In each of the last three decades of the 20 th century there were unprecedented expansions of sea-ice over the Labrador Sea basin and influxes of cold fresh water into the subpolar gyre which have been described as the Great Salinity Anomalies (GSAs). Employing data for sea surface temperature, salinity and sea ice cover, we propose that these events were downstream consequences of the expansion and subsequent melting of so-called 'Odden' ice formed over the deep basin of the Greenland-Iceland-Norway (GIN) Sea in the 1960s, 1970s and1980s and additional to the normal East Greenland shelf sea-ice. We expand previous findings that Odden ice expansions were linked to winter episodes of high atmospheric pressure north of Greenland that directed freezing Arctic winds across the GIN Sea and may also have been associated with increased Arctic sea-ice volume leading to enhanced ice export through Fram Strait. We show that cold water and ice derived from Odden melting in the summer passed through Denmark Strait and along the East Greenland shelf, and accumulated in the Labrador Sea, creating favourable conditions for winter ice formation during particularly cold years in southwest Greenland. Meltwater from Odden and Labrador Sea ice appeared to break out into the subpolar gyre in the fall of 1982 and 1984 respectively and this cold water represents the likely source of the 1982-1985 GSA. These findings further our understanding of the physical processes linking ice formation and melt with ocean circulation in this key component of the climate system.

The hydrogen isotopic compositions of the leaf-wax n-alkanes (δDalk) and n-alkanoic acids (δDacid) are known to reflect ambient climatic conditions (including precipitation δD values, δDPrecip). However, the climatic conditions of exactly which period (i.e. early or entire period of the leaf’s lifespan) these biomarkers represent, i.e. the seasonality in δDalk and δDacid records, is still evolving. The seasonality studies on the δDalk and δDacid values, done only in extra-tropical regions, mostly indicate the δDalk values are biased towards the early growing season whereas δDacid values are not biased towards any season. To decipher the seasonality in the δDalk and δDacid records from the tropics, we conducted a long-duration experiment wherein deciduous and evergreen species were grown using normal water (δD = −2‰) during the early stages of the leaf's growth and later using isotopically-labeled water (δD = 1000‰). Our experiment revealed (i) in deciduous and evergreen species, δDalk and δDacid values reflect δDPrecip during the early stages of the leaf's growth, (ii) synchronous synthesis of n-alkanes and n-alkanoic acids, and (iii) in deciduous species, minor incorporation of the previous year’s photosynthates in the leaf wax pool of the current year’s mature leaves. Our study suggests that the δDalk and δDacid records in the tropics are biased towards the climatic conditions prevailing during the early stages of the leaf's growth. This bias should be considered while comparing the δDPrecip values generated from the leaf wax proxy records and isotope-enabled atmospheric circulation models.

A document by 張澤宇. Click on the document to view its contents.

The Southern Ocean (SO) plays a fundamental role in the planet’s climate system, due to its ability to absorb and redistribute heat and CO2 (an important greenhouse gas). Besides, the SO interconnects three large oceanic basins the Pacific, the Atlantic, and the Indian Oceans, and it has an important role in the nutrient distribution into these oceans. However, the SO is poorly sampled with most measurements made in austral spring and summer. The variability of the air-sea CO2 flux is estimated, as well as the role of atmospheric and oceanic variables in this variability. The CO2 fluxes are calculated by the bulk parameterization method, in the Atlantic sector of the Southern Ocean, from 2003 to 2022, using in situ measurements, satellites and reanalysis data set. A neural network model is built to produce maps of the partial pressure of CO2 in seawater (pCO2sea). The CO2 flux varies from -0.05 to 0.05 gC m-² month-1. The Atlantic sector of the SO is a sink of CO2 in summer and spring and becomes a source in austral winter and autumn. The CO2 absorption intensifies from 2003 to 2022 by 7.6 mmol m-²month-1, due stronger westerly winds, related to the trend of the positive phase of the Antarctic Oscillation and the extreme events of El Niño and La Niña.

Climate change is anticipated to considerably reduce global marine fish biomass, driving marine ecosystems into unprecedented states with no historical analogues. The Time of Emergence (ToE) marks the pivotal moment when climate conditions (i.e. signal) deviate from pre-industrial norms (i.e. noise). Leveraging ensemble climate-to-fish simulations, this study examines the ToE of epipelagic, migratory and mesopelagic fish biomass, alongside their main environmental drivers, for two contrasted climate-change scenarios. Globally-averaged biomass signals emerge over the historical period. Epipelagic biomass decline emerges earlier (1950) than mesozooplankton decline (2000) due to a stronger signal in the early 20th century, possibly related to trophic amplification induced by an early-emerging surface warming (1915). Trophic amplification is delayed for mesopelagic biomass due to postponed warming in the mesopelagic zone, resulting in a later emergence (2000). ToE displays strong size class dependence, with medium sizes (20 cm) experiencing delays compared to the largest (1 m) and smallest (1 cm) categories. Regional signal emergence lags behind the global average, with median ToE estimates of 2029, 2034 and 2033 for epipelagic, mesopelagic and migrant communities, respectively, due to systematically larger local noise compared to global one. These ToEs are also spatially heterogeneous, driven predominantly by the signal pattern, akin to mesozooplankton. Additionally, our findings underscore that mitigation efforts (i.e. transitioning from SSP5-8.5 to SSP1-2.6 scenario) have a potential to curtail emerging ocean surface signals by 40%.

We present an application of the latest UCL-AGA magnetodisc model (MDISC) to the study of the magnetic and plasma conditions in the near-Ganymede space. By doing this, we provide a comparison with measurements from Juno’s most recent flyby of the Jovian moon, perijove 34 (PJ34). We find good agreement between the model results and the magnetometer data, pointing towards a hot plasma index value $K_h = \SI{2.719(24)e7}{\pascal\,\meter\,\tesla^{-1}}$ and an effective magnetodisc radius $r_{\text{max}} = \SI{79.5(11)}{}$ Jupiter radii for the Jovian magnetosphere, for the duration of the trajectory, suggesting a configuration with middling levels of expansion. We also predict the plasma conditions observed by Juno during the same flight-path, as well as the typical conditions over the orbit of Ganymede, with the magnetic and hot plasma pressures assuming dominant roles. Finally, these results are compared with functional fits of a compilation of Galileo flyby data obtained in the vicinity of Ganymede’s orbit, suggesting Juno experienced somewhat similar conditions, despite a systematic overestimation in magnetic field intensity in the near-Ganymede space.

Oceanic crust at fast-spreading ridges is formed by melt percolating through the Mohorovičić Transition Zone (MTZ), the boundary between crust and mantle. However, the relationship between the crustal structures and MTZ remains elusive. Applying full waveform inversion to wide-angle seismic data acquired near the 9oN East Pacific Rise, we show that the variations in crustal MTZ thicknesses are inversely correlated along the segment, although their total cumulative thickness shows little variations. These variations could be attributed to different melt migration efficiency through MTZ or variation in mantle thermal structures. Thin MTZ could be due to rapid percolation of melt from mantle to crust whereas the thick MTZ results from the crystallization of melt within the transition zone. On the other hand, for relatively hot segments, melt will accumulate at shallower depth within the lower crust. In contrast, melt could freeze at Moho depth for relatively cold segments thickening the MTZ.

A document by Shiqi Fang. Click on the document to view its contents.

Increasing anthropogenic CO2 emissions to the atmosphere are partially sequestered into the global oceans through the air-sea exchange of CO2 and its subsequent movement to depth, and this collective large-scale absorption is commonly referred to as the global ocean carbon sink. Quantifying this ocean carbon sink provides a key component for closing the global carbon budget which is used to inform and guide policy decisions. These estimates are typically accompanied by an uncertainty budget built by selecting what are perceived as critical uncertainty components based on selective experimentation. However, there is a growing realisation that these budgets are incomplete and may be underestimated, which limits their power as a constraint within global budgets. In this study, we present a methodology for quantifying spatially and temporally varying uncertainties in the air-sea CO2 flux calculations and data that allows an exhaustive assessment of all known sources of uncertainties, including decorrelation length scales between gridded measurements, and the approach follows standard uncertainty propagation methodologies. The resulting standard uncertainties are higher than previously suggested budgets, but the components are consistent with previous work, and they identify how the significance and importance of key uncertainty components change in space and time. For an exemplar method (the UEP-FNN-U method) the work identifies that we can currently estimate the annual ocean carbon sink to an accuracy of ±0.72PgCyr-1 (1 standard deviation uncertainty). Due to this method having been built on established uncertainty propagation and approaches, it appears applicable to all data-product assessments of the ocean carbon sink.

The north-south seismic belt of China poses a high risk of earthquakes, necessitating the need for accurate and rapid prediction of intensity measures (IMs) to prevent and mitigate potential damage. We have developed a new multi-task model, CRAQuake, to predict IMs for the north-south seismic belt of China. Using initial arrival seismic waves recorded at a single station as input, CRAQuake simultaneously predicts six IMs without relying on pre-configured parameters such as earthquake source, path, and location. The model was trained on 4281 sets of strong motion records datasets at 822 stations and tested to show highly correlated results with the target IMs. The prediction performance continues to improve as the input initial arrival seismic wave time window increases. CRAQuake promises to enhance the accuracy and timeliness of IMs prediction in the north-south seismic belt of China.

This paper continues the exploration of Machine Learning (ML) parameterization for radiative transfer for the ICOsahedral Nonhydrostatic weather and climate model (ICON). Three ML models, developed in Part I of this study, are coupled to ICON. More specifically, a UNet model and a bidirectional recurrent neural network (RNN) with long short-term memory (LSTM) are compared against a random forest. The ML parameterizations are coupled to the ICON code that includes OpenACC compiler directives to enable GPUs support. The coupling is done through Infero, developed by ECMWF, and PyTorch-Fortran. The most accurate model is the bidirectional RNN with physics-informed normalization strategy and heating rate penalty, but the fluxes above 15 km height are computed with a simplified formula for numerical stability reasons. The presented setup enables stable aquaplanet simulations with ICON for several weeks at a resolution of about 80 km and compare well with the physics-based radiative transfer solver ecRad. However, the achieved speed up when using the emulators and the minimum required memory usage relative to the GPU-enabled ecRad depend strongly on the Neural Network (NN) architecture. Future studies may explore physics-constraint emulators that predict heating rates inside the atmospheric model and fluxes at the top.

As climate modellers prepare their code for kilometre-scale global simulations, the computationally demanding radiative transfer parameterization is a prime candidate for machine learning (ML) emulation. Because of the computational demands, many weather centres use a reduced spatial grid and reduced temporal frequency for radiative transfer calculations in their forecast models. This strategy is known to affect forecast quality, which further motivates the use of ML-based radiative transfer parameterizations. This paper contributes to the discussion on how to incorporate physical constraints into an ML-based radiative parameterization, and how different neural network (NN) designs and output normalisation affect prediction performance. A random forest (RF) is used as a baseline method, with the European Centre for Medium-Range Weather Forecasts (ECMWF) model ecRad, the operational radiation scheme in the Icosahedral Nonhydrostatic Weather and Climate Model (ICON), used for training. Surprisingly, the RF is not affected by the top-of-atmosphere (TOA) bias found in all NNs tested (e.g., MLP, CNN, UNet, RNN) in this and previously published studies. At lower atmospheric levels, the RF can compete with all NNs tested, but its memory requirements quickly become prohibitive. For a fixed memory size, most NNs outperform the RF except at TOA. The most accurate emulator is a recurrent neural network architecture that closely imitates the physical process it emulates. The shortwave and longwave fluxes are normalized to reduce their dependence on the solar angle and surface temperature respectively. The model are, furthermore, trained with an additional heating rates penalty in the loss function.

River meanders are one of the most recurrent and varied patterns in fluvial systems. Multiple attempts have been made to detect and categorise patterns in meandering rivers to understand their shape and evolution. A novel data-driven approach was used to classify single-bend meanders. A dataset containing approximately 10 million single-lobe meander bends was generated using the Kinoshita curve. A neural network autoencoder was trained over the curvature energy spectra of Kinoshita-generated meanders. Then, the trained network was then tested on real meander bends extracted from satellite images, and the energy spectrum in the meander curvature was reconstructed accurately thanks to the autoencoder architecture. The meander spectrum reconstruction was clustered, and three main bend shapes were found associated with the meander datasets, namely symmetric, upstream-skewed, and downstream-skewed. The autoencoder-based classification framework allowed bend shape detection along rivers, finding the dominant pattern with implications on migration trends. By studying the shift in the prevailing bend shape over time, cutoff events were approximately forecast along the Ucayali River, whose migration was remotely sensed for 32 years. Overall, the method proposed opens the venue to data-driven classifications to understand and manage meandering rivers. Bend shape classification can thus inform restoration and flood control practices and contribute to predicting meander evolution from satellite images or sedimentary records.

In this study, we develop two dimensional (2-D) box models to identify the most viable reasons for the destruction of the North China Craton (NCC). We examine the role of flat slab-induced hydration, high-density lower crust, and weak mid-lithospheric discontinuity in our models. Results indicate that flat slab-induced hydration weakening of the eastern part of the NCC can lead to rapid craton destruction if hydration weakening rates are sufficiently fast. This accelerated hydration rate may be attributed to the extensive carbonatite magmatism within the eastern part of the NCC, facilitating a faster pathway for water diffusion throughout the craton. Craton destruction is contingent upon the craton’s density exceeding the surrounding mantle density, and its viscosity decreasing below 1022 Pa s. We observe that the presence of a dense lower crust or a weak mid-lithospheric discontinuity fail to destroy the NCC unless it is weakened.

A document by Mira Adhikari. Click on the document to view its contents.

Typhoons, fueled by warm sea surface waters, heighten concern as they increasingly interact with frequent Marine Heatwaves (MHWs) in a changing climate. Typhoon Hinnamnor (2022) weakenned and re-intensified as it approached the Korean Strait, interacting with underlying MHW in the northern East China Sea (nECS). Here, we found a substantial increase in latent heat loss from the nECS during the MHW period, contributing to the typhoon re-intensification from in-situ observations supplemented by reanalysis products. Strong sea surface wind forcing associated with the typhoon enhanced vertical mixing and upwelling, resulting in a pronounced (0.90°C) sea surface cooling after the typhoon passage, facilitating MHW dissipation with reduced thermal stratification. Such changes in background stratification, furthermore, significantly weakened semidiurnal internal tides due to unfavorable condition for generation from a nearby source. These findings underscore the importance of continuous time-series observations for monitoring interaction processes among the extremes in a changing climate.

Jacob M Muinde

The compositional diversity of primitive arc basalts has long inspired questions regarding the drivers of magmatism in subduction zones, including the roles of decompression melting, mantle heterogeneity, and amount and compositions of slab-derived materials. This contribution presents the volatile (H2O, Cl, and S), major, and trace element compositions of melt inclusions from basaltic magmas erupted at three volcanic centers in the Washington Cascades: Mount St. Helens (two basaltic tephras, 2.0–1.7 ka), Indian Heaven Volcanic Field (two <600 ka basaltic hyaloclastite tuffs), and Glacier Peak (late Pleistocene to Holocene basaltic tephra from Whitechuck and Indian Pass cones). Compositions corrected to be in equilibrium with mantle olivine display variability in Nb and trace element ratios indicative of mantle source variability that impressively span nearly the entire range of arc magmas globally. All volcanic centers have magmas with H2O and Cl contributions from the downgoing plate that overlap with other Cascade Arc segments. Volatile abundances and trace element ratios support a model of melting of a highly variably mantle wedge driven by a subduction component of either variably saline fluids and/or partial slab melts. Magmas from Glacier Peak have Th/Yb ratios similar to Lassen region basalts, which may be consistent with contributions of “subcreted” metasediments not found in central Oregon and southern Washington magmas that overly the Siletzia Terrane. This dataset adds to the growing inventory of primitive magma volatile concentrations and provides insight into spatial distributions of mantle heterogeneity and the role of slab components in the petrogenesis of arc magmas.