<p>Uncontrolled exploitation of natural resources as human needs increase over time triggers an ecological environment increasingly vulnerable. Spatial evaluation of ecological vulnerability is an effort to reduce the adverse impacts of internal and external disturbances on the ecosystem. We present a comprehensive and systematic evaluation system and analysis method. The Mondangan Watershed is located in Wonosari Sub-village, Tugurejo Village, Wates District, Blitar Regency, East Java, Indonesia. The Mondangan watershed represents a complex coastal ecosystem of natural and secondary forests associated with anthropogenic disturbances such as agricultural activities and constructing a new national road. Based on the ecological vulnerability framework such as sensitivity-resilience-pressure (SRP), 15 relevant indicators were selected. Their weighting was carried out using a hybrid method with high objectivity in building a spatial model of ecological vulnerability. Fuzzy evaluation is used to represent indicator data realistically and mapped to calculate the ecological vulnerability index (EVI) evaluation score. Spatial autocorrelation analysis is carried out to obtain a clear spatially aggregated ecological vulnerability map. Finally, the ecological vulnerability map under the control of the weighting coefficient is carried out based on the ordered weighting average (OWA) operator. The results show that the ecological vulnerability of the Mondangan Watershed shows a high class in the upstream to the east and a low class in the middle and extends to the south. There is a clear spatial aggregation effect. The high-high spatial aggregation zone is distributed upstream and extends to the east and south, while the low-low aggregation is in the middle area and extends to the west and south. NPP (net primary production), soil erosion, annual precipitation, and NTL (nighttime light) have the highest contribution rates to the EVI model. The increase in the weighting coefficient represents the gradual increase in the ecological vulnerability of the Mondangan Watershed. The findings of this study provide an initial reference in socio-economic development and ecological protection to achieve sustainable development in the Mondangan Watershed. The EVI assessment framework in this study offers reference value, especially for studies in small area coverage with coastal characteristics.</p> Graphical Abstract <p></p> <p>The graphical abstract describes an overview of the study of ecological vulnerability (EVI) modeling in the Mondangan Watershed, Blitar Regency, East Java Province, Indonesia using multi-source remote sensing data. A total of 15 geospatial indicators were used to model EVI in the study area, as presented in the graphical abstract. All geospatial data were converted into 10-meter resolution rasters to maintain consistency, then data normalization was based on their effects on EVI (positive or negative). In turn, all indicators were superimposed to perform principal component analysis (10 PCs) to produce a decision matrix used to calculate information entropy, so that the indicator weights were known. Spatial fuzzy raster was performed to polish the interval boundaries naturally on all indicators using triangular linear (high, medium, and low) and trapezoidal linear (highest and lowest) functions, and defuzzification was performed to convert fuzzy values into crisp values (expected scores are 100, 80, 60, 40, and 20). Finally, the weight value of PCA-entropy is applied to each indicator that has gone through the fuzzification and defuzzification process to produce an EVI map. The results show that the upstream part of the watershed to the volcanic hills (settlement and dryland agriculture) has high ecological vulnerability compared to the middle part of the watershed to the south (more woodlands). This is reinforced by the results of the LISA analysis which shows high-high and low- low clusters in the mentioned areas with statistical significance of 95–99%. NPP, soil erosion, annual precipitation, and NTL have the greatest contribution to the EVI model. The OWA operator analysis aims to build 8 indicator weighting scenarios for the EVI map, i.e. the smallest gamma value produces a very optimistic EVI map (all areas are low vulnerability), the vulnerability class varies increasingly closer to the original when the gamma value approaches 1, and finally the vulnerability class gradually changes to high vulnerability (pessimism) when the gamma value is more than 1. The OWA results help policy makers for socio-economic development and ecological protection.</p>