**Background:** The COVID-19 pandemic, declared on March 11, 2020, led to global restrictions that negatively impacted psychosocial health and health behaviors, including declines in psychological well-being, increases in loneliness, reduced physical activity, and adverse dietary changes. Prior to the pandemic, extensive evidence demonstrated that exposure to natural environments (green space, blue space, gardens) improves mental health, well-being, and promotes physical activity. However, no review had synthesized the global evidence on the role of natural environments for health during a public health crisis like COVID-19. This scoping review aimed to map the available research on the relationship between natural environments and psychosocial health as well as health-related behaviors in the COVID-19 context.
**Methods:** The review followed the Arksey and O'Malley five-step framework and PRISMA-ScR reporting guidelines. A systematic search was conducted in Web of Science, Scopus, PubMed, Embase, CINAHL, and Greenfile on April 14, 2021, and updated on July 14, 2022. Search strings combined terms related to natural environments (e.g., green space, park, digital nature) and COVID-19 (e.g., Sars-Cov-2, pandemic, lockdown). Inclusion criteria were: published since 2020 with data collected in the COVID-19 context; peer-reviewed; original empirical data on human participants; investigated the association between natural environments and psychosocial health or health behaviors; written in English, German, or Scandinavian languages. Out of 9126 articles screened, 188 articles (representing 187 studies) were included. Two reviewers independently screened titles, abstracts, and full texts. Data extraction covered authors, year, location, sample characteristics, study design, methodology, nature operationalization, health outcomes, and main findings. A systematic thematic analysis using deductive and inductive coding was performed in MAXQDA.
**Key Results:** Most studies used quantitative methods (n=132), were cross-sectional (n=150), and observational (n=171). Data collection peaked in April (n=60), May (n=54), and June (n=47) of 2020. Studies were predominantly conducted in the USA (n=36), UK (n=18), China and Spain (n=16 each), Italy (n=15), Canada (n=13), Germany (n=11), and Australia (n=10). The general population was the most common sample (n=84), followed by urban residents (n=42) and university students (n=15). Vulnerable populations were underrepresented (healthcare workers n=5, people with physical health problems n=6, mental health problems n=4, nursing home residents n=2). Adults (n=158) and older adults (n=128) were studied far more than children (n=20) or adolescents (n=24). Ethnicity was not reported in 149 studies.
Three main themes emerged from thematic analysis: 1) Nature types investigated: Public nature (n=111 codes, especially parks n=33 and urban natural areas n=28) dominated over private nature (n=32 codes, gardens n=25). Green space/vegetation (n=24) and general nature (n=21) were most common; digital nature (n=8), nature quality (n=6), and nature sounds (n=3) were rarely studied. Gardening was the most frequent nature-based activity (n=17). 2) Health outcomes and behaviors: Psychological health received the most attention (n=325 codes), with well-being (n=108), stress (n=79), mood/emotions (n=51), and depression/anxiety (n=42) most studied. Health behaviors (n=102 codes) focused heavily on physical activity (n=80 of 88 movement behavior codes). Social health was least studied (n=68 codes). Across all outcomes, 423 codes showed favorable associations with nature versus 72 showing null or unfavorable associations. 3) Heterogeneity: The nature–health relationship varied by human characteristics (n=56 codes, especially sociodemographics n=34), nature characteristics (n=19 codes, especially nature type n=13), and geographic region (n=11 codes).
**Clinical Implications:** The findings suggest that natural environments are a valuable public health resource that can mitigate the negative psychosocial and behavioral impacts of a crisis like COVID-19. The replication of pre-pandemic findings during a global crisis strengthens the evidence that nature exposure supports mental health and physical activity even under stressful conditions. However, the predominance of cross-sectional designs limits causal inference. The review identifies critical research gaps: the need for studies on vulnerable populations (who may benefit most), investigation of specific nature characteristics and quality indicators, research on digital/virtual nature for populations with limited access, examination of health behaviors beyond physical activity (e.g., diet, sleep), and longitudinal studies to assess long-term effects. The authors recommend developing a 'FITT principle' for human–nature interactions (frequency, intensity, time, type) to guide urban planning and interventions. Given that societal crises (financial, climate-related, future pandemics) are likely to recur, natural environments should be considered a resource for building resilience before, mitigating negative impacts during, and promoting health after crises.