Additive Trees Research Articles

SMARTboost Learning for Tabular Data

We introduce SMARTboost (boosting of symmetric smooth additive regression trees), an extension of gradient boosting machines with improved accuracy when the underlying function is smooth or the sample small or noisy. In extensive simulations, we find that the combination of smooth symmetric trees and of carefully designed priors gives SMARTboost a large edge (in comparison with XGBoost and BART) on data generated by the most common parametric models in econometrics, and on a variety of other smooth functions. XGBoost outperforms SMARTboost only when the sample is large, and the underlying function is highly discontinuous. SMARTboost’s performance is illustrated in two applications to global equity returns and realized volatility prediction.

Development of a recommendation engine for university study programme selection: A regression-based approach

This paper is the third in a series focused on bridging the gap between secondary and higher education. Our primary objective is to develop a robust theoretical framework for an innovative e-business model called the Undergraduate Study Programme Search System (USPSS). This system considers multiple criteria to reduce the likelihood of exam failure or the need for multiple retakes, while maximizing the chances of successful program completion. Testing of the proposed algorithm demonstrated that the Stochastic Gradient Boosted Regression Trees method outperforms the current method used in Lithuania for admitting applicants to 47 educational programs. Specifically, it is more accurate than the Probabilistic Neural Network for 25 programs, the Ensemble of Regression Trees for 24 programs, the Single Regression Tree for 18 programs, the Random Forest Regression for 16 programs, the Bayesian Additive Regression Trees for 13 programs, and the Regression by Discretization for 10 programs.

Bayesian additive regression trees for predicting childhood asthma in the CHILD cohort study

BackgroundAsthma is a heterogeneous disease that affects millions of children and adults. There is a lack of objective gold standard diagnosis that spans the ages; instead, diagnoses are made by clinician assessment based on a cluster of signs, symptoms and objective tests dependent on age. Yet, there is a clear morbidity associated with chronic asthma symptoms. Machine learning has become a popular tool to improve asthma diagnosis and classification. There is a paucity of literature on the use of Bayesian machine learning algorithms to predict asthma diagnosis in children. This paper develops a prediction model using the Bayesian additive regression trees (BART) and compares its performance to various machine learning algorithms in predicting the diagnosis of childhood asthma.MethodsClinically relevant variables collected at or before 3 years of age from 2794 participants in the CHILD Cohort Study were used to predict physician-diagnosed asthma at age 5. BART and six other commonly used machine learning algorithms, namely adaptive boosting, logistic regression, decision tree, neural network, random forest, and support vector machine were trained. Measures of performance including sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve were calculated. The confidence intervals were calculated using Bootstrapping samples. Important predictors and interaction effects associated with asthma were also identified using BART.ResultsBART, logistic regression and random forest showed the highest area under the ROC curve compared to other machine learning algorithms. Based on BART, recurrent wheeze, respiratory infection and food sensitization at 3 years of age were the most important predictors. The three most important interaction effects were found to be interaction terms of respiratory infection at 3 years and recurrent wheezing at 3 years, maternal asthma and paternal asthma, and maternal wheezing and inhalant sensitization of child at 3 years.ConclusionsBART demonstrated promising prediction performance when compared to other machine learning algorithms. Future research could validate the BART in an external cohort to evaluate its reliability and generalizability.

Modeling Crash Risk on Roadway Networks using Bayesian Regression Trees

Statistical modeling of vehicle crashes leads to a better understanding of how and why such crashes occur. Due to the irregular network structure of roadways, analyses are typically confined to a single roadway rather than considering the entire network collectively. Here, we present methodology to model crash risk of vehicle crashes on irregular roadway networks and estimate how that risk varies with road characteristics. We model vehicle crashes observed on a road network as a Poisson point pattern with a piecewise linear intensity surface. Further, we combine Bayesian additive regression trees (BART) and spatial data analysis to accurately explain the intensity surface allowing inference on the effect of road characteristics on crash risk. We illustrate the methodology using a dataset of vehicle crashes on Interstate highways in Utah.

Heterogeneous Mediation Analysis for Cox Proportional Hazards Model With Multiple Mediators.

This study proposes a heterogeneous mediation analysis for survival data that accommodates multiple mediators and sparsity of the predictors. We introduce a joint modeling approach that links the mediation regression and proportional hazards models through Bayesian additive regression trees with shared typologies. The shared tree component is motivated by the fact that confounders and effect modifiers on the causal pathways linked by different mediators often overlap. A sparsity-inducing prior is incorporated to capture the most relevant confounders and effect modifiers on different causal pathways. The individual-specific interventional direct and indirect effects are derived on the scale of the logarithm of hazards and survival function. A Bayesian approach with an efficient Markov chain Monte Carlo algorithm is developed to estimate the conditional interventional effects through the Monte Carlo implementation of the mediation formula. Simulation studies are conducted to verify the empirical performance of the proposed method. An application to the ACTG175 study further demonstrates the method's utility in causal discovery and heterogeneity quantification.

Application of artificial intelligence to study the causality in public health: a systematic review

Abstract Introduction In epidemiology, establishing causation requires meeting rigorous criteria beyond mere association. Traditional methodologies often struggle to provide definitive causal conclusions due to confounding variables. However, leveraging artificial Intelligence (AI) and machine Learning (ML) offers a promising solution by modeling complex interactions among relevant factors. This systematic review explores the AI techniques utilized to uncover causal relationships in public health applications. Methods A literature search was conducted in Pubmed, Web of Science, and Scopus, employing the following search strategy: (Causalit*[Title/abstract] OR Causation*[Title/abstract] OR Causal[Title/abstract]) AND (‘Machine Learning’[Title/abstract] OR “Deep Learning” [Title/abstract] OR ‘Artificial Intelligence’ [Title/abstract] OR Algorithm* [Title/abstract]). The search aimed to identify studies encompassing causal inference techniques utilizing ML and AI methodologies in public health. Results From a total of 28,230 articles published up to August 2023, after removing duplicates (11,158) and articles that did not meet the inclusion criteria (16,168 post title/abstract screening, and 856 post full-text screening), the systematic review included 48 articles. Bayesian additive regression trees, Bayesian network models, causal forests analysis, and graphical causal models emerged as the prevailing causal methodologies. AI-driven causal methodologies have been investigated across various domains within public health, encompassing disease prediction, evaluation of treatment efficacy, identification of risk factors, analysis of health behaviors, precision Public Health initiatives, and environmental health assessment. Discussion While our systematic review affirms the potential of AI-driven causal approaches across several public health domains, further research is warranted before these emerging techniques can be effectively integrated into real settings. Key messages • In epidemiology, establishing causation requires meeting rigorous criteria beyond mere association and traditional methodologies often struggle to provide definitive causal conclusions. • This systematic review explores the AI techniques utilized to uncover causal relationships in public health applications.

Cultivar and tree size, but not climate, are principal factors affecting stem quality of Populus tomentosa plantations in the North China Plain

In order to explore the management strategies for cultivating and improving the stem quality of Populus tomentosa plantations under the background of climate change, this study focuses on P. tomentosa plantations over 10 years old in the North China Plain. Using linear mixed models and ordered logistic models, the impacts of cultivar, tree size, stand age, competition, and climate on the stem quality of P. tomentosa (including crown base height, tapering, branching grade, and straightness grade) were analyzed. The study found that: cultivar significantly affected all stem quality indicators (P < 0.05). Compared to other cultivars, the P. tomentosa f. yixianensis had a 23 % increase in branch height, an 8 % reduction in taper, and the risk of having poorer branches and stem form decreased by 96 % and 80 %, respectively. In addition, taller and bigger-diameter trees had better external stem quality. The impacts of competition-related indicators on stem quality were inconsistent: reduced canopy openness could improve stem quality by enhancing light competition, however, increased tree density increased the risk of deteriorating branching and straightness grade by 1.2 % and 0.9 %, respectively. Among all factors, cultivar and individual tree size had the greatest relative importance for various stem quality indicators, followed by competition-related factors, while stand age and climate factors have no significant impact on P. tomentosa stem quality (P > 0.05). Currently, climate change has little impact on the external stem quality characteristics during the cultivation of P. tomentosa plantations. Management strategies for stem quality can focus on cultivar selection and competition regulation. It is worth noting that increasing tree density in the North China Plain may not necessarily improve stem quality of P. tomentosa plantations, so caution is needed in the process of regulating competition.

Considerations for using tree-based machine learning to assess causation between demographic and environmental risk factors and health outcomes.

Evaluation of the heterogeneous treatment effect (HTE) allows for the assessment of the causal effect of a therapy or intervention while considering heterogeneity in individual factors within a population. Machine learning (ML) methods have previously been employed for HTE evaluation, addressing the limitations associated with modelling complex systems. In this work, three tree-based ML algorithms, causal random forest (CRF), causal Bayesian additive regression trees (CBART), and causal rule ensemble (CRE), are used to analyze the potential causation of benzene exposure to cause childhood acute myeloid leukemia (AML). Data for this analysis is generated by drawing samples from a previously developed model that estimates AML probability given as input demographic information and benzene exposure. Comparison is drawn between the three tree-based algorithms in terms of the predicted average treatment effect (ATE), the regression coefficient of determination, and the computational time of each algorithm. Minimal difference is reported between the three tree-based algorithms in terms of the ATE, as well as the regression coefficient of determination. However, CRF outperforms CBART in terms of algorithm computational time. Moreover, CRF allows for both continuous and binary treatment variables, as opposed to CBART and CRE, making it better suited to environmental health studies, where exposure levels of pollutants shall be considered continuous. Following the comparison of all three algorithms, the influence of adding Gaussian noise to the treatment and outcome variables, as well as outliers, is investigated using CRF. A set of considerations is drawn to guide researchers in using these algorithms. These considerations detail the simulation settings, applications, and results interpretation and aim to provide prompt information in decision-making surrounding the establishment of pollutant exposure thresholds in environmental risk assessments.

Novel machine learning technique further clarifies unrelated donor selection to optimize transplantation outcomes

AbstractWe investigated the impact of donor characteristics on outcomes in allogeneic hematopoietic cell transplantation (HCT) recipients using a novel machine learning approach, the Nonparametric Failure Time Bayesian Additive Regression Trees (NFT BART). NFT BART models were trained on data from 10 016 patients who underwent a first HLA-A, B, C, and DRB1 matched unrelated donor HCT between 2016 and 2019, reported to the Center for International Blood and Marrow Transplant Research, then validated on an independent cohort of 1802 patients. The NFT BART models were adjusted based on recipient, disease, and transplant variables. We defined a clinically meaningful impact on overall survival (OS) or event-free survival (EFS; survival without relapse, graft failure, or moderate to severe chronic graft-versus-host disease) as >1% difference in predicted outcome at 3 years. Characteristics with <1% impact (within a zone of indifference) were not considered to be clinically relevant. Donor cytomegalovirus, parity, HLA-DQB1, and HLA-DPB1 T-cell epitope matching fell within the zone of indifference. The only significant donor factor that associated with OS was age, in which, compared with 18-year-old donors, donors aged ≥31 years old were associated with lower OS. Both donor age (≤32 years) and use of a male donor, regardless of recipient sex, improved EFS. We, therefore, recommend selecting the earliest available donor within the 18 to 30 years age range for HCT to optimize OS. If several donors in the 18 to 30 years age range are available, a male donor may be chosen to optimize EFS.

Effects of interval treadmill training on spatiotemporal parameters in children with cerebral palsy: A machine learning approach

Quantifying individualized rehabilitation responses and optimizing therapy for each person is challenging. For interventions like treadmill training, there are multiple parameters, such as speed or incline, that can be adjusted throughout sessions. This study evaluates if causal modeling and Bayesian Additive Regression Trees (BART) can be used to accurately track the direct effects of treadmill training on gait. We developed a Directed Acyclic Graph (DAG) to specify the assumed relationship between training input parameters and spatiotemporal outcomes during Short Burst Locomotor Treadmill Training (SBLTT), a therapy designed specifically for children with cerebral palsy (CP). We evaluated outcomes after 24 sessions of SBLTT for simulated datasets of 150 virtual participants and experimental data from four children with CP, ages 4–13 years old. Individual BART models were created from treadmill data of each step. Simulated datasets demonstrated that BART could accurately identify specified responses to training, including strong correlations for step length progression (R2 = 0.73) and plateaus (R2 = 0.87). Model fit was stronger for participants with less step-to-step variability but did not impact model accuracy. For experimental data, participants’ step lengths increased by 26 ± 13 % after 24 sessions. Using BART to control for speed or incline, we found that step length increased for three participants (direct effect: 13.5 ± 4.5 %), while one participant decreased step length (−11.6 %). SBLTT had minimal effects on step length asymmetry and step width. Tools such as BART can leverage step-by-step data collected during training for researchers and clinicians to monitor progression, optimize rehabilitation protocols, and inform the causal mechanisms driving individual responses.

Identifying suitable areas for plenter forest management

Plenter forests, also known as uneven-aged or continuous cover forests enhance forest resilience and resistance against disturbances compared to even-aged forests. They are considered as an adaptation option to mitigate climate change effects. In this study, we present a conceptual approach to determine the potentially suitable area for plenter forest management within central European mixed species forests and apply our approach to the case study area in Styria, the south-eastern Province of Austria. The concept is based on ecological and technical-economic constraints and considers expected future climate conditions and its impact on plenter forest management. For each 1 ha forest pixel, we assess the ecological conditions for plenter forest management according to the autecological growth conditions of silver fir, and at least one additional shade tolerant tree species. The technical-economic constraints are defined by slope (≤ 30%) and distance to the next forest road (≤ 100 m) to ensure cost-efficient harvesting. The results show that under current climate conditions 28.4% or 308,349 ha of the forests in Styria are potentially suitable for plenter forest management. For the years 2071 to 2100 and under the climate change scenario RCP 4.5, the potential area decreases to 286,098 ha (26.3% of the total forest area) and for the scenario RCP 8.5 to 208,421 ha (19.1% of the total forest area). The main reason for these changes is the unfavourable growing conditions for silver fir in the lowlands, while in the higher elevations silver fir is likely to expand. Our results may serve forest managers to identify areas suitable for plenter forests and assist in the transformation of even-aged pure forests to uneven-aged forests to increase resistance, resilience, and biodiversity under climate change.

Longitudinal Impacts of High Versus Low Greenness on Cardiovascular Disease Conditions.

Greenness-or vegetative presence-has been identified as a factor in chronic disease. The present study examines the longitudinal relationship between objective measures of greenness at the residential block level and incidence of 6 cardiovascular disease conditions. Analyses examined the impact of consistently high versus consistently low "precision" greenness at the Census block level on the 5-year incidence of cardiovascular disease conditions, including acute myocardial infarction, atrial fibrillation, heart failure, ischemic heart disease, stroke/transient ischemic attack, and hypertension, among 229 034 US Medicare beneficiaries in Miami-Dade County, Florida, USA. Zero-inflated Poisson regression was used to model the odds of developing any new cardiovascular disease and number of new cardiovascular disease conditions based on greenness tertiles computed across 2011 and 2016 Normalized Difference Vegetation Index values, adjusting for individual age, sex, race, ethnicity, baseline cardiovascular disease conditions, neighborhood income, and walkability in 2011 and 2016. When compared with individuals consistently in the low greenness tertile in 2011 and 2016, those consistently in the high greenness tertile in 2011 and 2016 had a 9% lower odds of having any new cardiovascular conditions (odds ratio [OR], 0.91 [95% CI, 0.84-0.99]; P=0.021). Over a 5-year period, consistently high greenness, when compared with consistently low greenness, was associated with lower odds of any new cardiovascular disease conditions. Identifying the role of greenness exposure in such a small geographic area, the Census block on which the older adult resides, allows for more precise, strategic decisions on where additional trees can be added-by selecting at-risk blocks rather than entire neighborhoods for tree-planting interventions.

Machine learning-based prediction of tree crown development in competitive urban environments

In urban forestry, managing trees is crucial for sustainable urban environments, especially in the context of climate change and the urban heat island effect. This research explores the complex dynamics of tree crown geometry development by asking the question: how do surrounding objects, such as nearby trees, buildings, and other urban structures, affect the shape of tree crowns? It aims to uncover how competition for light and space influences tree crown development in competitive urban environments. Our study employs machine learning models on six main species in Munich, using the measured data from the LiDAR scans, with the Hist Gradient Boosting Regressor (HGBR) emerging as the most promising performer across various metrics. Notably, the evaluation of 13 models reveals the HGBR’s consistent ranking as the best or second-best across all tree crown dimensions assessed, with R2 values reaching 0.83 for the tree height model and 0.7 on average for crown radiuses in eight directions. Employing SHapley Additive exPlanations (SHAP) values elucidate factors influencing model predictions, emphasising the significant impact of adjacent trees and buildings. After evaluating the models to include additional tree species in Munich, the models show strong predictive capabilities for some additional species. Despite the studies’ limitations - the models are only valid for selected species, and there are constraints in predicting tree crown start height - our findings contribute valuable insights for urban forestry management and planning.

Does living close to allergenic street trees affect lung function in German adults?

IntroductionStudies on greenspace and lung function in adults produced divergent results. Some of the adverse findings could be due to long-term exposure to allergenic tree pollen. We investigated whether having more birch trees or more allergenic trees around home is related to worse lung function and whether these exposures confound the association between greenspace and lung function. MethodsThe analytic sample consisted of 874 adults aged 20–44 years at baseline from the German study centers, Erfurt and Hamburg, of the ECRHS cohort study. Spirometric lung function was measured in 1991/92, 2000/01, and 2011/12. We counted trees based on tree registries and classified them into allergenic and non-allergenic. We assessed exposure to greenspace with the normalized difference vegetation index (NDVI), tree cover density, and total number of trees in a 300 m buffer around home. Linear mixed models were used. ResultsThe forced expiratory volume in 1 s (FEV1) and the forced vital capacity (FVC) were decreased in the presence of more birch trees after adjusting for confounders and co-exposures. For every 10 additional birch trees in a 300 m buffer around home, the average change in FEV1 was −27.6 mL (95% confidence interval (CI): [-58.7, 3.5]). For FVC the average change was −28.2 mL (95% CI: [-62.0, 5.6]). No consistent associations were found for allergenic trees, total trees, tree cover density, or NDVI. Unlike other associations, those of birch trees with FEV1 and FVC were not moderated by allergic sensitization to birch pollen, history of asthma symptoms or nasal allergies including hay fever, ozone, NO2, or age. DiscussionLiving close to birch trees had an adverse long-term association with lung function. That tree registries were limited to street trees prevented us from answering the question of a potential confounding of greenspace effects by allergenic neighborhood trees.

Enhanced intrusion detection model based on principal component analysis and variable ensemble machine learning algorithm

The intrusion detection system (IDS) model, which can identify the presence of intruders in the network and take some predefined action for safe data transit across the network, is advantageous in achieving security in both simple and advanced network systems. Several IDS models have various security problems, such as low detection accuracy and high false alarms, which can be caused by the network traffic dataset's excessive dimensionality and class imbalance in the creation of IDS models. Principal Component Analysis (PCA) has proven to be a helpful feature selection technique for dimensionality reduction. As a result, because it is a linear transformation, it has challenges capturing non-linear relationships between feature properties in the network traffic datasets. This paper proposes a variable ensemble machine learning method to solve the problem and achieve a low variance model with high accuracy and low false alarm. First, PCA is combined with the AdaBoost ensemble machine learning algorithm, which acts as stagewise additive modelling to compensate for PCA's deficiency in feature selection in network traffic by minimizing the exponential loss function. Secondly, PCA is used for feature selection, and a LogitBoost classifier algorithm can be used for multiclass classification and acts as an additive tree regression to compensate for the PCA's weakness by minimizing the Logistic Loss to provide an optimal classifier output. Finally, the low variance ability of RandomForest, which employs the bagging approach, is applied to eliminate overfittings. The experiments of the IDS model developed from the proposed methods were evaluated on the WSN-DS, NSL-KDD, and UNSW-N15 datasets. The performance of the methods, PCA with AdaBoost, on the WSN-DS dataset has an accuracy score of 92.3 %, an 89.0 % accuracy score on the NSL-KDD dataset, and a 67.9 % accuracy score on UNSW-N15, which is the least accurate score. PCA and RandomForest surpassed them by scoring 100 % accuracy on all three datasets. PCA and Bagging have an accuracy score of 99.8 % on the WSN-DS dataset, 100 % on the NSL-KDD dataset, and 93.4 % on the UNSW-N15 dataset. In comparison, PCA and LogitBoost have an accuracy score of 98.9 % on the WSN-DS dataset, 100 % on the NSL-KDD dataset, and 88.7 % on the UNSW-N15 dataset.

How to estimate causal effects associated with family planning? An introduction to Prince BART, a new approach to effect estimation based on principal stratification and Bayesian non-parametric models

There is interest in learning about the causal effect of family planning (FP) on empowerment-related outcomes. Experimental data related to this question are available from trials in which FP programs increase access to FP. However, estimating the effect of interest is challenging because FP and empowerment-related variables are likely to be confounded. We propose a new approach, Prince BART, to estimate the causal effect of FP on other outcomes of interest, among women affected by a FP program. Prince BART combines principal stratification, to account for differences in FP behavior based on exposure to a FP program, with Bayesian Additive Regression Trees (BART), to non-parametrically model stratum membership and relationships between variables. Prince Bart allows for estimation of effect heterogeneity. We use Prince BART to assess the causal effect of modern contraceptive use on employment in six cities in Nigeria, based on quasi-experimental data from a FP program trial during the first half of the 2010s. In the sample, we find a positive overall effect of contraceptive use on employment and substantial differences in effect sizes across subgroups defined by baseline covariates such as marital status, work at baseline, and wealth or education. The modeling approach and results can be used to inform future analyses seeking to understand the causal relationship between family planning use and other outcomes, e.g., related to empowerment.

Disaggregating census data for population mapping using a Bayesian Additive Regression Tree model

Population data is crucial for policy decisions, but fine-scale population numbers are often lacking due to the challenge of sharing sensitive data. Different approaches, such as the use of the Random Forest (RF) model, have been used to disaggregate census data from higher administrative units to small area scales. A major limitation of the RF model is its inability to quantify the uncertainties associated with the predicted populations, which can be important for policy decisions. In this study, we applied a Bayesian Additive Regression Tree (BART) model for population disaggregation and compared the result with a RF model using both simulated data and the 2021 census data for Ghana. The BART model consistently outperforms the RF model in out-of-sample predictions for all metrics, such as bias, mean squared error (MSE), and root mean squared error (RMSE). The BART model also addresses the limitations of the RF model by providing uncertainty estimates around the predicted population, which is often lacking with the RF model. Overall, the study demonstrates the superiority of the BART model over the RF model in disaggregating population data and highlights its potential for gridded population estimates.

Mediators of racial and ethnic inequities in clinical trial participation among patients with cancer, 2011-2023.

Although racially and ethnically minoritized populations are less likely to participate in cancer trials, it is unknown whether social determinants of health (SDOH) explain these inequities. Here we identify SDOH factors that contribute to racial and ethnic inequities in clinical trial participation among patients with 22 common cancers. This retrospective cohort study used electronic health record data (2011-2023) linked to neighborhood (US Census tract) data from multiple sources. Patients were followed from diagnosis to clinical study drug receipt (proxy for trial participation), death, or last recorded activity. Associations were assessed using Cox proportional hazards models adjusted for clinical factors (year of diagnosis, age, sex, performance status, disease stage, cancer type). To elucidate which area-level SDOH underlie racial and ethnic inequities, mediation analysis was performed using nonlinear multiple additive regression tree models. This study included 250 105 patients (64.7% non-Latinx White, 8.9% non-Latinx Black, 5.2% Latinx). Black and Latinx patients were more likely to live in economically or socially marginalized areas (eg, disproportionately minoritized [measure of segregation], limited English proficiency, low vehicle ownership) than White patients. Black (3.7%; hazard ratio = 0.55, 95% confidence interval [CI] = 0.52 to 0.60) and Latinx patients (4.4%; hazard ratio = 0.63, 95% CI = 0.58 to 0.69) were less likely to participate in trials than White patients (6.3%). Fewer patients in economically or socially marginalized neighborhoods participated in trials. Mediators explained 62.2% (95% CI = 49.5% to 74.8%) of participation inequities between Black and White patients; area-level SDOH-including segregation (29.9%, 95% CI = 21.2% to 38.6%) and vehicle ownership (11.6%, 95% CI = 7.0% to 16.1%)-were the most important mediators. Similarly, Latinx-White participation inequities were mediated (65.1%, 95% CI = 49.8% to 80.3%) by area-level SDOH, such as segregation (39.8%, 95% CI = 28.3% to 51.3%), limited English proficiency (11.6%, 95% CI = 2.8% to 20.4%), and vehicle ownership (9.6%, 95% CI = 5.8% to 13.5%). To improve racial and ethnic diversity in cancer trials, efforts to address barriers related to adverse neighborhood SDOH factors are necessary.

A 3D photon-to-digital converter readout for low-power and large-area applications

A new trend in large area noble liquid experiments is to measure the scintillation light with photodetectors and their electronics inside the active volume. Compared to the typical approach of using silicon photomultipliers (SiPM) with an analog readout chain leading to an analog-to-digital converter, this paper presents a new 3D photon-to-digital converter (PDC) readout that takes advantage of the binary nature of the single-photon avalanche diodes (SPAD). The readout contains 4096 pixels over 25 mm2, each including a 3D bonding pad and a quenching circuit. The readout features three different outputs: a fast flag to get the timestamp of each event from an external time-to-digital converter, a digital sum to retrieve the number of pixels triggered during an event, and an analog monitor to generate an analog SiPM-like output. The analog monitor is also used to validate the two former digital outputs. The readout also includes 61 2D CMOS SPADs for validation purpose prior to the final 3D integration with SPADs custom made according to our design by Teledyne DALSA (Bromont, Canada). As a first system integration toward large-area detector applications, a mini-tile of 2 × 2 readouts has been developed to test all the functionalities. The measured single-photon timing resolution ranges from 72 to 93 ps FWHM across the mini-tile SPAD channels population (i.e. 4 × 61 channels). The flag timing resolution is below 95 ps RMS, which includes the contribution of the optimized flag H-tree but also an additional trigger tree that replaces the 3D SPAD array at this stage of development. Once bonded with the 3D SPADs, the trigger tree won't be required to measure the flag timing resolution. With the removed contribution of the trigger tree, the estimated flag timing resolution should be below 45 ps RMS. The extent of the benefits of the digital sum output depend on the application, and this paper focuses on two cases. First, a low-power coincidence scheme such as required by the nEXO liquid xenon experiment, leading to a power consumption as low as 140 μW per PDC. With a finer sampling of the scintillation light such as required for pulse shape discrimination in liquid argon, the power consumption remains below 100 μW per PDC. Overall, this readout is designed as a replacement for a typical analog SiPM chain, without compromise on the performances.

Global analysis of forest tipping points leading to changing water cycle dynamics

Forest cover loss is increasing at unprecedented rates, affecting the hydrologic systems of major freshwater-producing regions of the world. However, quantification of the tipping points of forest cover loss before hydrologic changes manifest and their impact in water yield and climatic conditions has remained elusive. In this study, we aim to systematically document the critical thresholds of tree cover loss leading to changing hydrologic functioning within regions that experienced extensive drought, fire, or clear-cutting events spanning different climates during the period from 2001 to 2016. Using the Hydrologic Sensitivity Index based on Budyko’s curve, we analyzed the changes in hydrologic responses to climate variability as landcover changes across the affected forests. Critical thresholds were derived by fitting Richard’s Curve function to the observed relationship between growing sensitive area and tree cover loss. Our analysis reveals decrease in water yields and warming trends during the early stages of tree cover loss in tropical forests (c = 16 %), with negative anomalies observed in rainforests of Central Africa and Maritime Southeast Asia. Boreal forests also show low thresholds (c = 18 %) with a strong tendency toward a warmer climate state and no clear tendency in water yields. Mixed forests show moderate thresholds (c = 25 %) with unclear water yield and climate trends. Conversely, Temperate forests exhibit the most resilience to hydrologic regime shifts with high critical thresholds of tree cover loss (c = 46––54 %), but a rapid alteration once their threshold is surpassed resulting primarily in increased water yields and a shift toward cooler climate conditions. As the potential for additional tree cover loss heightens, due to expected increases in the frequency and intensity of droughts and wildfires, the analyses presented provides a quantitative framework to monitor and assess the impacts of changing forest cover conditions on the water cycle behavior of some of the largest freshwater producing regions of the world.