نوع مقاله : مقاله پژوهشی
نویسندگان
1 گروه باغبانی و گیاهپزشکی، دانشکده کشاورزی، دانشگاه صنعتی شاهرود، شاهرود، ایران
2 گروه مهندسی آب، دانشکده کشاورزی، دانشگاه صنعتی شاهرود، شاهرود، ایران
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
The Moroccan locust (Dociostaurus maroccanus), as one of the native pests of Iran, is recognized for its migratory capability, population outbreaks, and severe damage to crops and rangelands. In this study, meteorological data (temperature, precipitation, and relative humidity), field-measured data (soil moisture and locust abundance), and indices derived from Landsat 8 satellite imagery (LST, TCI, OSAVI, and BSI) were used as input variables for a multilayer perceptron (MLP) neural network model to estimate locust abundance at the pixel scale. To this end, the dataset was randomly divided into training (70%) and testing (30%) subsets. Model performance was evaluated using the correlation coefficient (R) and the Nash–Sutcliffe efficiency (NSE). The results indicated that the input variables of the simulation model were appropriately selected and that the model structure was capable of simulating locust abundance with very high accuracy, such that NSE values of approximately 0.98 and 0.99 were obtained for the training and testing phases, respectively. Furthermore, the results showed that soil moisture exhibited the lowest correlation with measured locust abundance (R=0.9), while the TCI index showed the highest correlation (R=0.98). Overall, the findings of this study demonstrate that the combined use of field data and satellite-derived indices within machine learning frameworks can be highly effective for simulating the habitat of the Moroccan locust. The proposed approach enables large-scale and multi-temporal monitoring of Moroccan locust populations, given the availability of satellite imagery, thereby providing valuable support for experts and decision-makers in pest management and control planning.
کلیدواژهها [English]
Extended Abstract
Introduction
The Moroccan locust (Dociostaurus maroccanus) is a native, migratory, and polyphagous pest in Iran, mainly distributed in dry and low-productivity habitats surrounding the Mediterranean region. In years with favorable climatic conditions, this species transitions from a solitary to a gregarious phase, becoming a serious threat to rangelands and crops. Climatic factors such as temperature, precipitation, and soil moisture are the key drivers of its population dynamics. Meanwhile, vegetation cover and soil type determine the suitability of oviposition sites and nymphal development. Considering the increasing climatic fluctuations and recent droughts in Golestan Province, accurate modeling of potential habitats using remote sensing and artificial intelligence is essential for targeted management and early prediction of locust outbreaks.
Materials and Methods
This study was conducted in Gonbad-e Kavus County, located in the northern and central parts of Golestan Province, which has a warm and dry Mediterranean climate (150–200 dry days per year). Field data included locust density and volumetric soil moisture (0–10 cm depth), collected between 10:00 and 11:00 a.m. Climatic data (temperature, precipitation, relative humidity) were obtained from the Incheh Borun meteorological station, while satellite-based indices, including Land Surface Temperature (LST), Temperature Condition Index (TCI), Optimized Soil-Adjusted Vegetation Index (OSAVI), and Bare Soil Index (BSI)-were derived from atmospherically and geometrically corrected Landsat-8 images. A total of 151 data points were used for modeling (70% for training and 30% for testing). Locust density was predicted using a Multilayer Perceptron (MLP) neural network employing sigmoid and ReLU activation functions. Model training was performed using the Levenberg–Marquardt backpropagation algorithm, and performance was evaluated using the coefficient of determination (R²) and the Nash–Sutcliffe Efficiency (NSE) indices.
Results and Discussion
Analysis of the satellite-derived indices revealed that the Bare Soil Index (BSI) was the most influential factor in identifying suitable habitats for the Moroccan locust. High BSI values, representing dry and sparsely vegetated soils, were directly associated with increased locust density and oviposition activity, whereas negative BSI values corresponded to densely vegetated and cooler areas. Land Surface Temperature (LST) was also identified as a key predictor, with the 38.26–47.97°C range providing optimal conditions for locust activity and reproduction. Strong correlations between LST, BSI, and OSAVI confirmed the synergistic role of these indices in habitat modeling. High TCI values (>0.8) effectively identified thermally favorable regions with dense populations. The OSAVI index showed an inverse relationship with locust abundance and reproduction, indicating that higher vegetation density corresponded to lower locust populations. activity-higher vegetation density reduced locust abundance and reproduction. These findings are consistent with previous studies from North Africa and Central Asia. The MLP model demonstrated very high accuracy in predicting locust density (training R² = 0.99, testing R² = 0.98; NSE = 0.99 and 0.98), explaining over 98% of observed variation, confirming the model’s robustness and predictive reliability.
Conclusion
The MLP-based modeling approach demonstrated high capability in accurately predicting the potential habitats of the Moroccan locust under variable climatic conditions. Satellite-derived indices (LST, OSAVI, TCI, and BSI) proved to be effective tools for understanding environmental heterogeneity and identifying high-risk areas. The results highlight the importance of integrating remote sensing and artificial intelligence within national locust management systems, providing a foundation for early warning, targeted monitoring, and optimized pesticide application. With the increasing frequency of droughts and vegetation degradation in northern Iran, combining climate monitoring, satellite-based habitat modeling, and sustainable resource management is essential to mitigate future outbreaks and safeguard food security. The findings of this study can serve as a model for developing early warning systems for pest management in other arid regions of the country.
Author Contributions
“Conceptualization: Roozbeh Moazenzadeh and Masoud Hakimitabar; methodology: Daryoush Mansouri Razi; software: Roozbeh Moazenzadeh; validation: Roozbeh Moazenzadeh and Masoud Hakimitabar, formal analysis: Roozbeh Moazenzadeh and Masoud Hakimitabar; investigation: Daryoush Mansouri Razi, Roozbeh Moazenzadeh and Masoud Hakimitabar; writing-original draft preparation: Masoud Hakimitabar and Roozbeh Moazenzadeh; writing-review and editing: Daryoush Mansouri Razi, Roozbeh Moazenzadeh and Masoud Hakimitabar; project administration: Masoud Hakimitabar; funding acquisition: Shahrood University of Technology. All authors have read and agreed to the published version of the manuscript.”
All authors contributed equally to the conceptualization of the article and writing of the original and subsequent drafts.
Data Availability Statement
Not applicable.
Acknowledgements
The authors would like to thank Shahrood University of Technology for preparing the fund of research and also all participants of the present study.
Ethical considerations
This study only included arthropod material, and all required ethical guidelines for the treatment and use of animals were strictly adhered to in accordance with international, national, and institutional regulations. No human participants were involved in any studies conducted by the authors for this article. The authors avoided data fabrication, falsification, plagiarism, and misconduct.
Conflict of interest
The authors declare no conflict of interest.
)