Multi-criterion assessment of the priority of environmental protection measures based on open online data

Abstract

The purpose of the study is to develop and test a methodology for multi-criteria assessment of environmental protection priorities using open online data under conditions of incomplete information and interval uncertainty. The study addresses a practical problem in environmental management: how to justify priority actions when several measures are relevant simultaneously, and the available data have different spatial resolutions, reliability, and completeness. The research uses systems analysis, multi-criteria assessment, an additive utility function, expert weighting of criteria, normalization of heterogeneous indicators, interval analysis, and scenario analysis. Open online sources were selected as the information basis: NASA FIRMS for active fires, JRC Global Surface Water for water dynamics, ESA WorldCover and Sentinel-2/Copernicus Data Space for land cover and vegetation indicators, OpenAQ for air quality, HDX for administrative boundaries, Global Forest Watch for tree cover loss, and EkoZagroza as an auxiliary source of environmental threat reports. A pilot computational experiment was conducted for a conditional urbanized area in the Kharkiv region. Six environmental alternatives were assessed: greening, cleaning water bodies, reclamation of disturbed lands, strengthening air monitoring, improving waste management, and fire-prevention measures. Six criteria were used: environmental effect, implementation cost, implementation time, social significance, risk of inaction, and availability of online data. The highest baseline utility scores were obtained for waste management improvement, Uc = 0.650, and fire-prevention measures, Uc = 0.642. The following priorities were water body cleaning, Uc = 0.530, greening, Uc = 0.523, strengthening of air monitoring, Uc = 0.497, and reclamation of disturbed lands, Uc = 0.359. Scenario analysis showed that the first two alternatives remain the highest-priority group under different weighting assumptions. The proposed methodology enables formalizing preliminary environmental decision-making even when online data are incomplete. Interval representation does not eliminate uncertainty but makes it explicit and useful for assessing the reliability of ranking results. The method can support the preliminary justification of local environmental protection programs and should be further tested using actual datasets for specific territorial communities.