We use weakly supervised structured learning to track and disambiguate the identity of multiple indistinguishable, translucent and deformable objects that can overlap for many frames. For this challenging problem, we propose a novel model which handles occlusions, complex motions and non-rigid deformations by jointly optimizing the flows of multiple latent intensities across frames. These flows are latent variables for which the user cannot directly provide labels. Instead, we leverage a structured learning formulation that uses weak user annotations to find the best hyperparameters of this model.The approach is evaluated on a challenging dataset for the tracking of multiple Drosophila larvae which we make publicly available. Our method tracks multiple larvae in spite of their poor distinguishability and minimizes the number of identity switches during prolonged mutual occlusion.

Historically, humans have cleared many forests for agriculture. While this substantially reduced ecosystem carbon storage, the impacts of these land cover changes on terrestrial gross primary productivity (GPP) have not been adequately resolved yet. Here, we combine high-resolution datasets of satellite-derived GPP and environmental predictor variables to estimate the potential GPP of forests, grasslands, and croplands around the globe. With a mean GPP of 2.0 kg C m−2 yr−1 forests represent the most productive land cover on two thirds of the total area suitable for any of these land cover types, while grasslands and croplands on average reach 1.5 and 1.8 kg C m−2 yr−1, respectively. Combining our potential GPP maps with a historical land-use reconstruction indicates a 4.4% reduction in global GPP from agricultural expansion. This land-use-induced GPP reduction is amplified in some future scenarios as a result of ongoing deforestation (e.g., the large-scale bioenergy scenario SSP4-3.4) but partly reversed in other scenarios (e.g., the sustainability scenario SSP1-1.9) due to agricultural abandonment. Comparing our results to simulations from state-of-the-art Earth System Models, we find that all investigated models deviate substantially from our estimates and from each other. Our maps could be used as a benchmark to reduce this inconsistency, thereby improving projections of land-based climate mitigation potentials.

When tracking and segmenting multiple objects under heavy occlusion, a large class of algorithms can greatly benefit from a preprocessing that reliably assesses the number of individuals in each cluster. This is a difficult task when relying on local information only, due to scarcity of training examples and lack of strongly predictive features. In this paper, we develop a deterministic graphical model to address the problem of counting the number of objects in each foreground region as global inference across the entire video sequence. We show that global inference improves over local predictions, and is able to produce an accurate and coherent output within an useful runtime.

Forests mitigate climate change by storing carbon and reducing emissions via substitution effects of wood products. Additionally, they provide many other important ecosystem services (ESs), but are vulnerable to climate change; therefore, adaptation is necessary. Climate‐smart forestry combines mitigation with adaptation, whilst facilitating the provision of many ESs. This is particularly challenging due to large uncertainties about future climate. Here, we combined ecosystem modeling with robust multi‐criteria optimization to assess how the provision of various ESs (climate change mitigation, timber provision, local cooling, water availability, and biodiversity habitat) can be guaranteed under a broad range of climate futures across Europe. Our optimized portfolios contain 29% unmanaged forests, and implicate a successive conversion of 34% of coniferous to broad‐leaved forests (11% vice versa). Coppices practically vanish from Southern Europe, mainly due to their high water requirement. We find the high shares of unmanaged forests necessary to keep European forests a carbon sink while broad‐leaved and unmanaged forests contribute to local cooling through biogeophysical effects. Unmanaged forests also pose the largest benefit for biodiversity habitat. However, the increased shares of unmanaged and broad‐leaved forests lead to reductions in harvests. This raises the question of how to meet increasing wood demands without transferring ecological impacts elsewhere or enhancing the dependence on more carbon‐intensive industries. Furthermore, the mitigation potential of forests depends on assumptions about the decarbonization of other industries and is consequently crucially dependent on the emission scenario. Our findings highlight that trade‐offs must be assessed when developing concrete strategies for climate‐smart forestry.

Over the last six decades the terrestrial biosphere has sequestered on average about 29% of the anthropogenic CO 2 emissions each year (Friedlingstein et al., 2022). While large part of this net carbon sink is likely driven by elevated CO 2 concentrations (Fernández-Martínez et al., 2019), many other factors influence the uptake capacity of the land, including variations in temperature and water availability, which are expected to change with global warming. Many of these effects and their implications for carbon dynamics and vegetation distribution are not well quantified. The effects of higher temperatures and higher CO 2 concentrations, for example, may counteract each other (Peñuelas et al., 2017). Reduced productivity due to higher evaporative demand and stomatal closure (Friend et al., 2014) as a consequence of higher temperatures may be compensated by increased water use efficiency