Urban forests are widely recognised as a nature-based solution to mitigate the effects of climate change; however, urban forests are also vulnerable to climate change. Therefore, there is a need to improve species selection to ensure the delivery of ecosystem services by urban forests now and in the future. Research on the impacts of climate change on urban forests requires investigation to capture the complexities associated with species identity and growing conditions. Yet, such studies remain rare in urban contexts, highlighting the need for expanding collaborative research in cities. Here, we present a nation-wide urban trial network established across four states in Australia, showcasing stakeholder collaboration aimed at advancing urban forest research. The network consists of 11 standardised plantings of tree and/or shrub species aimed at testing species' growth and performance (i.e., stress tolerance) in cities across a range of climatic conditions. To test these differences, we measured height and diameter relative growth rates (RGR) and leaf damage caused by stress at each site one month after planting (2018-2020) and at the end of the austral summer in 2024. We used generalised linear mixed-effects models for RGR and ordinal logistic regressions for leaf damage to test the effects of annual maximum temperature (TMAX) and the Pinna Combinative Index (IP, a climate-drought index). By 2024, across all sites, we found 23 % of the originally planted individuals had died or were missing. We recorded significant differences in height and diameter RGR and leaf damage among sites, and IP was significantly and negatively related to both RGR and leaf damage. The network serves as an example of how stakeholder collaboration can broaden the scope of urban forest research that evaluates plant growth and performance across regions and environmental conditions.
Tree failure can pose significant challenges to green-infrastructure planning for potentially jeopardizing ecosystem services provision, infrastructure safety, and citizens' well-being. The city-wide disturbance caused by the loss of over 2000 trees annually in Sa similar to o Paulo, Brazil, impelled local authorities to collect detailed field-data on tree failure from 2016 to 2018 at the city center, a hotspot of tree failure, and then engage with the academia to support risk management. We aimed at building predictors and defining guidelines to reduce branch, trunk, and root failure based on species, wood status, root collar constrictions, conflicts with overhead cables, pruning methods, and site characteristics of 456 trees using Classification Trees and Bagging. These algorithms commonly used in decision-making yielded up to 70% accuracy, identifying wood status, root collar constrictions, and pruning as the main predictors. Branch failure represents 46% of the dataset. In the absence of wood degrada-tion, branches were the most likely mode of failure. Root failure comes next representing 33% of the dataset, common to trees without wood degradation but with constricted root collars by pavement, compacted soil, or girdling roots. Root failure also dominates in trees with clear signs of wood decay and trunk cavities. Trunk failure only represents 21% of the events, common to trees with wood decay and subject to poor pruning practices. Thus, effective management of trees requires a collaborative approach to collecting data, analyzing, and establishing roles and guidelines. This study points to the role of local authorities in undertaking a detailed assessment of trees' wood status throughout the city, while the municipality and private companies responsible for their management must adopt appropriate pruning practices. Lastly, those engaged in planting trees must guarantee enough space for the root collar to grow. Neglecting these guidelines can incur the cost of twice as much damage to the city.