Paper Harvest Report

Date range: March 20, 2026

4 top-tier papers selected out of 76 total publications

Today’s Highlights

A precipitation-themed slate from GRL: Feldman et al. show that “fewer but more intense” daily rainfall trends are widespread across global land, complicating annual-total interpretations and pointing toward water-supply and flood implications. Two complementary modeling/monitoring papers examine why CMIP6 misses heavy monsoon rainfall (zonal moisture-advection biases) and how anthropogenic seismic noise can be repurposed as a high-resolution proxy for rainfall infiltration and groundwater pumping. A BAMS essay rounds out the day with a timely call for tighter terminology and documentation around “reanalysis” datasets used to train ML models.

Today’s Highlights
Top-Tier Journal Papers
Statistics
1. Papers by journal
Filtering Criteria

Top-Tier Journal Papers

Widespread Co‐Location of Less Frequent and More Intense Daily Precipitation Over Land

Authors: Andrew F. Feldman, Xue Feng, Andrew J. Felton, Wade T. Crow, Joel A. Biederman, Alexandra G. Konings et al.

Journal: Geophysical Research Letters · DOI: 10.1029/2025gl120745

Matched topics: flood

Under increasingly variable rainfall, trends toward more intense and less frequent daily‐scale precipitation have been identified using regional and global averages. However, it has not been explicitly demonstrated whether and where these trends are co‐located, which is important given their potential impacts on land surface processes. Here, using global observation and model‐based data sets, we find that trends toward fewer, larger daily precipitation events are common and relatively distributed across terrestrial ecosystems; they are approximately as common as trends toward more, larger daily precipitation events (which underpin increases in annual precipitation totals). Therefore, widespread precipitation intensification is not consistently increasing annual precipitation totals partly because precipitation events, especially of small‐to‐moderate depths (<10 mm/day), are simultaneously becoming less frequent. Independent of the consequences of changes in mean annual precipitation, these daily‐scale precipitation alterations can substantially impact water resource availability, floods, land‐atmosphere interactions, crop yields, wildfire fuel loads, and carbon sequestration.

Monitoring Near‐Surface Changes Using Anthropogenic Seismic Vibrations

Authors: Sayan Mukherjee, Yunyue Elita Li, Andrew Stumpf, Heng Zhang

Journal: Geophysical Research Letters · DOI: 10.1029/2025gl120455

Matched topics: irrigation

The abundance of anthropogenic seismic noise provides a valuable resource for monitoring dynamic subsurface property changes. We utilize railway and wind turbine‐induced vibrations recorded by a geophone array to estimate surface wave attenuation and velocity in the underlying glacial deposits of Illinois. Over the 12‐month monitoring period, attenuation (Q⁻¹) varied over ±25%, demonstrating stronger sensitivity to environmental changes than velocity, which varied over ±2%. We observe sharp attenuation increases typically following heavy rainfall episodes and rapid attenuation decreases during intense groundwater pumping for irrigation. These findings demonstrate that seismic attenuation, measured from persistent anthropogenic noise sources, provides a robust, highly sensitive, and high–temporal–resolution proxy for continuous near‐surface hydrological processes, complementary to traditional seismic velocity‐based methods.

Mean Biases Dominate CMIP6 Model Deficiencies in Simulating Heavy Rainfall During Monsoon Intraseasonal Oscillation

Authors: Qingjian Shi, Jianhuang Qin, Baosheng Li

Journal: Geophysical Research Letters · DOI: 10.1029/2025gl119176

Matched topics: seasonal

The monsoon intraseasonal oscillation (MISO) modulates heavy precipitation during the Indian summer monsoon, but accurate rainfall simulation remains a challenge in state‐of‐the‐art models. Our analysis reveals a moisture bias in MISO using simulations from the sixth phase of the Coupled Model Intercomparison Project models. Despite an overestimated local moisture supply from the ocean in the Bay of Bengal (BoB), atmospheric moisture content before MISO‐induced heavy rainfall is underestimated. Model diagnoses reveal deficient zonal moisture advection in the low‐level troposphere as the primary constraint, attributable to a weaker intraseasonal Arabian Sea moisture source and a bias in the mean zonal winds. The latter inhibits the development of barotropic instability and the energy transfer to intraseasonal timescales. Consequently, the weakly simulated MISO leads to insufficient moisture supply over the BoB, resulting in failed simulation of heavy rainfall during MISO events. These findings highlight potential approaches for model improvements to enhance monsoon precipitation simulations.

Reanalyses in the Age of Machine Learning: Why Dataset Curation Matters Now More Than Ever

Authors: Mimi Rose Abel, Alexander J. Thompson, Ethan D. Gutmann, Kelly Mahoney, Rachel Rose McCrary, Russ S. Schumacher et al.

Journal: Bulletin of the American Meteorological Society · DOI: 10.1175/bams-d-25-0149.1

Matched topics: earth system model

As machine learning becomes ever more prevalent within earth and atmospheric science, clear and consistent descriptions of models, observations, and observations-based datasets, particularly reanalyses, are increasingly vital. Reanalyses remain foundational for climate and weather research, but advancements in data assimilation and model nudging methods, as well as increasingly complex physical parameterization options, mean that not all variables within reanalyses are equally constrained by observations. Because machine learning models are often trained and evaluated on such datasets, imprecise terminology and inadequate documentation can lead to a loss of information content, mislead users unfamiliar with data nuances, lead to the training of flawed machine learning models, and ultimately result in model evaluations that do not realistically describe performance relative to observations. This essay argues for more careful use of the term “reanalysis”, emphasizing that it should be reserved for datasets that explicitly blend observations with models through data assimilation. It highlights the rise of “reanalysis-adjacent” datasets, as well as the growing disconnect between data producers and increasingly interdisciplinary users, particularly within the machine learning community. It offers guidance for dataset producers and users, alongside recommendations to enhance transparency, including renewed use of variable classification systems, better documentation of variable-specific uncertainties, and greater community-wide emphasis on data transparency. Without such efforts, Earth science datasets may be applied indiscriminately, regardless of fitness for purpose. Ensuring trustworthy and interpretable data is essential for maintaining the scientific integrity of Earth system modeling in the machine learning age.

Statistics

Metric	Count
Journals searched	11
Total papers fetched	76
Passed deterministic filter	6
After LLM relevance filtering	4
Rejected (not relevant)	2

Papers by journal

Journal	Papers
Geophysical Research Letters	3
Bulletin of the American Meteorological Society	1

Filtering Criteria

Topics: hydrology, hydrological, watershed, river, streamflow, runoff, flood, drought, reservoir, dam, irrigation, groundwater, precipitation, snow, glacier, water resources, water management, climate change, land surface, earth system model, MOSART, ELM, E3SM, hydropower, water cycle, evapotranspiration, soil moisture, seasonal, lake

Fields: Environmental Science, Geology, Geography, Engineering