Agro ParisTech

Photo-interpretation is widely used for land-use mapping but has not yet been operationally applied to agroforestry systems. One way to monitor these systems is to collect georeferenced samples at fine spatial scales, which can be used to evaluate existing land-use maps, develop agroforestry inventories, and train machine learning models for large-scale mapping. We propose a photo-interpretation-based classification method to identify three agroforestry types—hedgerows, alley cropping, and scattered agroforestry—and two non-agroforestry categories: woody monocrops and unmanaged trees outside forest (TOF)/forest trees (FT). Five interpreters tested the method on 300 georeferenced sample points distributed across multiple global regions. Overall accuracy ranged from 0.75 to 0.81, depending on the interpreter, increasing to 0.86 after adjusting for interpreter errors. Hedgerows were consistently well classified (2% omission, no commission error), whereas alley cropping and woody monocrops were more frequently confused. Identifying scattered agroforestry systems proved more challenging, with 19% of points misclassified as unmanaged TOF/ FT. After validation, the method was applied at a global scale to produce 7,131 georeferenced sample points representing four tree-based agricultural systems: alley cropping, hedgerows, scattered agroforestry, and woody monocrops. Overlaying these points with the ESA Above-Ground Biomass Map (2020) revealed distinct above-ground carbon (AGC) median densities: woody monocrops (9.7 MgC ha⁻¹, alley cropping (7.3), scattered agroforestry (4.2), and hedgerows (2.9). These results demonstrate that the proposed classification method, combined with photo-interpretation, can effectively generate large-scale georeferenced agroforestry datasets, support map validation and agroforestry inventories, and enable scalable carbon monitoring.

Water stress can cause declines in plant function that persist after rehydration. Recent work has defined 'resilience' traits characterizing leaf resistance to persistent damage from drought, but whether these traits predict resilience in whole-plant function is unknown. It is also unknown whether the coordination between resilience and 'resistance' - the ability to maintain function during drought - observed globally occurs within ecosystems. For eight rainforest species, we dehydrated and subsequently rehydrated leaves, and measured water stress thresholds for declines in rehydration capacity and maximum quantum yield of photosystem II (Fv /Fm ). We tested correlations with embolism resistance and dry season water potentials (ΨMD ), and calculated safety margins for damage (ΨMD - thresholds) and tested correlations with drought resilience in sap flow and growth. Ψ thresholds for persistent declines in Fv /Fm , indicating resilience, were positively correlated with ΨMD and thresholds for leaf vein embolism. Safety margins for persistent declines in Fv /Fm , but not rehydration capacity, were positively correlated with drought resilience in sap flow. Correlations between resistance and resilience suggest that species' differences in performance during drought are perpetuated after drought, potentially accelerating shifts in forest composition. Resilience to photochemical damage emerged as a promising functional trait to characterize whole-plant drought resilience.

Urban nature has played an important role in cities around the world since industrialisation in the 18th century. The responsibility for urban nature has since then primarily lied with local governments, however with no formal obligation to develop or manage such areas. A number of concepts and approaches have been suggested over past decades to address planning and management of urban nature, but to gain maximum benefit of urban nature in forms of e.g. Green Infrastructure and Urban Forests, there are still a number of generic challenges to be overcome. These challenges includes fragmented organizational structures leading to disconnections between planning and management. Also governance related dimensions are often perceived as limiting factors for the improvement of urban nature. This chapter describes approaches to future management of Biocities in forms of Strategic Management and Adaptive Management, while also introducing new governance approaches. Three cases exemplifies novel local government approaches to inclusive governance and management of the urban nature—the future BioCities.