Predicting the first seasonal occurrence of <i>Lobesia botrana</i> and <i>Eupoecilia ambiguella</i> in Austria using new multiple linear regression models
Climate change will cause new challenges for sustainable crop production, as increasing temperatures may accelerate the development of thermophilic insect pests and promote their spread and overwintering capacities. Improved or new forecasting models to determine the potential future temporal and s...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
International Viticulture and Enology Society
2025-07-01
|
| Series: | OENO One |
| Subjects: | |
| Online Access: | https://oeno-one.eu/article/view/8269 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Climate change will cause new challenges for sustainable crop production, as increasing temperatures may accelerate the development of thermophilic insect pests and promote their spread and overwintering capacities. Improved or new forecasting models to determine the potential future temporal and spatial shift in the occurrence of the European grapevine moth, Lobesia botrana (Denis and Schiffermüller) and the European grape berry moth, Eupoecilia ambiguella (Hübner) (Lepidoptera: Tortricidae) could help to better assess these future risks in Austrian wine-growing regions. Additionally, the timing of monitoring and control measures for both these pest species could be optimised to limit crop damages. In this context, prediction models for Lobesia botrana and Eupoecilia ambiguella were generated using long-term monitoring data (1980 to 2022) from 60 selected monitoring sites in 4 federal states in Austria, which had been collected using two different monitoring methods. Prediction models for the first seasonal occurrence of the different developmental stages (egg, larvae and adult) of the first and second flight/generation of both of the grape moth species were generated by applying stepwise multiple linear regression (MLR) analysis. The validation results showed high prediction accuracy for all six newly generated MLR models for L. botrana and for two out of six newly generated MLR models for E. ambiguella (R2 > 0.6; RMSE < 4.0; | BIAS | < 2.5). Depending on the developmental stage and generation of L. botrana, the validation results displayed an average prediction range of 0.89 days too early to 0.95 days too late. For E. ambiguella the predictions were on average 2.85 days too early to 0.20 days too late. To further improve model prediction accuracy, additional datasets should be included in the analysis, especially those from years in which extreme weather events occurred.
|
|---|---|
| ISSN: | 2494-1271 |