Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.06476

Evaluating LiDAR Data Sources, Predictor Resolution, and Spatial Random Effects in Bayesian Change-of-Support Models for Forest Inventory

Arthur Spitta·Andrew O. Finley·Christoph Gollob·Ralf Kraßnitzer·Tim Ritter·Andreas Tockner·Arne Nothdurft
View original source ↗Read source document ↗Export metadata ↓

Abstract

Forest managers require timely stand-level information for operational planning. Model-based estimation combines sparse field samples with remotely sensed auxiliary data to estimate growing stock volume (GSV) for small-area units. This is particularly relevant in mixed-species and structurally heterogeneous forests, where timely structural information can support management decisions under climate change and associated disturbance pressures. Uncrewed aerial vehicle laser scanning (ULS) provides flexible access to high-resolution LiDAR data, but its benefits over conventional airborne laser scanning (ALS) for model-based inference remain insufficiently understood. We compared public ALS and new ULS data using Bayesian change-of-support models to estimate GSV in a mixed-species forest in north-eastern Germany. We assessed the effects of distributional LiDAR metrics and spatial random effects. ULS models consistently outperformed ALS models, with cross-validated root mean squared prediction errors (RMSPEs) of 68.5 m^3/ha and 79.5 m^3/ha , respectively, a 13.8 % reduction in prediction error. Distributional metrics benefited ULS models more than ALS models, reducing RMSPE by up to 10.1 %, whereas spatial effects yielded minor improvements at greater computational cost. ULS models also exhibited lower stand-level predictive uncertainty for latent stand-mean GSV. This advantage may partly reflect closer temporal alignment with field data and finer-scale predictor information. These findings suggest that timely, information-rich LiDAR data may be more beneficial for stand-level GSV estimation than increasingly complex spatial model structures. ULS is promising where timely acquisition and high-resolution canopy characterization are operationally feasible. Controlled comparisons with temporally matched ALS and ULS data are needed to distinguish platform effects from temporal mismatch.

Explore related subjects

stat.AP

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Arthur Spitta, Andrew O. Finley, Christoph Gollob, Ralf Kraßnitzer, Tim Ritter, Andreas Tockner, Arne Nothdurft. 2026-10-05. Evaluating LiDAR Data Sources, Predictor Resolution, and Spatial Random Effects in Bayesian Change-of-Support Models for Forest Inventory. https://arxiv.org/abs/2610.06476

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Addressing the Between-Group Comparison Problem: Detecting Differences Between Correlation Matrix Populations due to Single-variable Perturbations for Resting State fMRI

Resting-state fMRI has been known for decades as a promising method for evaluating cognitive and mental states, both in health and especially in disease, due to its ease of implementation as a short, standard MRI protocol. In clinical settings, a group of patients with a given disorder is typically compared to a group of healthy controls. This poses an inherent challenge of between-group comparison. We propose a new efficient model for characterizing changes to the temporal synchronization of brain activity measured using RS-fMRI between groups, summarized as individual correlation matrices. Our model posits that the between-group differences are the product of single-region effects describing the increase or decay of synchronization with the rest of the brain. This parsimonious model pools the correlation coefficients of each region with all others, and therefore can detect differences between groups even in small samples. Inference for this model accounts for the variability in individual correlation matrices, the within-group differences across individuals, and for the approximation error of the single-region model. This results in per-region estimates and confidence intervals for the parameters governing the difference between groups. In simulations, our model shows increased power to detect model-aligned alternatives compared with competing approaches. To demonstrate feasibility of the method in a clinical application, we use the model to analyze RS-fMRI correlation matrices in patients with transient global amnesia and healthy controls. Our model detects significant decreases in synchronization for the patient population in the amygdala after multiplicity correction as well as borderline decreases in memory-related brain regions that were not detected using mass-univariate tests without prior knowledge, suggesting its usefulness in the application of RS-fMRI in clinical settings.

stat.AP↗
arXiv2026

Evaluating cross-encoders for semantic similarity assessment in psychological questionnaires

Correlations between rating scales are commonly interpreted as evidence of convergent or discriminant validity, yet prior studies suggest that part of these associations may be attributable to semantic similarity between item wordings rather than to genuine construct overlap alone. Building on this evidence, largely derived from bi-encoders, the present study explores whether cross-encoders, which jointly encode item pairs, offer a suitable technique for detecting semantic overlapping between questionnaire items. Using response data from the NEO-FFI and the PID5BF+M (N = 502, Labek et al., 2024), we examined whether cross-encoder-derived semantic similarity estimates are associated with empirical item correlations, and whether cross-encoders offer a systematic advantage over bi-encoders. Across twelve cross-encoder models, semantic distance was consistently negatively associated with absolute item correlations, reaching statistical significance in two-thirds of the models, with R2 values of up to .37. However, cross-encoders did not consistently outperform bi-encoders based on the same base models. These findings extend prior evidence for semantic components in scale intercorrelations to cross-encoder architectures, while indicating that predictive value depends more on model-specific training characteristics than on encoder architecture itself.

stat.AP↗
arXiv2026

Spatio-Temporal Stochastic Interventions for Causal Inference in Climate Science

Estimating causal effects in climate science, such as the effect of anthropogenic warming on crop loss, is challenging because of complex spatio-temporal dependence and the high-dimensional nature of the treatment. To address this dependence and the resulting poor overlap between observed and counterfactual scenarios, we develop a spatio-temporal stochastic-intervention framework for estimating causal effects from climate observations. We introduce a regularized estimator of the stochastic-intervention treatment effect that trades a controlled bias for a reduction in the weight variance caused by poor overlap. Simulation studies show that this estimator attains lower mean squared error than alternative weighting estimators and removes the confounding bias of an unadjusted estimator. We apply the framework to estimate the effect of historical warming on vapor-pressure deficit, a driver of crop stress, adjusting for precipitation, which confounds the effect by affecting both temperature and humidity. In GISS-E2-1-G climate-model simulations, the global effect is distinguishable from zero in every year from 1995 onward, and omitting the precipitation adjustment inflates the global estimate by 47%. Adjustment reverses the sign of the estimate over 8% of global cropland (125 million hectares), where an unadjusted analysis could misdirect adaptation between heat-focused and moisture-focused measures.

stat.AP↗