Generative AI (GenAI), particularly text-to-image (TTI) models, is reshaping landscape representation by transforming textual descriptions into visual outputs. However, these models often reinforce biases embedded in their training datasets, shaping how landscapes are perceived and represented. This research examines the biases in GenAI-generated landscape imagery through the lens of Edward Said’s “imaginative geographies”, focusing on how geographic references, cultural archetypes, and methodological factors influence AI outputs. We employed a structured approach to create prompts based on the Global Ecosystem Typology (GET), using Midjourney V6.1 as the primary tool for image generation. We extracted landscape descriptors from GET classifications and structured them into both simple and detailed prompts. The analysis involved comparing AI-generated images to ecological classifications and reference images to assess biases. The findings reveal the following three key types of biases: (1) geographic biases, where certain locations act as semantic triggers, leading to culturally stereotypical portrayals; (2) representation biases, where landscapes frequently depicted in AI training datasets appear with greater accuracy, while underrepresented landscapes are simplified into essentialized visual tropes; and (3) methodological biases, where prompt complexity influences representational accuracy but does not eliminate pre-existing cultural hierarchies. To address the bias challenges, the research presents the following four key recommendations for future research and practice: (1) incorporating finer-scale ecosystem classifications; (2) diversifying training datasets; (3) engaging local communities in participatory data collection; and (4) refining prompt-writing methodologies. These insights contribute to broader discussions on AI bias, emphasizing the necessity of critically evaluating the role of generative models in shaping landscape imaginaries.

Armed conflicts and resulting displacement disrupt people’s sense of place, leading to an imbalance in the people–place relationship, exaggerated by rehabilitation efforts that overlook the sense of place among conflict- and displacement-impacted communities. A continuous landscape narrative that extends from pre- to post-conflict times contributes to recreating the essence of lost landscapes and therefore reconnecting their sense of place. Focusing on a Syrian city that hosted internally displaced persons (IDPs) in the aftermath of the Syrian conflict, this study aims to structure a virtual landscape and narrative depiction of conflict-impacted landscapes. This study proposes a storytelling approach for narrative construction and an AI-powered visualisation approach to revive the image of the elusive landscapes. This study utilised qualitative research methods through in-depth, semi-structured interviews for data collection and an online survey for exploring the perception of virtual landscape images generated with AI text-to-image models (DALL.E 2 and Bing Image Creator). This study indicates that narratives, supported by AI visualisation, are reliable for comprehending landscape transformation and changes in the sense of place. The two approaches can serve as rehabilitation initiatives in post-crisis settings to recall images of elusive landscapes to save them from being lost forever.

Landscape Character Assessments (LCAs) support planning decisions by offering structured descriptions of landscape character. However, producing these texts is often resource-intensive and shaped by subjective judgement. This study explores whether Generative Artificial Intelligence (GenAI), specifically ChatGPT, can support the drafting of LCA descriptions using a structured, prompt-based framework. Applied to Harku Municipality in Estonia, the method integrates spatial input, reference material, and standardised prompts to generate consistent descriptions of landscape character areas (LCAs) and facilitate scenario building. The results show that ChatGPT outputs align with core LCA components and maintain internal coherence, although variations in terminology and ecological specificity require expert review. A stakeholder role play using ChatGPT highlighted its potential for enhancing early-stage planning, education, and participatory dialogue. The limitations include the reliance on prompt quality, static inputs, and the absence of real-time community validation. Recommendations include piloting AI-assisted workflows in education and practice, adopting prompt protocols, and prioritising human oversight, both experts and stakeholders, to ensure contextual relevance and build trust. This research proposes a practical framework for embedding GenAI into planning processes while preserving the social and interpretive dimensions central to landscape governance.

In an era of systemic risks from natural and human-induced hazards, we need approaches that address such risks at different governance levels, including those at the community level. Citizen science (CS)—public participation in research—is one such approach that has the potential to benefit risk assessment and management. Building on systematic evidence synthesis methodologies, we examine the individual and collective benefits from CS to environmental risk assessment and management. From systematic search results (9277 records), we found 133 publications dealing with the topic in-depth. Results show that CS is increasingly being recognised for risk assessment and management and several individual-level benefits (e.g., improved scientific skills) support outcomes for communities (e.g., increased community capacity). Identified knowledge gaps include (i) unidentified potential of CS for different risk governance phases, and (ii) scattered evidence of CS community outcomes in different geographic regions. We conclude that CS can increase resilience amongst vulnerable communities.

Background: The deleterious effects of built environments on mental wellbeing are a global health concern, especially impacting deprived neighbourhoods. Green and blue spaces (GBS) are increasingly recognised as health assets, but the role of their aesthetic qualities in health protection remains unclear.Methods: Combining large-language modelling and machine learning, we developed an automated scoring system to score GBS photos based on the Contemplative Landscape Model (CLM). The tool was applied to >17,000 GBS photos from nine European study-areas identified in the GreenME Horizon Europe project for high rates of socioeconomic deprivation. These scores were used to distinguish two groups of aesthetic landscape features: CLM-static, including long-distance views, natural topography, diversified skyline, and archetypal elements such as water bodies and large solitary trees, and CLM-dynamic, such as richness in species. We analysed data from 6,173 neighborhood residents who provided GBS-visit locations and wellbeing self-reports as part of the GreenME survey. Ordinal mixed-effects regression linked GBS landscape qualities with wellbeing. Models were validated through sensitivity analysis.Findings: CLM-static features were consistently associated with greater Life Satisfaction, Life Meaning and Happiness, while CLM-dynamic features were detrimentally associated with Life Satisfaction only. For life satisfaction, these effects were attenuated among individuals with higher perceived nature access. Importantly, findings were robust across sensitivity analyses, independent of socio-demographic variables, and demonstrated relatively low risk of unmeasured confounding.Interpretation: Preserving and introducing CLM-static features may support widespread health protection. Urban planning strategies focusing narrowly on CLM-dynamic features may not, on their own, deliver broader mental health benefits. Strategic integration of scenic landscape features into public health and urban planning may support population health and wellbeing, especially in the context of deprived neighbourhoods.