AGU Advances最新文献

Cratonic lithospheres carry a long history of tectonic modifications that result in heterogeneous structures, as revealed by an increasing number of geophysical observations. The existence of cratonic basins indicates protracted periods of tectonic modification, causing subsidence within global continental interiors. An enigmatic aspect of this process is the cessation of subsidence in cratonic basins with unclear mechanisms. Here, using full-wave ambient noise tomography, we reveal distinct seismic low-velocity anomalies below 60 km beneath the Illinois and Michigan Basins, where subsidence terminated in the late Paleozoic to the early Mesozoic. These low-velocity volumes, surrounded by distinctly higher velocities, are attributed to asthenospheric materials upwelling to shallow mantle depths during lithospheric foundering or delamination. This lithospheric modification may be associated with a major regional tectonic exhumation in the early Mesozoic that could have terminated basin subsidence and unroofed upper portions of basin stratigraphy. This timing coincides with the passage of this region over mantle plumes, which likely triggered lithospheric delamination and asthenospheric upwelling. Geodynamic modeling shows that the emplacement of these buoyant asthenospheric materials would lead to an uplift of about 3.5 km, sufficient to terminate the subsidence in the cratonic basins within this region. These findings document evidence of lithospheric delamination in the North American midcontinent and present important links between geodynamic drivers and geological records of the evolution of the cratonic lithosphere in North America and beyond. They also offer broader implications for understanding how deep Earth processes shape surface environments, influencing resource distribution and long-term landscape evolution.

Geomorphic and stratigraphic studies of Mars prove that extensive liquid water flowed and pooled on the surface early in Mars' history. Martian paleoclimate models, however, have difficulty simulating climate conditions warm enough to maintain liquid water on early Mars. Reconciling the geologic record and paleoclimatic simulations of Mars is critical to understanding Mars' early history, atmospheric conditions, and paleoclimate. This study uses an adapted lake energy balance model to investigate the connections between Martian geology and climate. The Lake Modeling on Mars for Atmospheric Reconstructions and Simulations (LakeM²ARS) model is modified from an Earth-based lake model to function in Martian conditions. We use LakeM²ARS to investigate the conditions necessary to simulate a lake in Gale crater. Working at a localized scale, we combine climate input from the Mars Weather Research & Forecasting general circulation model with geologic constraints from Curiosity rover observations to identify potential climatic conditions required to maintain a seasonally ice-free lake. Our results show that an initially small lake system (10 m deep) with ∼50 mm monthly water input and seasonal ice cover would retain seasonal liquid water for over 100 years, demonstrating conditions close to long-term lake survivability. These results are an important step in resolving the historic disconnect between climate and geology on Mars. Continued use and iteration of LakeM²ARS will strengthen connections between Mars' paleoclimate and geology to inform climate models and enhance our understanding of conditions on early Mars.

Root distributions are typically based on root mass per soil volume. This plant-focused approach masks the biogeochemical influence of fine roots, which weigh little. We assert that centimeter-scale root presence-absence data from soil profiles provide a more soil-focused approach for probing depth distributions of root-regolith interfaces, where microsite-scale processes drive whole-ecosystem functioning. In 75 soil pits across the continental USA, Puerto Rico, and the Alps, we quantified fine and coarse root presence as deep as 2 m. In 70 of these pits we estimated root mass and created standardized metrics of both data sets to compare their depth distributions. We addressed whether: (a) depth distributions of root presence-absence data differ from root mass data, thus implying different degrees of root-regolith interactions with depth; and (b) if root presence or any depth-dependent differences between these data sets vary predictably with environmental conditions. Presence of fine roots exhibited diverse depth-dependent patterns; root mass generally declined with depth. In B and C horizons, standardized root presence was greater than standardized root mass; random forest analyses suggest these discrepancies are greater in B horizons with increasing mean annual precipitation and in C horizons with increasing mean annual temperature. Our work suggests that deep in the subsurface, biogeochemical and reactive transport processes result from more numerous root-regolith interfaces than mass data suggest. We present a new paradigm for discerning patterns in depth distributions of root-regolith interfaces across multiple biomes and land uses that promotes understanding of the roles of those interfaces in driving key critical zone processes.

A recent report released by the U.S. Department of Energy concludes that U.S. tide-gauge data in aggregate provide no evidence for relative sea-level (RSL) acceleration above the historical mean trend. However, that conclusion rests largely on cursory analysis of a small number of tide-gauge records that are known to be unrepresentative of large-scale RSL behavior. Here I analyze all long active tide-gauge RSL data records on the contiguous U.S. (CONUS) coast to make a comprehensive estimate of spatially averaged RSL changes over the CONUS (CONUS RSL) during the past 125 years. I find that long-term rates of CONUS RSL rise doubled in the past century, from about 1.7 mm $��{\text{yr}}^{�-�1�}�$ in 1900 to roughly 4.3 mm $��{\text{yr}}^{�-�1�}�$ in 2024, and that recent rates are higher than the longterm historical mean rate since 1900, which is approximately 3.0 mm $��{\text{yr}}^{�-�1�}�$. That is, CONUS tide gauges give obvious evidence of RSL acceleration, which is likely related to ongoing climate change.

Our understanding of the formation and evolution of planetary systems has made major advances in the past decade. This progress has been driven in large part by the Atacama Large Millimeter/submillimeter Array (ALMA), which has given us an unprecedented view of solar system bodies themselves, and of the structure and chemistry of forming exoplanetary systems. Within our own solar system, ALMA has enabled the detection of new molecules and isotopologues across moons and comets, as well as placing new constraints on the compositions and histories of small bodies through thermal emission observations. In this article, we highlight some key areas where ALMA has contributed to a deeper understanding of our solar system's formation and evolution, and place these discoveries in the context of our evolving understanding of protoplanetary disks.

Peat decomposition is progressing in Southeast Asia due to lowered groundwater levels (GWL) caused by drainage. Additionally, droughts during El Niño events significantly lower the GWL, the main environmental factor that controls greenhouse gas (GHG; carbon dioxide (CO₂) and methane) emissions in peatlands. Consequently, tropical peatlands have been recognized as a significant source of carbon emissions, and these emissions have been estimated for the region using constant decomposition rates of peat for each land use (Tier 1 emission factors). However, these factors hardly reflect the spatiotemporal variation of the GWL. Furthermore, these estimates do not account for CO₂ uptake through photosynthesis. To reduce uncertainty, we developed a method to estimate spatiotemporal GWL variation from satellite-derived antecedent precipitation. Using the estimated GWL, we calculated the monthly net ecosystem-scale GHG emissions from peat forests and managed peatlands using the observed relationship between eddy covariance GHG fluxes and GWL, though carbon losses from deforestation, fires, and fluvial export were not covered in this study. Spatiotemporal variations in GHG emissions across Sumatra, Borneo, and the Malay Peninsula over a decade revealed the following: (a) Peat forests are a net source of CO₂-equivalent GHGs, even when undrained, (b) Decadal mean annual GHG emission rates increase 2.8-fold when forests are drained and 6.4-fold when undrained forests are converted to managed peatlands, (c) Droughts increase total annual GHG emissions by 16% across the study area. Additionally, climate models projected precipitation increase in the mid-21st century, suggesting an increase in GWL and a consequent reduction in peat decomposition.

Managing carbon stocks in the land, ocean, and atmosphere under changing climate requires a globally-integrated view of carbon cycle processes at local and regional scales. The growing Earth Observation (EO) record is the backbone of this multi-scale system, providing local information with discrete coverage from surface measurements and regional information at global scale from satellites. Carbon flux information, anchored by inverse estimates from spaceborne Greenhouse Gas (GHG) concentrations, provides an important top-down view of carbon emissions and sinks, but currently lacks global continuity at assessment and management scales (<100 km). Partial-column data can help separate signals in the boundary layer from the overlying atmosphere, providing an opportunity to enhance surface sensitivity and bring flux resolution down from that of column-integrated data (100–500 km). Based on a workshop held in September 2024, the carbon cycle community envisions a carbon observation system leveraging GHG partial columns in the lower and upper troposphere to weave together information across scales from surface and satellite EO data, and integration of top-down/bottom-up analyses to link process understanding to global assessment.

Atmospheric rivers (ARs) play a dominant role in water resource availability in many regions, and can cause substantial hazards, including extreme precipitation, flooding, and moist heatwaves. Despite this, there is substantial uncertainty about recent and ongoing changes in AR frequency and impacts. Here, we place recent observed trends in their longer-term context using AR records extending back to 1940. Our results show that AR frequency has increased broadly across the midlatitudes, bridging the apparent discrepancy between the observed satellite-era poleward shift and the general increase simulated in climate change projections. This increase in AR frequency enhances AR-associated precipitation and snowfall across their region of influence in the mid- and high-latitudes. We also find that, despite warmer surface temperatures associated with ARs, there is a decrease in the magnitude of AR-associated temperature anomalies in high-latitude regions due to Arctic amplification. An increase in AR-associated humid heatwaves underscores the societal importance of changing AR activity.

With its 1 day lasting 243 days on Earth, Venus is the slowest-spinning planet in the Solar System and its rotational bulge is anomalously small. A rotational bulge stabilizes the orientation of planets. Having only a tiny stabilizer, the rotational pole of Venus has been expected to separate from the figure pole in response to mantle flow, which has been used to explain why both poles are observed to be 0.5° apart. Here, we couple 3D mantle-convection simulations and polar motion dynamics to explore how mantle flow, and in particular surface mobilization, drives Venus's polar motion. We provide a predictive framework for polar motion on slow rotators and show that the spin/figure pole separation (or offset) follows a simple law: it scales with the figure-axis drift rate times the planet's Chandler period. Contrary to prior expectations, stronger internal loading does not amplify the offset, and the mantle-driven polar motion is smooth rather than wobbly, more similar to that of fast rotators. In models matching Venus's geoid, figure-axis drift rates reach only up to a few °/Myr, too slow compared to ca. 60°/Myr that is needed to match the observed offset. We therefore exclude mantle convection as the cause of Venus' spin and figure poles separation, and suggest that atmospheric and solid tides are not balanced instead.

Earth System Models (ESM) are crucial for quantifying climate impacts across Earth's interconnected systems and supporting science-based adaptation and mitigation. However, not including end-users, especially decision-makers representing communities vulnerable to climate change, can limit model utility, increase epistemic risks, and lead to information misuse in decision-making. While the ESM community increasingly values broad community engagement, end-users may not initially perceive models as useful for local planning. Co-designing models with end-users fosters two-way learning: users better understand models and their outputs, while modelers gain insights into fine-scale local processes like monitoring practices and management priorities. Higher-level co-design can lead to more customized, priority-driven, and useful modeling products. Despite these benefits, modelers often struggle to initiate meaningful partnerships with local communities. Therefore, this paper explores model co-design from the perspective of modelers. This study presents two case studies where modelers and social scientists collaborated with Indigenous communities' decision-makers to reflect their priorities in model design and application. In the Arctic Rivers Project, high-resolution climate and hydrology data sets for Alaska were developed with guidance from an Indigenous Advisory Council, using optimized, coupled land-atmosphere models. In the Mid-Klamath Project, we partnered with the Karuk Tribe's Department of Natural Resources to assess climate change and prescribed burning impacts on terrestrial hydrology in the Klamath River Basin. Drawing from these studies, we introduce a four-level framework: (a) Co-design Configuration; (b) Model Tuning; (c) Incorporate Contextual Knowledge; (d) Co-develop New Model Functions. We aim to help researchers consider and compare co-design across diverse modeling projects systematically and coherently.