Increasing stability of northern Austrian Lepidoptera populations over three decades DOI
Werner Ulrich, Thomas Schmitt,

Patrick Gros

и другие.

Ecological Entomology, Год журнала: 2024, Номер unknown

Опубликована: Ноя. 27, 2024

Abstract Agricultural intensification has led to landscape homogenisation across major parts of Europe and reduced diversity in flora fauna. In Central Europe, the species composition insect groups is increasingly dominated by a few ecologically generalist mobile species. So far, however, degree stability population sizes today's anthropogenic landscapes comparison pre‐Anthropocene hardly been analysed. Here, we studied large museum records Lepidoptera from northern Austria spanning years 1990–2022 infer trends sizes. On average, dynamics decreased increased significantly over time. This trend was most pronounced lowland regions, where agricultural transformed former heterogeneous into intensively managed grasslands fields. Community structures are now ubiquitous Habitat specialist existing isolated patches particularly A metapopulation structure appeared have stabilising effect on dynamics. We conclude that altered community might not only stem selective decline but also patterns stochasticity. Higher associated with faunal homogenisation. Precise butterfly sensitivity analyses require long‐term data average composition.

Язык: Английский

AI-Driven Insect Detection, Real-Time Monitoring, and Population Forecasting in Greenhouses DOI Creative Commons
Dimitrios Kapetas, Panagiotis Christakakis, Sofia Faliagka

и другие.

AgriEngineering, Год журнала: 2025, Номер 7(2), С. 29 - 29

Опубликована: Янв. 27, 2025

Insecticide use in agriculture has significantly increased over the past decades, reaching 774 thousand metric tons 2022. This widespread reliance on chemical insecticides substantial economic, environmental, and human health consequences, highlighting urgent need for sustainable pest management strategies. Early detection, insect monitoring, population forecasting through Artificial Intelligence (AI)-based methods, can enable swift responsiveness, allowing reduced but more effective insecticide use, mitigating traditional labor-intensive error prone solutions. The main challenge is creating AI models that perform with speed accuracy, enabling immediate farmer action. study highlights innovating potential of such an approach, focusing detection prediction black aphids under state-of-the-art Deep Learning (DL) models. A dataset 220 sticky paper images was captured. system employs a YOLOv10 DL model achieved accuracy 89.1% (mAP50). For prediction, random forests, gradient boosting, LSTM, ARIMA, ARIMAX, SARIMAX were evaluated. ARIMAX performed best Mean Square Error (MSE) 75.61, corresponding to average deviation 8.61 insects per day between predicted actual counts. visualization results, embedded mobile application. holistic approach supports early intervention strategies while offering scalable solution smart-agriculture environments.

Язык: Английский

Процитировано

0
Data‐driven approach to weekly forecast of the western flower thrips (Frankliniella occidentalis Pergande) population in a pepper greenhouse with an ensemble model DOI Open Access
Kin Ho Chan, Rob Moerkens, Nathalie Brenard

и другие.

Pest Management Science, Год журнала: 2025, Номер unknown

Опубликована: Фев. 21, 2025

Abstract BACKGROUND Integrated pest management (IPM) in European glasshouses has substantially advanced automated insect detection systems lately. However, transforming such an enormous data influx into optimal biological control strategies remains challenging. In addition, most forecast studies relied on the single‐best model approach, which is susceptible to overconfidence, and they lack validation over sufficient sampling repetitions where robustness questionable. Here we propose employing unweighted ensemble model, by combining multiple forecasting models ranging from simple (linear regressions Lotka–Volterra model) machine learning (Gaussian process, Random Forest, XGBoost, Multi‐Layer Perceptron), predict 1‐week‐ahead population of western flower thrips ( Frankliniella occidentalis ), a notorious glasshouses, under influence its agent Macrolophus pygmaeus pepper‐growing glasshouses. RESULTS Models were trained with only 1 year data, validated 3 years monitoring compartments evaluate their robustness. The full outperformed Naïve Forecast 10 out 14 for validation, around 0.451 26.6% increase coefficient determination R 2 ) directional accuracy, respectively. It also extended 0.096 best single equivalent 27% while maintaining 75% accuracy. CONCLUSION Our results demonstrated benefits traditional ‘single‐best model’ avoiding structural biases minimizing risk overconfidence. This showcased how minimal training can assist growers fully utilizing support decision‐making IPM. © 2025 Society Chemical Industry.

Язык: Английский

Процитировано

0
Aphids and their parasitoids persist using temporal pairing and synchrony DOI
Eduardo Engel, Douglas Lau, Wesley Augusto Conde Godoy

и другие.

Environmental Entomology, Год журнала: 2025, Номер unknown

Опубликована: Апрель 13, 2025

The study analyzed the population dynamics of aphids and their parasitoids in winter cereals southern Brazil, using wavelet transform (WT) to detect patterns periodicity synchronization over a decade (2011 2020). analysis revealed different peaks between aphid species parasitoids. Aphids, such as Rhopalosiphum padi L., Sitobion avenae (Fabricius), Schizaphis graminum (Rondani), Metopolophium dirhodum (Walker), showed varied peak frequencies, with M. consistently exhibiting shortening interval outbreaks. In contrast, maintained more-constant patterns, frequencies predominantly around 12 mo. Cluster identified 4 highly synchronized aphid-parasitoid pairs: S. graminum-Diaeretiella rapae (MacIntosh), R. padi-Aphidius platensis Brèthes, avenae-Aphidius uzbekistanicus Luzhetzki, dirhodum-Aphidius rhopalosiphi De Stefani-Perez. coherence (WC) significant correlations time series these pairs, ranging from in-phase anti-phase relationships time. results indicate that is viable tool for characterizing non-stationary series, parasitoid populations. Understanding can support integrated pest-management strategies, enabling more effective sustainable agricultural interventions.

Язык: Английский

Процитировано

0
Using random forest algorithm to improve Ceutorhynchus napi GYLL. (Coleoptera: Curculionidae) occurrence forecasting DOI
Quentin Legros,

Célia Pontet,

Céline Robert

и другие.

Journal of Applied Entomology, Год журнала: 2024, Номер unknown

Опубликована: Июнь 29, 2024

Abstract Random Forest algorithm was used to predict on‐field presence probability of rape stem weevil in France as a function climatic and landscape variables, based on long‐term multisite data set. A first version the model included set 342 variables. variable selection procedure retain only 15 most influential variables without significant drop predicting performances. Most retained were temperature related results showed that sum maximum daily above 9°C during week preceding observation predictor with largest influence occurrence. This reached mean AUC 0.77 outperformed some other published models. As such, this can help farmers precisely time insecticide application. It has been integrated decision support system freely available Terres Inovia (French applied agricultural research development institute dedicated oilseed crops) website.

Язык: Английский

Процитировано

1
Future of Information Systems for Pest Management: Data Acquisition and Integration to Guiding Management Decisions DOI
Mahendra Bhandari, Pankaj Pal, Michael J. Brewer

и другие.

CABI eBooks, Год журнала: 2024, Номер unknown, С. 251 - 262

Опубликована: Авг. 22, 2024

Язык: Английский

Процитировано

1
Future of Information Systems for Pest Management: Data Acquisition and Integration to Guiding Management Decisions DOI
Mahendra Bhandari, Pankaj Pal, Michael J. Brewer

и другие.

CABI eBooks, Год журнала: 2024, Номер unknown, С. 251 - 262

Опубликована: Авг. 23, 2024

Язык: Английский

Процитировано

1
Increasing stability of northern Austrian Lepidoptera populations over three decades DOI
Werner Ulrich, Thomas Schmitt,

Patrick Gros

и другие.

Ecological Entomology, Год журнала: 2024, Номер unknown

Опубликована: Ноя. 27, 2024

Abstract Agricultural intensification has led to landscape homogenisation across major parts of Europe and reduced diversity in flora fauna. In Central Europe, the species composition insect groups is increasingly dominated by a few ecologically generalist mobile species. So far, however, degree stability population sizes today's anthropogenic landscapes comparison pre‐Anthropocene hardly been analysed. Here, we studied large museum records Lepidoptera from northern Austria spanning years 1990–2022 infer trends sizes. On average, dynamics decreased increased significantly over time. This trend was most pronounced lowland regions, where agricultural transformed former heterogeneous into intensively managed grasslands fields. Community structures are now ubiquitous Habitat specialist existing isolated patches particularly A metapopulation structure appeared have stabilising effect on dynamics. We conclude that altered community might not only stem selective decline but also patterns stochasticity. Higher associated with faunal homogenisation. Precise butterfly sensitivity analyses require long‐term data average composition.

Язык: Английский

Процитировано

0