuations in the control area. The reduction in larval feeding activity following Bt treatment was associated with the maintenance of higher NDVI values in the treated area, indicating reduced defoliation and improved vegetation condition. Furthermore, regression analysis showed that the relationship between NDVI and larval density became statistically insignificant after Bt application in the treated area, whereas it remained significant in the control area. Therefore, integrating field-based observations with NDVI derived from Sentinel-2 imagery can provide an effective approach for evaluating biological pest control and monitoring vegetation health at large spatial scales.