Paper Harvest Report

Date range: June 18, 2026

4 top-tier papers selected out of 116 total publications

Today’s Highlights

Tropical cyclones turn out to be a far more pervasive driver of extreme hourly precipitation across the contiguous U.S. than previously recognized — often via remote atmospheric-river moisture transport, affecting 76% of gauge stations since 1980 (GRL). On the Klamath River, post-wildfire debris flows reshaped river metabolism, spiking then suppressing productivity while leaving turbidity elevated for over a year (GRL). A global risk assessment finds compound drought, flood, and heatwave exposure to croplands intensifying fastest across South Asia, Sub-Saharan Africa, and major U.S./European breadbaskets (Nature Communications), while a companion land-modeling study warns that Earth System Models underestimate by 2-3x how rising atmospheric dryness suppresses tree growth (GRL).

Today’s Highlights
Top-Tier Journal Papers
AI for Science
1. How AI is changing research
Statistics
1. Papers by journal
Filtering Criteria

Top-Tier Journal Papers

Contribution of Tropical Cyclones to Hourly Precipitation Extremes in the Contiguous United States

Authors: Dmitri A. Kalashnikov, John T. Abatzoglou, Alyssa M. Stansfield

Journal: Geophysical Research Letters · DOI: 10.1029/2026gl122815

Matched topics: river, flood

Tropical cyclones (TCs) are major drivers of contiguous United States (CONUS) flooding, yet their contribution to hourly precipitation extremes remains poorly quantified. Here we link observations from 420 gauges (1980–2024) with TC track data to attribute extreme hourly precipitation to both local and remote TCs. Remote TC contributions are identified via atmospheric river (AR) objects using the TempestExtremes framework applied to a global AR database. We identify 254 TCs that contributed to extreme hourly precipitation, affecting 76% of stations. While local contributions from Atlantic TCs are prevalent in eastern CONUS, remote influences from Pacific TCs contribute to precipitation extremes in the rest of CONUS. The number of TCs resulting in extreme hourly precipitation has increased significantly in northeast and southeast CONUS since 1980. These results reveal that TC‐linked moisture represents a contributor to short‐duration precipitation extremes across a broader area of CONUS than previously recognized.

Debris Flows Suppressed Riverine Productivity and Respiration Following High‐Severity Wildfire on the Klamath River, California

Authors: Laurel Genzoli, John R. Oberholzer Dent, Robert O. Hall, Grant Johnson

Journal: Geophysical Research Letters · DOI: 10.1029/2026gl122846

Matched topics: river

Wildfires alter terrestrial material and energy inputs to rivers, but impacts to ecosystem metabolism and dissolved oxygen (DO) are not well known due to few river ecosystem response studies during and after wildfires. In summer 2022, wildfire and heavy rains triggered debris flows into the Klamath River, where we compared immediate (days) and mid‐term (1.5 y) post‐debris flow metabolism (gross primary production; GPP and ecosystem respiration; ER) to pre‐debris flow magnitudes (2018–2023). The sediment pulse temporarily spiked ER, leading to anoxia that lasted <1 d. Following the debris flows, turbidity remained elevated for the duration of the study, limiting riverbed light and suppressing GPP. ER magnitude decreased, but less than GPP, reducing net ecosystem production. Despite more heterotrophic days after the debris flows, average DO minima increased due to lower metabolic fluxes. While postfire debris flows can cause catastrophic effects to water quality, small increases in turbidity may improve DO in eutrophic rivers.

Multi-hazard exposure and vulnerability drive rising agricultural risk in global breadbasket regions

Authors: Hossein Tabari, Parisa Hosseinzadehtalaei

Journal: Nature Communications · DOI: 10.1038/s41467-026-74617-5

Matched topics: flood, drought

Climate extremes increasingly threaten global food production, yet global multi-hazard agricultural risk assessments that jointly consider multiple hazards, crop exposure and socioeconomic vulnerability are still missing. This gap limits our ability to identify where climate shocks to croplands are most likely to translate into high agricultural risk. Here we present a global agricultural risk assessment that quantifies the exposure of croplands and ten major crop groups to heatwaves, droughts, floods and compound hot–dry events, and combines this with a vulnerability index constructed from socio-economic and institutional indicators using latent factor modelling. Using ensembles of climate and hydrological models under multiple future scenarios, we show widespread increases in multi-hazard exposure, with the strongest intensification across South Asia, Sub-Saharan Africa and major U.S. and European breadbaskets. A significant relationship between vulnerability and cropland exposure emerges in future scenarios but is absent historically, indicating a growing role for vulnerability in shaping future agricultural risk. These results demonstrate that reliably identifying future agricultural risk hotspots requires jointly accounting for multi-hazard exposure and socioeconomic vulnerability, so that adaptation efforts can be targeted to regions where intensifying extremes coincide with high vulnerability.

Land Models Likely Underestimate the Impact of Future Atmospheric Dryness on European Tree Growth

Authors: Brendan Clark, Daniele Visioni, Shan Kothari, Manuel Lerdau

Journal: Geophysical Research Letters · DOI: 10.1029/2026gl122759

Matched topics: river, climate change, earth system model

Understanding how the land carbon sink will be altered by climate change is critical for projecting future atmospheric CO2 concentrations under different emission pathways in Earth System Models (ESMs). Most land models assume that woody growth, the main driver of land carbon storage, is driven by photosynthetic carbon supply. In reality, growth is often constrained by physiological processes limiting cell division independent of carbon availability, like reduced turgor pressure. Rising vapor pressure deficit (VPD) under climate change reduces turgor pressure and suppresses actual growth more strongly than photosynthesis, suggesting ESMs may overestimate future terrestrial carbon uptake. We compare simulations of the Community Land Model (CLM) to a statistical model built on dendrometer observations, showing that CLM underestimates decreases to growth under elevated VPD by a factor of 2–3. This highlights the need for land models to represent physiological constraints on tree growth to improve projections of future terrestrial carbon uptake.

AI for Science

How AI is changing research

As better chatbots get harder to build, AI turns to simulated worlds (Science, 2026-06-25) — As text-pretraining gains plateau, AI labs are shifting toward agents that learn by acting in virtual environments — a methodology shift relevant to anyone building agentic research workflows (e.g. RL-trained surrogate models or simulation-in-the-loop agents for hydrologic forecasting).

Statistics

Metric	Count
Journals searched	11
Total papers fetched	116
Passed deterministic filter	9
After LLM relevance filtering	4
Rejected (not relevant)	5
AI for Science items picked	1

Papers by journal

Journal	Papers
Geophysical Research Letters	3
Nature Communications	1

Filtering Criteria

Topics: hydrology, hydrologic model, river, runoff, streamflow, reservoir, water management, flood, drought, seasonal, land surface model, climate change, hydropower, surface water, irrigation, earth system model, estuary, coastal, freshwater discharge, river plume, ocean biogeochemistry, marine heatwave, paleohydrology, paleoclimate, Quaternary, Holocene, Pleistocene, fluvial geomorphology, river terrace, loess, drainage network, river capture, landscape evolution, luminescence dating

Fields: engineering, environmental science, computer science, geology, geography