Climate science increasingly relies on statistical tools to evaluate and synthesize diverse datasets, including observations, reanalyses, paleoclimate proxies, and model simulations. Climate data differentiation, which quantifies discrepancies between datasets, and climate integration, which combines information across them, play central roles in these efforts. This article reviews modern statistical approaches for climate data differentiation, including methods for quantifying discrepancies in means, dependence structures, distributions, extremes, spatial signals, and shapes. It also reviews approaches for climate data integration that combine information across models and observations through ensemble methods, bias correction, hierarchical modeling, and machine learning. We highlight how these methods address challenges such as high dimensionality, spatial dependence, limited observations, and structural model uncertainty. Together, these tools provide complementary perspectives for understanding differences among climate data sources and for synthesizing information across them, ultimately improving estimates of past, present, and future climate conditions.
Estimating conditional extreme quantiles is essential for assessing tail risks in complex systems, with broad applications in finance, climate science, engineering, and beyond. While classical extreme value theory provides a foundational framework, recent advances, particularly semiparametric and nonparametric methods, including approaches based on quantile regression, machine learning, and deep learning, have greatly enriched the methodological landscape. This modern review synthesizes these developments, covering traditional likelihood-based methods, semiparametric approaches, and tree-based and deep learning techniques, including higher-order refinements.
With increased data collection, the need to fuse data sources has emerged as an important and rapidly growing research activity in the statistical community. In considering spatial and spatio-temporal datasets to examine complex environmental and ecological processes of interest, we often have multiple sources that are jointly informative about features of interest of the processes. Model-based data fusion aims to leverage information from these sources to improve inference and prediction. In the spatial statistics setting, these data could be geostatistical; areal; or point patterns with varying spatial resolutions, supports, and domains. Given two or more sources, we explore stochastic modeling to implement a suitable fusion with full inference and uncertainty quantification. We illustrate these ideas using three environmental and ecological examples: precipitation, marine mammal abundance, and joint species distributions.
We review recent advances in change-point detection methods across three important fields of statistics: (a) We first present a subgroup identification method based on a multi-threshold change plane model where the subgroup boundaries are defined by a high-dimensional hyperplane in the covariate space. Subjects grouped into different regions may receive more individualized treatments in medical research studies and achieve improved health outcomes. (b) We then consider the estimation of discontinuity for functional process data. Many longitudinal or functional responses may exhibit abrupt jumps, and our methodology effectively accommodates such complicated nonsmooth features. (c) Finally, we explore change-point estimation within dynamic networks using a recently proposed network autoregressive model. This framework demonstrates that community structures in networks can shift similarly to changes observed in time series data. These reviews highlight the wide-ranging applications of change-point detection methodologies in modern data analysis.
This article examines the role of statistics in the humanitarian sector, with a particular focus on disasters caused by natural hazards. It begins by outlining current applications, including primary data collection, anticipatory action frameworks, Earth observation, mobile positioning data, and artificial intelligence. It then highlights key challenges such as gaps and biases in disaster impact and response data, difficulties in communicating statistical findings clearly, inequities in aid allocation, and the widespread outsourcing of statistics-related work. In exploring future applications, the article discusses the potential of impact-based early warning models, dynamic population data, and artificial intelligence to enhance communication and decision-making. Throughout, emphasis is placed on the need for interoperable systems as well as ethical and inclusive data practices. In doing so, the article presents statistics as both a diagnostic and strategic tool for strengthening the effectiveness, fairness, and responsiveness of humanitarian action in disaster contexts.
