Improving the climate resilience of European cities via socially acceptable nature-based solutions

The paper addresses how to improve the climate resilience of European cities through nature-based solutions (NBS) that are socially acceptable. With urbanization rising to an expected 80% in Europe by 2030, resilient urban planning is critical. NBS are defined by the European Commission as nature-inspired solutions that are cost-effective and provide environmental, social, and economic co-benefits while building resilience. Traditional assessments of mitigation and adaptation measures emphasize technical and economic feasibility, but successful implementation also requires social acceptance. Social acceptance has often been relegated to late planning stages, leading to a social gap where acceptance factors are not integrated into design. NBS implementations occur within complex socio-ecological systems with uncertain responses, long time horizons, and multi-stakeholder collaboration requirements, making trust, inclusivity, and transparency crucial. Existing acceptance studies are fragmented and cross-disciplinary, complicating comparisons. This paper proposes a comprehensive, dynamic, and adaptable framework grounded in behavioral theories to quantify and compare social acceptance of NBS and integrate these insights into planning and sustainability assessments. Four diverse NBS cases (Ankara, Szeged, Alcalá de Henares, Milan) are used to demonstrate how data-driven evidence can support tailored policy instruments and management strategies that reflect socio-spatial and cultural contexts.

The paper synthesizes cross-disciplinary literature on social acceptance and adapts it to NBS. Social acceptance is variously defined, ranging from positive attitudes or behaviors at a point in time to welfare-based definitions balancing costs and benefits. The authors adopt Upham et al.’s broader definition emphasizing favorable attitudes, intentions, behaviors, and use within a defined social unit. Wüstenhagen et al.’s three dimensions—socio-political, community, and market acceptance—are applied to NBS. Given the examined NBS are implemented or in progress, community acceptance is the focal dimension. The theoretical model draws on the Theory of Planned Behavior (attitudes, subjective norms, perceived behavioral control), affect theories (positive and negative affect as independent drivers of attitudes), and Norm Activation Theory (personal norms shaped by awareness of adverse consequences and outcome efficacy). Additional determinants include trust in implementers (shaping perceived risks, costs, benefits), procedural fairness (inclusive decision-making), distributive fairness (fair distribution of costs/benefits), knowledge, and experience. The framework adapts and extends Huijts et al.’s renewable energy acceptance model to the specific context and sensitivities of NBS.

The study develops a replicable decision-support framework that incorporates social acceptance into NBS planning. Steps include: (1) literature review to identify common values and challenges across NBS types; (2) identification of situational factors through field research and public consultations; (3) translation of challenges into survey items (drawing on Nature4Cities typology of 75 NBS types) and tailoring items to pre- or post-implementation contexts; (4) data collection via zoned random sampling of local communities exposed to each NBS; and (5) quantitative analysis to identify key acceptance drivers. The structural model operationalizes latent constructs including trust, perceived benefits, risks, costs, affects (positive/negative), personal and social norms, procedural and distributive fairness, knowledge, experience, problem perception, and outcome efficacy. Partial Least Squares Structural Equation Modeling (PLS-SEM) using SmartPLS 3.0M2 tests the hypothesized relationships. Target sample sizes were 200–300 respondents per case. Bootstrapping (1,000 resamples) was used to obtain t-statistics and p-values. Measurement model evaluation followed standard criteria: standardized loadings >0.4 (items below removed), Cronbach’s alpha >0.7, composite reliability >0.7, AVE >0.5 for convergent validity, and Fornell–Larcker criterion for discriminant validity. Inner VIF values were <3 (below recommended maximum of 5), indicating no multicollinearity. Common method bias was assessed via full collinearity VIF (≤3.3) and Harman’s single-factor test (first factor explained 12–20.7% variance, <50% threshold), suggesting CMB was not a threat. The framework and analyses were applied to four cases: METU Forest (Ankara, Turkey), Tisza River Bank (Szeged, Hungary), Forest Garden (Alcalá de Henares, Spain), and Quarries (Milan, Italy).

Across four European NBS cases, social acceptance is shaped by fairness (procedural and distributive), trust, perceived risks/costs/benefits, knowledge, experience, and personal norms, with context-specific variations. Perceived benefits emerged as a common direct driver across cases. Case-specific results: 1) Tisza River Bank (Szeged, pre-implementation): Trust increased positive affect (r=0.529, p=0.001) which raised acceptance (r=0.274, p=0.001); trust decreased negative affect (r=-0.388, p=0.001) which lowered acceptance (r=-0.146, p=0.001). Lower trust raised perceived cost (r=-0.263, p=0.001) and risk (r=-0.271, p=0.001), both reducing acceptance (cost→acceptance r=-0.146, p=0.001; risk→acceptance r=-0.117, p<0.05). Trust increased perceived benefits (r=0.331, p=0.001), which increased acceptance (r=0.119, p=0.05). Procedural fairness increased trust (r=0.324, p=0.001) and directly increased acceptance (r=0.145, p=0.01). Distributive fairness directly increased acceptance (r=0.136, p=0.01). Experience increased knowledge (r=0.244, p=0.001); knowledge→trust was not significant. Personal and social norms did not significantly predict acceptance. 2) Quarries (Milan, mixed implementation stages): Trust increased positive affect and benefits (r=0.367, p=0.001; r=0.288, p=0.001) and decreased negative affect and risks (r=-0.301, p=0.001; r=-0.222, p=0.01); these translated to acceptance. Perceived benefits influenced acceptance indirectly via personal norms; perceived risks, costs, and outcome efficacy also influenced acceptance via personal norms. Procedural fairness increased trust (r=0.138, p=0.1), but neither procedural nor distributive fairness directly affected acceptance. Experience increased knowledge (r=0.392, p=0.001); knowledge→trust was not significant. 3) Forest Garden (Alcalá de Henares, post-implementation): All hypothesized trust outcomes held—trust positively related to positive affect and benefits, and negatively to negative affect, cost, and risk. Of these, only positive affect, perceived cost, and perceived benefits significantly predicted acceptance; negative affect and risk did not. Procedural fairness influenced acceptance both directly and indirectly via trust. Experience→knowledge→trust pathway was supported: more experience increased knowledge, which increased trust. Personal norm significantly increased acceptance; among its antecedents, only perceived risk and perceived benefits significantly increased personal norms; social norm had no impact. 4) METU Forest (Ankara, post-implementation): Trust increased positive affect and benefits; only benefits significantly increased acceptance. Trust decreased negative affect and risk; only risk significantly decreased acceptance. Experience increased knowledge, which increased trust; procedural fairness increased trust (indirectly supporting acceptance). Personal norm did not significantly predict acceptance; social norm had a marginal effect per tabulated results. Overall, trust consistently shaped cognitions and affect, with procedural fairness a robust antecedent of trust. Perceived benefits are the only universal direct driver of acceptance identified across all cases.

The findings demonstrate that quantifying social acceptance and integrating it into NBS planning is feasible and highly informative. The proposed framework reveals that acceptance drivers vary by context and project stage. In pre-implementation settings (e.g., Szeged), trust and fairness are paramount because communities lack direct experience, making perceptions of process inclusivity and fair distribution of impacts central to acceptance. In post-implementation settings (e.g., Ankara, Alcalá de Henares), experience and knowledge become critical through their influence on trust, and cognitions (perceived risks and benefits) may outweigh affect in shaping acceptance. Procedural fairness consistently builds trust, while distributive fairness directly supports acceptance in some contexts. Policy implications include: fostering inclusive, transparent decision-making; targeted communication to clarify risks, costs, and co-benefits; facilitating experience and knowledge-building (site visits, virtual tours, educational outreach); and leveraging feedback loops to iteratively refine NBS designs. Embedding the framework into planning enables early identification of local barriers and tailored strategies, thereby enhancing long-term public support and climate resilience outcomes.

This study contributes a dynamic, data-driven framework to measure and compare social acceptance of NBS and to embed acceptance evidence into planning and decision-making. Applying the framework in four European cities shows that procedural and distributive fairness, trust, perceived risks/costs/benefits, knowledge, experience, and personal norms are key, with perceived benefits the only universal direct driver of acceptance. The framework supports replicable, comparable analyses to guide tailored interventions, stakeholder engagement, and communication strategies that enhance climate resilience. Future research should expand to include group-level perception formation dynamics alongside individual perceptions, extend applications to mitigation and low-carbon technologies, and refine stage-specific strategies (pre- vs post-implementation) to optimize acceptance trajectories over time.

The study focuses on community acceptance near each NBS and does not generalize to national populations. Sampling targeted residents exposed (or to be exposed) to each NBS via zoning; response-rate challenges inherent to surveys remain. Some determinants (e.g., knowledge→trust) vary with implementation stage, limiting cross-context generalizability. While common method bias checks indicate minimal risk, reliance on self-reported perceptual data is an inherent limitation. The framework identifies significant paths in specific contexts; transferability requires re-validation in new settings and NBS types.