Why do forests keep disappearing despite conservation interventions? The answer requires an unobservable counterfactual: a world in which these interventions never occurred. For 28 countries that account for 87% of global deforestation, we simulate counterfactual deforestation driven solely by agricultural profits. Under this scenario, deforestation rates would have increased by more than twice the observed rate relative to 2001-2004, so deforestation reduced relative to the counterfactual. However, correlations between reductions in deforestation in some countries and additional deforestation in others suggest that 43% of the reductions we have identified (23.9 ± 0.744 million hectares) were offset by leakage. If all 28 countries had achieved reductions comparable to the six most successful, global deforestation would have been 15.8% lower than in 2001-2004. Eliminating profit-oriented deforestation would require annual conservation payments of US$59 billion (<1% of sample GDP). Our framework enables transparent evaluation of conservation progress based on a parsimonious land-change model and supports monitoring at continental and global scales.

The crown of a tree determines current growth and reflects past growth patterns, competition and ecological stress factors such as drought, nutrient deficiency, or competitive pressure. In the context of climate change, assessing the acclimatization potential of trees to drought stress is crucial. This study investigates the relation of crown morphology with the basal area increment of 151 Norway spruces (Picea abies L. Karst), 151 Scots pines (Pinus sylvestris L.), 217 European beeches (Fagus sylvatica L.) and 61 oaks (Quercus petraea (Matt.) Liebl./Quercus robur L.) in southern Germany. Using terrestrial laser scans, eleven metrics describing the crown and competition were calculated from the tree point clouds. We linked these with tree-ring widths of increment cores to model basal area increment and relative growth decline during the Central European drought period 2015-2022. Using correlation analyses, we investigate the strength of different crown traits' association with growth over periods of different lengths (1-30 years). We found that trees develop species-specific crowns exhibiting temporally differentiated structural memory (i). Crown morphology is thus not only functional but also retrospectively interpretable. Furthermore, we found a considerable influence of various crown metrics on the drought stress response of trees (ii). However, the nature of the influence is highly differentiated and suggests a nonlinear stress response and trade-offs in acclimation. Our study highlights the relevance of crown morphology in drought response and provides a basis for further investigations into tree resilience and forest acclimation.

Heavy thinning from above promotes coarse root and stem diameter growth in both Scots pine and sessile oak. Coarse root-stem allometry indicates a greater allocation of resources to coarse roots relative to the stem in both species. The effects of heavy thinning from above on stem and coarse root growth, as well as on coarse root–stem allometry, were investigated in mature Scots pine (Pinus sylvestris L.) and sessile oak (Quercus petraea (Matt.) Liebl.) growing in monospecific and mixed stands in Bavaria, Germany. Thinning was conducted in autumn 2017, and growth was analyzed over a five-year pre-thinning (2013–2017) and post-thinning (2018–2022) period using replicated thinned and unthinned plots. Thinning substantially enhanced diameter growth in both species. On average, stem increments increased by 26–61

Since the 1970ies, future crop tree release is a prominent silvicultural approach in Central Europe. In a nutshell, it means an early commitment to a relatively small number of trees. Interestingly, the risk of losing, i.e. having to give up such trees later has been hotly discussed, but never scrutinized on a strong empirical basis so far. On 228 long-term experimental plots in Southern Germany, surveys partly dating back to the mid 1970ies, we investigated the risk of losing future crop trees which had been selected and committed to earlier. Focusing on the species Norway spruce, Scots pine, and Douglas fir, we identified inferior growth performance and mechanical damages as the most important loss reasons. Using generalized linear mixed effect models, we could show that the average loss probability for all three species amounted to 5-6% per decade. A detailed analysis revealed several tree-level indicator variables associated with the loss probability. Most of these variables described tree size, social status, or competition, often interacting significantly with the time since a future crop tree had been selected. Distinct, silviculturally relevant patterns were observed among the three species. We discuss the results with regard to implications for forest management, also under changing environmental conditions.

Abstract Tree density can have both positive (facilitation) and negative (competition) impacts on tree growth. Evidence indicates that non‐linear growth–density patterns often better capture the complexity of these interactions than linear forms. How such patterns vary along density and climatic conditions, including drought, remains unclear. In this study we used a unique and long‐term density manipulation experimental network to assess the impact of site climatic conditions and temporal climatic variability on the tree growth–density relationship. For two decades we monitored growth of pedunculate oak ( Quercus robur L.) trees planted with systematic spacing (Nelder design) along a climatic gradient in Europe. We quantified relative growth (normalized to solitary trees) and absolute growth responses to density across age classes. We further disentangled the impact of site aridity and drought periods on growth. Our results revealed that trees growing within intraspecific neighbourhoods can facilitate each other's above‐ground biomass growth, achieving a maximum relative growth increase of 30% compared with solitary trees. This unimodal optimum relationship was most pronounced during early developmental stages (ages 5–10), with its effect diminishing as trees matured. The observed shift of interaction, from facilitation to competition, could be attributed to increased canopy cover in a trees' neighbourhood. We identified aridity‐specific density thresholds (3000–4000 N ha −1 ), beyond which growth declines sharply during water‐limited periods. At dry sites, growth was highly responsive to drought, while the density threshold was relatively consistent across climatic conditions. On mesic to humid sites, drought and density were more tightly coupled, with density effects amplified beyond the initial low‐density optimum. Synthesis. We found that density‐dependent growth follows a unimodal, non‐linear pattern, with optimal density thresholds narrowing under increasing site aridity, stand age and climatic variability. These findings enhance our understanding of how climate and tree development jointly shape growth responses to density. They highlight the need for adaptive density management strategies that account for shifting facilitation–competition dynamics under climate change.

This study examined the effects of mature stands and tree stumps on the height growth of regeneration in a Bavarian experimental site initially established to evaluate the mixing effects of European beech (Fagus sylvatica L.) and Douglas fir (Pseudotsuga menziesii (Mirbel) Franco). Previous research has indicated that mature trees can both facilitate and hinder regeneration, potentially promoting growth due to reduced radiation-induced stress, while also causing competition for resources. Tree stumps were hypothesized to provide benefits through nutrient and water release but might also present risks by harboring pathogens. To investigate these facilitation and competition dynamics, data were collected on mature tree characteristics, regeneration height and diameter (specifically for regeneration trees over 1.3 m), stump diameter, position, and decomposition status. Statistical analyses were performed using linear mixed-effects models to account for the hierarchical structure of the data and spatial autocorrelation, and model selection was guided by Akaike's Information Criterion (AIC) along with likelihood ratio tests. The findings showed that the mature trees negatively impacted overall regeneration height growth, likely due to competition for light. At the same time, tree stumps, particularly those in the early stages of decomposition, positively influenced growth. Significant species differences were observed: Douglas fir height growth was positively influenced by less decomposed stumps, suggesting a benefit from the microhabitats and resources they provide, whereas beech height growth was less dependent on stump variables. These results highlight the complex interplay between competition and facilitation in forest regeneration, emphasizing the importance of considering stump decomposition stages and positions when planting in forests. Future research should expand to different forest types and species to deepen the understanding of these interactions and inform management practices.

Recent studies have shown that the current growth of a tree is strongly influenced by its past. In the face of climate change, assessing the acclimation potential of trees to drought stress is crucial. This study quantifies the influence of past growth on basal area increment during drought for 356 Norway spruces (Picea abies L. Karst), 223 Scots pines (Pinus sylvestris L.), 326 European beeches (Fagus sylvatica L.) and 190 oaks (Quercus petraea (Matt.) Liebl./robur L.) in southern Germany. Dendrochronological data were used to model basal area increment based on growth trends, growth variation, basal area, and the frequency of past low-growth years under average and dry weather conditions based on SPEI. We investigated whether i) progressive or degressive growth trends and ii) high or low growth variations had a positive effect on the drought response of trees, iii) whether this effect varies with basal area and iv) the occurrence of low-growth years influenced drought responses. Trees with progressive growth trends and low variability showed the smallest drought-related growth losses. Depending on growth trends and variability, the drought stress response on basal area growth could be reduced by up to 14 %. The effect of past growth is species-specific and varied with tree size for N. spruce, S. pine and E. beech. Frequent past low-growth years promoted acclimatisation of oaks but increased the drought sensitivity of E. beeches. These results underline the importance of past growth dynamics in determining tree resilience to drought stress and provide valuable insights for silvicultural strategies to enhance forest resilience.

Increasingly uncertain decision outcomes prevail in forest management and hamper choosing a single optimal management alternative. Confronting all management alternatives with multiple future scenarios and selecting an alternative minimising the regret under the worst scenario may provide suitable guidance under such uncertainty. Here, we search for future forested landscape compositions using regret minimisation for different objectives. We consider even-aged and uneven-aged stand types (called close-to-nature stand types) as management alternatives. Close-to-nature forest stand types supported the minimisation of regret for all objectives (represented by financial return, volume increment, C-storage, and two biodiversity indicators). However, closeto-nature stand types covered 18 % to 43 % of the future forest landscape in our study, which shows that evenaged stands are also necessary. For example, supporting biodiversity or multiple objectives simultaneously required large proportions of light-demanding and climate-change-tolerant Oak stands (even aged). Such Oak stands are difficult to achieve under shady conditions with limited canopy openings, which is typical for unevenaged systems. Building on robust Pareto frontiers, we show a substantial trade-off between supporting biodiversity and maximising financial return but only a moderate trade-off between supporting biodiversity and maximising the C storage in a forest landscape. We suggest that such landscape-level trade-offs be quantified and discussed more intensively.

Tree mortality is a crucial process in forest dynamics and a key component of forest growth models and simulators. Factors like competition, drought, and pathogens drive tree mortality, but the underlying mechanism is challenging to model. The current environmental changes are even complicating model approaches as they influence and alter all the factors involving mortality. However, innovative classification algorithms can go deep into data to find patterns that can model or even explain their relationship. We use Logistic binomial Regression as the reference algorithm for predicting individual tree mortality. However, different machine learning (ML) alternatives already applied to other forest modeling topics can be used for this purpose. Here, we compare the performance of five different ML algorithms (Decision Trees, Random Forest, Naive Bayes, K-Nearest Neighbour, and Support Vector Machine) against Logistic binomial Regression in individual tree mortality classification under 40 different case studies and a cross-validation case study. The data used corresponds to Norway spruce long-term experimental plots, which have a total of 75,522 tree records and a 10.28 % mortality rate on average. Through different case studies, when more variables were used, general performance improved as expected, while more extensive datasets decreased the performance level of the algorithms. Performance was also higher when plots remained without management compared to thinned ones. Random Forest outperformed the other algorithms in all the cases except cross-validation, where it was the weaker one. Our results demonstrate the potential of ML in assessing tree mortality. When the model application is not clearly defined and/or model interpretability is needed, Logistic binomial Regression is still the best tool for evaluating individual tree mortality.

Urban trees provide vital benefits such as carbon storage and urban cooling. However, limited knowledge on urban tree mortality hinders effective management of sustainable urban tree stock. Utilizing cadaster datasets, this study investigates urban tree mortality across 959,466 trees in four German cities from 2016 to 2022 and identified key factors affecting tree mortality. While assessing annual mortality rates, we found an average annual mortality rate of 1.3 %, with a peak in 2019. Street trees exhibited a 0.3 % higher mortality rate than nonstreet trees, underscoring their vulnerability to urban conditions. Additionally, annual mortality rates were higher for coniferous species (2.2 %) compared to broad-leaved species (1.3 %). Using multivariable logistic regression approach, significant correlations were found between mortality probability, drought tolerance indices, and local climate determined by de Martonne Index (dMI). Our findings showed that younger and drought-intolerant street trees are likely more vulnerable to urban environmental stressors than matured, drought-tolerant ones. The study successfully assessed cadaster data, demonstrating their utility for urban tree mortality monitoring but highlighted challenges such as difficulty in accessing data, inconsistent data collection intervals, various data formats and less frequent monitoring of non-street trees. This research underscores the need for long-term monitoring and highlights the importance of species-specific and environmental factors in urban forestry management.

We examined the effects of drought-induced stress on foliar litter production, nutrient contents, and nutrient masses in mature European beech (Fagus sylvatica [L.]) and Norway spruce (Picea abies [L.] Karst.) over a two-year period (October 2015–September 2017) in southern Bavaria. A rainfall exclusion experiment was conducted with six control plots receiving normal rainfall and six roof plots excluding rainfall. Abscised leaf and needle biomass, as well as the contents and masses of calcium (Ca), potassium (K), magnesium (Mg), nitrogen (N), and phosphorus (P), were monitored across two non-vegetation periods (October–February: NV1, NV2) and two vegetation periods (March–September: V1, V2).Foliar litter on control plots (set at 100 %) was 4000, 329, 4501, 403 kg/ha for European beech and 3534, 1146, 1352, 607 kg/ha for Norway spruce across the four observation periods (NV1, V1, NV2, V2). Roof plots yielded 2917 (73 %), 364 (111 %), 3710 (82 %), 358 (89 %) kg/ha for European beech and 5841 (165 %), 1040 (91 %), 899 (67 %), 447 (74 %) kg/ha for Norway spruce. Significant differences between control and roof plots were observed only during NV1.Foliar K contents were significantly lower under drought in both species. For European beech, values were 2.83, 3.83, 2.76, 4.37 g/kg (control plots), compared to 2.38 (84 %), 3.08 (80 %), 2.30 (83 %), 4.01 (92 %) g/kg (roof plots). For Norway spruce, values were 2.64, 2.77, 2.51, 2.13 g/kg (control plots), compared to 2.26 (86 %), 2.33 (84 %), 2.01 (80 %), 1.66 (78 %) g/kg (roof plots). Drought also significantly decreased foliar Ca content in Norway spruce during NV2, from 8.61 to 7.04 g/kg (82 %).Foliar nutrient masses aligned more closely with biomass abscission patterns than with nutrient translocation patterns. European beech predominantly exhibited significantly reduced abscised nutrient masses under drought during NV1 and NV2, while Norway spruce initially showed significantly increased abscised nutrient masses in NV1, followed by a marked decline in subsequent seasons.We concluded that European beech responded to drought stress by reducing foliage biomass production, suggesting a potential acclimation strategy, whereas Norway spruce mitigated water loss through transpiration by shedding its needles. However, Norway spruce failed to compensate for the initial high needle losses by regenerating sufficient new needles, indicating its lower resilience to drought.

When dealing with CO2 in- and outflows of forest ecosystems, the CO2 emissions due to harvest operations have been mostly not included into the analyses, so far. Therefore, to demonstrate an integrative and more holistic view, we used the generic simulation model care4cmodel for assessing relevant silvicultural concepts for pine species in different European (Pinus sylvestris, Pinus pinaster) and South African (Pinus pinaster, Pinus radiata) climatic conditions and management regimes. The concepts covered different thinning and regeneration systems, ranging from clear cut to target diameter harvest. Our focus was on the CO2 emissions due to forest operations (CEF) and their relation to the CO2 uptake due to wood increment (CUI). Simulations covered a time span of 50 years and different scenarios of the initial shares of stand development phases on a large virtual forest area. Our simulations suggest that, across all concepts and countries, the CEF are about 2–3 orders of magnitude smaller than the CUI. More importantly, while the initial situation strongly matters for development of the increment and harvest amounts, it does considerably less so for CEF and CEF/CUI. This provides leeway for silvicultural decisions regarding the above-mentioned CO2 flows.

Climate change and associated drought stress significantly impact tree root systems and canopies, essential for water and nutrient provision. We examined the root-shoot allometry of dominant European beech (Fagus sylvatica [L.]) and Norway spruce (Picea abies [L.] Karst.) in monospecific and mixed-species stands across five triplets along a climatic and geological gradient from North to South Bavaria, focusing on their drought responses. We used Lloret indices of resistance, resilience, and recovery to assess the effects of the exceptionally hot drought year of 2003 on root and shoot growth. Data were categorized by stand type: beech pure (BP), beech mixed (BM), spruce pure (SP), and spruce mixed (SM). Results showed that the allometric factor decreased in the order SP (− 2.986) > SM (− 3.960) > BP (− 7.434) > BM (− 8.628), and the allometric exponent in the order BM (3.159) > BP (2.795) > SM (1.914) > SP (1.657), with significant differences between species. For European beech, the allometric factor was significantly affected by stand density and site index, and the allometric exponent by tree species, mixed stand type, and site index. European beech exhibited higher resistance and resilience in shoot growth compared to Norway spruce, with no significant influence from stand type. Root growth patterns were similar, with European beech being more resilient and resistant than Norway spruce, although to a lesser extent. European beech also recovered faster than Norway spruce, with significant differences between their respective monospecific stands. We concluded that root growth for both species was enhanced compared to shoot growth, improving anchoring and the supply of water and nutrients to the aboveground tree organs. European beech experienced less drought stress than Norway spruce, attributed to differences in their physiology and root system characteristics.

Background: In Central Europe, forests are increasingly affected by various disturbances, resulting in an increasing gap formation in the canopy. In order to support goal-oriented management, more knowledge is required about the acclimation of the crown and its effects on the basal area growth of trees at the edge of a gap. Methods: This work compared trees' growth and crown structure at the edge of a transient gap, with a gap size of more than 80 m2, with trees in the stand that were at least 30 m away from the gap. A total of 249 European beeches (Fagus sylvatica L.), Norway spruces (Picea abies L. Karst), Scots pines (Pinus sylvestris L.), oaks (Quercus spp.; Quercus petraea (Matt.) Liebl., Quercus robur L.), and silver firs (Abies alba Mill.) were examined on long-term experimental plots in southern Germany. Various crown measures were developed and calculated using highresolution terrestrial laser scanning (TLiDAR) to capture the three-dimensional crown structures. Growth responses to edge conditions were measured based on tree rings. Using linear mixed models, we predict the basal area increment of edge trees relative to trees in the stand under wet and dry soil moisture conditions after the gap formation. Results: We identified i) species-specific acclimation of the crown of edge trees after the gap formation, ii) under wet soil moisture conditions a growth increase of 25%-45% for beech, pine, and oak edge trees and growth losses of 5%-60% for spruce and fir and iii) coniferous tree species benefited from the edge position regarding their basal area increment under dry soil moisture conditions and deciduous tree species grew regardless of the soil moisture conditions at the edge of a gap. Conclusion: Gaps have a species-specific effect on the habitus and growth of edge trees and can have both positive and negative impacts on silviculture.

The effects of site and climate on the self-thinning line, a key characteristic that defines forest dynamics, have been the subject of research for decades. However, contrasting results have generally been found. To adapt management practices for widely distributed species, especially considering the impact of climate change, it is crucial to understand the variables and their effects on the self-thinning line. We conducted a systematic analysis based on 77 trial plots from 62 long-term experiments across Europe, covering the distribution range of Scots pine. Our focus was on unthinned conditions since 1975. Using a linear mixed model, we examined the effects of each statistically significant variable, separating the influences on the slope and the intercept. Our observations revealed that parameters enhancing species tolerance, such as shortwave solar radiation, flatten the slope of the self-thinning line. Conversely, temperature and precipitation, which reduce self-tolerance and increase intraspecific competition, lead to an increase in the slope. Balancing the effects between these aspects results in a maximum negative slope at mid-latitudes. In terms of the intercept, we found compensating effects among the analyzed factors, indicating a monotonic increase with decreasing latitude and increasing radiation. Although there were no significant changes in the self-thinning line since the 1990 s, we observed an increase in mortality, suggesting an accelerated self-thinning process. Site and climatic differences across the distribution range of Scots pine influenced the self-thinning line, whereas no trends with time could be observed. Therefore, management strategies and models based on self-thinning need to be adapted to different latitudes. While climate changes have not yet impacted the trajectory significantly, a continuous rise in temperature, coupled with high precipitation, may accelerate self-thinning and result in increased biomass accumulation.

Robust and accurate point cloud registration is an essential part of many robotic tasks such as SLAM or object pose retrieval. In this paper, we address the problem of global 3D point cloud registration, i.e., the task of estimating the 3D rigid body transform between a source and a target point cloud without any initial guess. Typically, the problem is solved by extracting and matching features to find a data association and then computing a transform that minimizes the squared distance between points. Our approach combines the normal distributions transform and oriented point pair framework and introduces the NDT distance histogram to quickly generate and test candidate transforms. Our method further exploits semantic information if available for greater speed. We implement our algorithm in C++ and compare it to other state-of-the-art approaches on a diverse set of environments. Our evaluation shows that our method outperforms the other approaches, especially concerning run-time and compute efficiency.

Carbon stocks and flows in forest ecosystems play an important role in the context of climate change mitigation. Different aspects of the forest carbon balance, however, are often treated independently, leading to a fragmented, disciplinary knowledge. With the R-package care4cmodel we want to support a consolidated view of forest growth and forest operations with respect to carbon flows. The software is, in essence, a pragmatic simulation tool that allows juxtaposing the CO2 uptake due to wood increment and the CO2 emissions due to forest operations for given silvicultural concepts in an arbitrary area, over time. At the core, the approach is a dynamic forest area model where forest development stages are represented in a cyclical sequence, which can be broken by disturbances. The model scales up growth and yield information given per forest development stage and unit area to the dynamically simulated development stage areas. This allows to quantify the total forest area’s CO2 uptake, and to estimate the CO2 emissions caused by forest operations. The forest operations in our implementation include the maintenance of the forest road network, felling trees, debranching and bucking the stems, and extracting the timber to a landing at a forest road. The transport from there to the industry is beyond the system boundary. For the CO2 uptake of the forest system, the current model version focuses on the wood increment only. We use a practical example to demonstrate the basic features of the model and its plausible behaviour. Beyond the current focus of the model, we see a broad field of applications as a generic meta model, especially in the context of ecosystem service provision.

Against the backdrop of global change, the intensity, duration, and frequency of droughts are projected to increase and threaten forest ecosystems worldwide. Tree responses to drought are complex and likely to vary among species, drought characteristics, and site conditions. Here, we examined the drought response patterns of three major temper-ate tree species, s. fir (Abies alba), E. beech (Fagus sylvatica), and N. spruce (Picea abies), along an ecological gradient in the South - Central - East part of Germany that included a total of 37 sites with varying climatic and soil conditions. We relied on annual tree-ring data to assess the influence of different drought characteristics and (micro-) site conditions on components of tree resilience and to detect associated temporal changes. Our study revealed that nutrient regime, drought frequency, and hydraulic conditions in the previous and subsequent years were the main determinants of drought responses, with pronounced differences among species. Specifically, we found that (a) higher drought fre-quency was associated with higher resistance and resilience for N. spruce and E. beech; (b) more favorable climatic conditions in the two preceding and following years increased drought resilience and determined recovery potential of E. beech after extreme drought; (c) a site's nutrient regime, rather than micro-site differences in water availability, determined drought responses, with trees growing on sites with a balanced nutrient regime having a higher capacity to withstand extreme drought stress; (d) E. beech and N. spruce experienced a long-term decline in resilience. Our results indicate that trees under extreme drought stress benefit from a balanced nutrient supply and highlight the relevance of water availability immediately after droughts. Observed long-term trends confirm that N. spruce is suffering from per-sistent climatic changes, while s. fir is coping better. These findings might be especially relevant for monitoring, sce-nario analyses, and forest ecosystem management.