Document Type : Review article
Authors
1
Department of Environmental Engineering, Faculty of Natural Resources and Earth Sciences, Shahrekord University, Shahrekord, Iran
2
Department of Natural Engineering, Faculty of Natural Resources and Earth Sciences, Shahrekord University, Shahrekord, Iran
10.22034/iwm.2026.2085956.1266
Abstract
Extended Abstract
Introduction: Ecosystem services play a fundamental role in supporting human well-being, sustainable development, and natural resource management. Ecosystem service models have become essential tools for assessing, quantifying, and predicting these services, providing valuable information for environmental planning and policy-making. Despite substantial advances in ecosystem service modeling, model validation remains one of the most critical challenges affecting the reliability, accuracy, and practical applicability of model outputs. The performance of ecosystem service models depends not only on their internal structure but also on the quality of input data, validation approaches, uncertainty analysis, and the compatibility between the model and the study objectives. Considering the absence of a comprehensive framework for ecosystem service model validation, this study aimed to systematically review the existing literature, compare widely used ecosystem service models, identify major validation challenges and research gaps, and propose a conceptual framework to improve and standardize the ecosystem service model validation process.
Materials and Methods: This study was conducted as a systematic literature review. Scientific publications published between 2005 and 2026 were retrieved from Web of Science, Scopus, ScienceDirect, SpringerLink, and Google Scholar using keyword combinations related to ecosystem services, model validation, uncertainty analysis, remote sensing, and machine learning. The initial search identified 284 publications. After removing duplicate records and screening titles, abstracts, and full texts according to predefined inclusion and exclusion criteria, 85 studies were selected for detailed analysis. Information regardingwas systematically extracted and comparatively synthesized, including ecosystem service models, categories, validation approaches, data sources, performance indicators, uncertainty analysis, spatial and temporal scales, challenges, and key findings.
Results and Discussion: The results of the review of 85 studies indicated that no single model can be called suitable for assessing all ecosystem services, and that model selection depends on the type of service, the spatial and temporal scale of the study, and data availability. Among the models examined, InVEST was the most frequently applied, being used in 9 studies (10.7%), followed by ARIES, LUCI/Polyscape, SWAT, and SolVES, which were used in 3 (3.6%), 3 (3.6%), 2 (2.4%), and 2 (2.4%) studies, respectively. An examination of validation methods showed that field data (10 studies, 11.9%), remote sensing data (9 studies, 10.7%), and hybrid approaches (8 studies, 9.5%) were the most important methods for evaluating model performance. In addition, only 11 studies (13.1%) explicitly conducted uncertainty analysis, highlighting the need to develop standardised frameworks for managing uncertainty in ecosystem service models. The use of emerging technologies has also increased; specifically, 12 studies (14.3%) employed remote sensing, 5 studies (6.0%) used machine learning, and 3 studies (3.6%) applied artificial intelligence methods. Overall, our findings suggest that integrating ecosystem service models with multi-source data, emerging technologies, and comprehensive validation methods can enhance the reliability of model outputs.
Conclusion: This study demonstrates that ecosystem service model validation should be regarded as an integral component of ecosystem service modeling rather than a post-modeling procedure. Reliable validation requires the integration of appropriate model selection, independent and high-quality datasets, uncertainty assessment, multi-source validation approaches, and transparent reporting of methodologies and results. The principal contribution of this research extends beyond a conventional systematic review. It lies in synthesizing existing knowledge, critically comparing ecosystem service models, identifying key determinants of successful validation, highlighting current research gaps, and proposing a comprehensive conceptual framework for standardizing ecosystem service model validation. The findings provide valuable guidance for future research and offer practical support for researchers, environmental managers, and policy-makers seeking to improve ecosystem service assessments, natural resource management, biodiversity conservation, land-use planning, and evidence-based environmental decision-making.
Keywords
