During the 2022 New Mexico monsoon season, we deployed two X-ray scintillation detectors, coupled with a 180 MHz data acquisition system to detect X-rays from natural lightning at the Langmuir Lab mountain-top facility, located at 3.3 km above mean sea level. Data acquisition was triggered by an electric field antenna calibrated to pick up lightning within a few km of the X-ray detectors. We report the energies of over 240 individual photons, ranging between 13 keV and 3.8 MeV, as registered by the LaBr3(Ce) scintillation detector. These detections were associated with four lightning flashes. Particularly, four stepped leaders and seven dart leaders produced energetic radiation. The reported photon energies allowed us to confirm that the X-ray energy distribution of natural stepped and dart leaders follows a power-law distribution with exponent ranging between 1.09 and 1.96, with stepped leaders having a harder spectrum. Characterization of the associated leaders and return strokes was done with four different electric field sensing antennas, which can measure a wide-range of time scales, from the static storm field to the fast change associated with dart leaders.

Atmospheric rivers (ARs) significantly impact the hydrological cycle and associated extremes in western continental regions. Recent studies suggest ARs also influence water resources and extremes in continental interiors. AR detection tools indicate that AR conditions are relatively frequent in areas east of the Rocky Mountains. The origin of these ARs, whether from synoptic-scale waves or mesoscale processes, is unclear. This study uses meteorological composite maps and transects of AR conditions during the four seasons. The analysis reveals that ARs east of the Rockies are associated with a long-wave baroclinic Rossby wave. This result demonstrates that eastern and midwestern ARs are dynamically similar to their western coastal counterparts, though mechanisms for vertical moisture flux differ between the two. These findings provide a foundation for understanding future climate change and ARs in this region and offer new methods for evaluating climate model simulations.

Upscaling flux tower measurements based on machine learning (ML) algorithms is an essential approach for large-scale net ecosystem CO2 exchange (NEE) estimation, but existing ML upscaling methods face some challenges, particularly in capturing NEE interannual variations (IAVs) that may relate to lagged effects. With the capacity of characterizing temporal memory effects, the Long Short-Term Memory (LSTM) networks are expected to help solve this problem. Here we explored the potential of LSTM for predicting NEE across various ecosystems using flux tower data over 82 sites in North America. The LSTM model with differentiated plant function types (PFTs) demonstrates the capability to explain 79.19% (R2 = 0.79) of the monthly variations in NEE within the testing set, with RMSE and MAE values of 0.89 and 0.57 g C m-2 d-1 respectively (r = 0.89, p < 0.001). Moreover, the LSTM model performed robustly in predicting cross-site variability, with 67.19% of the sites that can be predicted by both LSTM models with and without distinguished PFTs showing improved predictive ability. Most importantly, the IAV of predicted NEE highly correlated with that in flux observations (r = 0.81, p < 0.001), clearly outperforming that by the random forest model (r = -0.21, p = 0.011). Among all nine PFTs, solar-induced chlorophyll fluorescence, downward shortwave radiation, and leaf area index are the most important variables for explaining NEE variations, collectively accounting for approximately 54.01% in total. This study highlights the great potential of LSTM for improving carbon flux upscaling with multi-source remote sensing data.

Machine-learned uncertainty quantification (ML-UQ) has become a hot topic in environmental science, especially for neural networks.  Scientists foresee the use of ML-UQ to make better decisions and assess the trustworthiness of the ML model.  However, because ML-UQ is a new tool, its limitations are not yet fully appreciated.  For example, some types of uncertainty are fundamentally unresolvable, including uncertainty that arises from data being out of sample, i.e., outside the distribution of the training data.  While it is generally recognized that ML-based point predictions (predictions without UQ) do not extrapolate well out of sample, this awareness does not exist for ML-based uncertainty.  When point predictions have a large error, instead of accounting for this error by producing a wider confidence interval, ML-UQ often fails just as spectacularly.  We demonstrate this problem by training ML with five different UQ methods to predict shortwave radiative transfer.  The ML-UQ models are trained with real data but then tasked with generalizing to perturbed data containing, e.g., fictitious cloud and ozone layers.  We show that ML-UQ completely fails on the perturbed data, which are far outside the training distribution.  We also show that when the training data are lightly perturbed -- so that each basis vector of perturbation has a little variation in the training data -- ML-UQ can extrapolate along the basis vectors with some success, leading to much better (but still somewhat concerning) performance on the validation and testing data.  Overall, we wish to discourage overreliance on ML-UQ, especially in operational environments.

A document by Martin Velez Pardo. Click on the document to view its contents.

This study develops an explainable variational autoencoder (VAE) framework to efficiently generate high-fidelity local circulation patterns in Taiwan, ensuring an accurate representation of the physical relationship between generated local circulation and upstream synoptic flow regimes. Large ensemble semi-realistic simulations were conducted using a high-resolution (2 km) model, TaiwanVVM, where critical characteristics of various synoptic flow regimes were carefully selected to focus on the effects of local circulation variations. The VAE was constructed to capture essential representations of local circulation scenarios associated with the lee vortices by training on the ensemble dataset. The VAE’s latent space effectively captures the synoptic flow regimes as controlling factors, aligning with the physical understanding of Taiwan’s local circulation dynamics. The critical transition of flow regimes under the influence of southeasterly synoptic flow regimes is also well represented in the VAE’s latent space.This indicates that the VAE can learn the nonlinear characteristics of the multiscale interactions involving the lee vortex. The latent space within VAE can serve as a reduced-order model for predicting local circulation using synoptic wind speed and direction. This explainable VAE ensures the accurate predictions of the nonlinear characteristics of multiscale interactions between synoptic flows and the local circulation induced by topography, thereby accelerating the assessments under various climate change scenarios.

Surface, aircraft, and satellite measurements indicate pervasive cold season CO2 emissions across Arctic regions, consistent with a hyperactive biosphere and increased metabolism in plants and soils. A key remaining question is whether cold season sources will become large enough to permanently shift the Arctic into a net carbon source. Polar orbiting GHG satellites provide robust estimation of regional carbon budgets but lack sufficient spatial coverage and repeat frequency to track sink-to-source transitions in the early cold season. Mission concepts such as the Arctic Observing Mission (AOM) advocate for flying imaging spectrometers in highly elliptical orbits (HEO) over the Arctic to address sampling limitations. We perform retrieval and flux inversion simulation experiments using the AURORA mission concept, leveraging a Panchromatic imaging Fourier Transform Spectrometer (PanFTS) in HEO. AURORA simulations demonstrate the benefits of increased CO2 sampling for detecting spatial gradients in cold season efflux and improved monitoring of rapid Arctic change.

Using the Weather Research and Forecasting (WRF) model with two planetary boundary layer schemes, ACM2 and MYNN, convection-permitting model (CPM) regional climate simulations were conducted for a 6-year period at a 15-km grid spacing covering entire South America and a nested convection-permitting 3-km grid spacing covering the Peruvian central Andes region. These two CPM simulations along with a 4-km simulation covering South America produced by National Center for Atmospheric Research, three gridded global precipitation datasets, and rain gauge data in Peru and Brazil, are used to document the characteristics of precipitation and MCSs in the Peruvian central Andes region. Results show that all km-scale simulations generally capture the spatiotemporal patterns of precipitation and MCSs at both seasonal and diurnal scales, although biases exist in aspects such as precipitation intensity and MCS frequency, size, propagation speed, and associated precipitation intensity. The 3-km simulation using MYNN scheme generally outperforms the other simulations in capturing seasonal and diurnal precipitation over the mountain, while both it and the 4-km simulation demonstrate superior performance in the western Amazon Basin, based on the comparison to the gridded precipitation products and gauge data. Dynamic factors, primarily low-level jet and terrain-induced uplift, are the key drivers for precipitation and MCS genesis along the east slope of the Andes, while thermodynamic factors control the precipitation and MCS activity in the western Amazon Basin and over elevated mountainous regions. The study suggests aspects of the model needing improvement and the choice of better model configurations for future regional climate projections.

The accurate prediction of the Fire Weather Index (FWI), a multivariate climate index for wildfire risk characterization, is crucial for both wildfire management and climate-resilient planning. Moreover, consistent multisite fire danger predictions are key for targeted allocation of resources and early intervention in high-risk areas, as well as for “megafire” risk detection. In this regard, Convolutional Neural Networks (CNNs) are known to capture complex spatial patterns in climate data. This study compares different CNN architectures and traditional Statistical Downscaling (SD) methods (regression and analogs) for predicting daily FWI across diverse locations in Spain, considering marginal, distributional and spatial coherence measures for validation. Overall, the CNN-Multi-Site-Multi-Gaussian configuration, which explicitly accounts for the inter-site variability in the output layer structure, showed a superior performance. These insights provide a methodological guidance for the successful application of CNNs in the context wildfire risk assessment, enhancing wildfire response strategies and climate adaptation planning.

Aerosol increases over the 20th century delayed the rate at which Earth warmed as a result of increases in greenhouse gases (GHGs). Aggressive aerosol mitigation policies arrested aerosol radiative forcing from ~1980 to ~2010. Recent evidence supports decreases in forcing magnitude since then. Using the approximate partial radiative perturbation (APRP) method, future shortwave aerosol effective radiative forcing changes are isolated from other shortwave changes in an 18-member ensemble of ScenarioMIP projections from phase 6 of the Coupled Model Intercomparison Project (CMIP6). APRP-derived near-term (2020-2050) aerosol forcing trends are correlated with published model emulation values but are 30-50% weaker. Differences are likely explained by location shifts of aerosol-impacting emissions and their resultant influences on susceptible clouds. Despite weaker changes, implementation of aggressive aerosol cleanup policies will have a major impact on global warming rates over 2020-2050. APRP-derived aerosol radiative forcings are used together with a forcing and impulse response model to estimate global temperature trends. Strong mitigation of GHGs, as in SSP1-2.6, likely prevents warming exceeding 2C since preindustrial but the strong aerosol cleanup in this scenario increases the probability of exceeding 2C by 2050 from near zero without aerosol changes to 6% with cleanup. When the same aerosol forcing is applied to a more likely GHG forcing scenario (i.e., SSP2-4.5), aggressive aerosol cleanup more than doubles the probability of reaching 2C by 2050 from 30% to 80%. It is thus critical to quantify and simulate the impacts of changes in aerosol radiative forcing over the next few decades.

This study demonstrates the advantages of scale- and variable-dependent localization (SDL and VDL) on three-dimensional ensemble variational data assimilation of the hourly-updated high-resolution regional forecast system, the Rapid Refresh Forecast System (RRFS). SDL and VDL apply different localization radii for each spatial scale and variable, respectively, by extended control vectors. Single-observation assimilation tests and cycling experiments with RRFS indicated that SDL can enlarge the localization radius without increasing the sampling error caused by the small ensemble size and decreased associated imbalance of the analysis field, which was effective at decreasing the bias of temperature and humidity forecasts. Moreover, simultaneous assimilation of conventional and radar reflectivity data with VDL, where a smaller localization radius was applied only for hydrometeors and vertical wind, improved precipitation forecasts without introducing noisy analysis increments. Statistical verification showed that these impacts contributed to forecast error reduction, especially for low-level temperature and heavy precipitation.

Super Dual Auroral Radar Network (SuperDARN) consists of more than 30 monostatic high-frequency (HF, 10-18~MHz) radars which utilise signals scattered from decameter-scale ionospheric irregularities for studying dynamic processes in the ionosphere. By combining line-of-sight velocity measurements of ionospheric scatter echoes from radars with overlapping fields of view, SuperDARN provides maps of ionospheric plasma drift velocity over mid and high latitudes. The conventional SuperDARN radars consecutively scan through sixteen beam directions with dwelling time of 3.5 s/beam, which places a lower limit of one minute to sample the entire field of view. In this work we remove this limitation by utilizing advanced capabilities of the recently developed Borealis digital SuperDARN radar system. Combining a wide transmission beam with multiple narrow reception beams allows us to sample all conventional beam directions simultaneously and to increase the sampling rate of the entire field of view by up to sixteen times without noticeable deterioration of the data quality. The wide-beam emission also enabled the implementation of multistatic operations, where ionospheric scatter signals from one radar are received by other radars with overlapping viewing areas. These novel operations required the development of a new model to determine the geographic location of the source of the multistatic radar echoes. Our preliminary studies showed that, in comparison with the conventional monostatic operations, the multistatic operations provide a significant increase in geographic coverage, in some cases nearly doubling it. The multistatic data also provide additional velocity vector components increasing the likelihood of reconstructing full plasma drift velocity vectors.

At mesoscale, trade wind clouds organize with various spatial arrangements, shaping their effect on Earth's energy budget. Representing their fine-scale dynamics even at 1 km scale climate simulations remains challenging. However, geostationary satellites (GS) offer high-resolution cloud observation for gaining insights into trade wind cumuli from long-term records. To capture the observed organizational variability, this work proposes an integrated framework using a continuous followed by discrete self-supervised deep learning approach, which exploits cloud optical depth from GS measurements. We aim to simplify the entire mesoscale cloud spectrum by reducing the image complexity in the feature space and meaningfully partitioning it into seven classes whose connection to environmental conditions is illustrated with reanalysis data. Our framework facilitates comparing human-labeled mesoscale classes with machine-identified ones, addressing uncertainties in both methods.  We highlight the potential to explore transitions between regimes, a challenge for physical simulations, and illustrate a case study of sugar-to-flower transitions.

Temperature measurements by vertically staring ground-based Rayleigh lidars are often used to detect middle atmospheric gravity waves. In time-height diagrams of temperature perturbations, stationary mountain waves are identifiable by horizontal phase lines. Vertically tilted phase lines, on the other hand, indicate that the wave source or the propagation conditions are transient. Idealized numerical simulations illustrate that and how a wave source moving in the direction of the mean wind entails upward-tilted phase lines. The inclination angle depends on the horizontal wavelength and the wave source’s propagation speed. On this basis, the goal is to identify and characterize transient non-orographic gravity waves (NOGWs), e.g., from propagating upper-level jet/front systems, in virtual and actual Rayleigh lidar measurements. Compositions of selected atmospheric variables from a meteorological forecast or reanalysis are thoughtfully combined to associate NOGWs with processes in the troposphere and stratosphere. For a virtual observation over the Southern Ocean, upward-tilted phase lines indeed dominate the time-height diagram during the passage of an upper-level trough. The example also emphasizes that temporal filtering of temperature measurements is appropriate for NOGWs, especially in the presence of a strong polar night jet that implies large vertical wavelengths. During two selected observational periods of the COmpact Rayleigh Autonomous Lidar (CORAL) in the lee of the southern Andes, upward-tilted phase lines are mainly associated with mountain waves and transient background wind conditions. One nighttime measurement by CORAL coincides with the passage of an upper-level trough, but large-amplitude mountain waves superpose the small-amplitude NOGWs in the middle atmosphere.

Precipitation nowcasting is a crucial element in current weather service systems. Data-driven methods have proven highly advantageous, due to their flexibility in utilizing detailed initial hydrometeor observations, and their capability to approximate meteorological dynamics effectively given sufficient training data. However, current data-driven methods often encounter severe approximation/optimization errors, rendering their predictions and associated uncertainty estimates unreliable. Here we develop a probabilistic diffusion model-based precipitation nowcasting methodology, overcoming the notorious blurriness and mode collapse issues in existing practices. Our approach results in a 3.7% improvement in continuous ranked probability score compared to state-of-the-art generative adversarial model-based method. Critically, we significantly enhance the reliability of forecast uncertainty estimates, evidenced in a 68% gain of spread-skill ratio skill. As a result, our approach provides more reliable probabilistic precipitation nowcasting, showing the potential to better support weather-related decision makings.

This study investigates the origin of the zonal asymmetry in stratospheric ozone trends at northern high latitudes, identified in satellite limb observations over the past two decades. We use a merged dataset consisting of ozone profiles retrieved at the University of Bremen from SCIAMACHY and OMPS-LP measurements to derive ozone trends. We also use TOMCAT chemical transport model (CTM) simulations, forced by ERA5 reanalyses, to investigate the factors which determine the asymmetry observed in the long-term changes. By studying seasonally and longitudinally resolved observation-based ozone trends, we find, especially during spring, a well-pronounced asymmetry at polar latitudes, with values up to +6 % per decade over Greenland and -5 % per decade over western Russia. The control CTM simulation agrees well with these observed trends, whereas sensitivity simulations indicate that chemical mechanisms, involved in the production and removal of ozone, or their changes, are unlikely to explain the observed behaviour. The decomposition of TOMCAT ozone time series and of ERA5 geopotential height into the first two wavenumber components shows a clear correlation between the two variables in the middle stratosphere and demonstrates a weakening and a shift in the wavenumber-1 planetary wave activity over the past two decades. Finally, the analysis of the polar vortex position and strength points to a decadal oscillation with a reversal pattern at the beginning of the century, also found in the ozone trend asymmetry. This further stresses the link between changes in the polar vortex position and the identified ozone trend pattern.

Overshooting tops (OTs) are manifestations of deep convective updrafts that extend above the tropopause into the stratosphere. They can induce dynamic perturbations and result in irreversible transport of aerosols, water vapor and other mass from the troposphere into the stratosphere, thereby impacting the chemical composition and radiative processes of the stratosphere. These and other effects of OTs depend on their characteristics such as depth and area, which are understood to connect to mid-tropospheric updraft speed and width, respectively. Less understood is how static stability in the upper troposphere/lower stratosphere (UTLS) potentially modulates these OT–updraft connections, thus motivating the current study. Here, UTLS static stability and observed OT characteristics are quantified and compared using a combination of reanalysis data, observed rawinsonde data and geostationary satellite data. A strong relationship between OT depth and UTLS lapse rate and Brunt-Väisälä frequency N2 (R > 0.9, < -0.9, respectively) is found, implying that OT depth is reduced with an increasingly stable UTLS. In contrast, a weak relationship (R > 0.5) is found between OT area and UTLS static stability, implying that OT area is controlled primarily by mid to upper tropospheric updraft area. OT duration has a weak relationship to UTLS lapse rate and N2 (R = -0.27, 0.27, respectively). These relationships may be useful in describing mid- and low-level storm dynamics from satellite-observed characteristics of OTs in near real-time.

In past global dust storms, no long lasting anomalies in the pressure cycle had been observed. The Global Dust Storm of Mars Year 34 (MY34), however, left behind an average surface pressure lower than what was expected based on the the values recorded on previous years by the Rover Environmental Monitoring Station (REMS) on Curiosity. The main signal contribution to the daily average surface pressure is the CO2 cycle, which is controlled by the Polar ice sublimation and freezing cycles. We used REMS and Mars Climate Sounder (MCS) data to search for correlations between the REMS anomaly and anomalies in the circulation compared to MCS observations from previous years. The findings include an early start of the retreat season for the Northern Polar cap, followed by the longest period of growth for the Southern Polar (SP) cap ice expansion since Curiosity had landed and then, during the dust storm, the longest retreat season of the Southern Polar cap. We also find a larger Northern Polar Cap extension after the storm, suggestive of a larger deposition of CO2 ice. The changes in length of the SP growth and retreat seasons might be consequence of the response of the zonal mean circulation to the dust storm. Changes in the structure of the zonal mean circulation compared to previous years are found in MCS data and presented. The combination of these anomalies constraint what physical processes may have caused this response in surface pressure after the dust storm.

A document by Olivia Sablan. Click on the document to view its contents.

Ocean–atmosphere interactions largely control the variabilities of the climate system on Earth. However, how much the atmospheric internal signals contribute to climate variabilities is not yet known. Here, we develop an interactive ensemble coupled model (called Hydra-SINTEX) to investigate the influences of atmospheric internal variations (AIV) on the mean-states and variability of the climate system. The results show that, while the climatological mean-states are little affected, the AIV can largely influence the variabilities of the climate over the globe. We pay particular attention to two regions, i.e., the tropical eastern Indian Ocean, which is the key area of the Indian Ocean Dipole (IOD), and the subtropical North Pacific. We found that the variabilities of sea surface temperature (SST) in these two regions are much reduced in the absence of the AIV but with distinct mechanisms. Without the AIV, the intensity of the IOD is largely reduced in association with weakened air–sea coupling in the tropics. This indicates the importance of atmospheric noise forcing on the development of the IOD. In contrast, the reduction of the SST variability in the subtropical North Pacific, where local air–sea interaction itself is weak, is caused by the absence of the AIV that is generated by both the mid-latitude atmospheric processes and the weakened remote influence of the tropical SST in accordance with the reduced SST signals there.

Circulation patterns linked to the East Asian winter monsoon (EAWM) affect precipitation, surface temperature, and air quality extremes over East Asia. These circulation patterns can in turn be influenced by aerosol radiative and microphysical effects through diabatic heating and its impacts on atmospheric vorticity. Using global model simulations, we investigate the effects of anthropogenic aerosol emissions and concentration changes on the intensity and variability of the EAWM. Comparison with reanalysis products indicates that the model captures the mean state of the EAWM well. The experiments indicate that anthropogenic aerosol emissions strengthen the Siberian High but weaken the East Asian jet stream, making the land areas of East Asia colder, drier, and snowier. Aerosols reduce mean surface air temperatures by approximately 1.5°C, comparable to about half of the difference between strong and weak EAWM episodes in the control simulation. The mechanisms behind these changes are evaluated by analyzing differences in the potential vorticity budget. Anthropogenic aerosol effects on diabatic heating strengthen anomalous subsidence over southern East Asia, establishing an anticyclonic circulation anomaly that suppresses deep convection and precipitation. Aerosol effects on cloud cover and cloud longwave radiative heating weaken stability over the eastern flank of the Tibetan Plateau, intensifying upslope flow along the western side of the anticyclone. Both circulation anomalies contribute to reducing surface air temperatures through regional impacts on thermal advection and the atmospheric radiative balance.

Abstract content goes here

A document by Michal Ostrowski. Click on the document to view its contents.

Key Points: • CMIP6 models can capture the timing of annual cycle (particularly in February) and spatial patterns of SIT resembling the observations. • Compared to sea-ice area, CMIP6 models exhibit larger negative biases in thickness/volume, with a higher degree of variation among models. • Seasonal variations in sea-ice show positive (negative) relationships between sea ice area and thickness during September (February). Abstract This study assesses less-explored Southern Ocean sea-ice parameters, namely Sea-ice Thickness and Volume, through a comprehensive comparison of 26 CMIP6 models with reanalyses and satellite observations. Findings indicate that models replicate the mean seasonal cycle and spatial patterns of sea-ice thickness, particularly during its maxima in February. However, some models simulate implausible historical mean states compared to satellite observations, leading to large inter-model spread. September sea-ice thickness is consistently biased low across the models. Our results show a positive relationship between modeled mean sea-ice area and thickness in September (i.e., models with more area tend to have thicker ice); in February this relationship becomes negative. While CMIP6 models demonstrate proficiency in simulating Area, thickness accuracy remains a challenge. This study, therefore, highlights the need for improved representation of Antarctic sea-ice processes in models for accurate projections of thickness and volume changes. Plain Language Summary In this study, we investigated sea-ice thickness and volume in the Southern Ocean using data from 26 different climate models and observation datasets. Our findings show that the models generally capture the seasonal cycle and spatial patterns of sea-ice thickness well, with the highest average thickness occurring in February. We also found that the models tend to perform better in simulating sea-ice area compared to thickness. Furthermore, simulated sea-ice area and thickness tend to behave differently during different seasons-positively (negatively) covarying in September (February). The models that performed well in simulating sea-ice area faced challenges in accurately representing thickness and volume. This raises the question regarding the overall performance of such models or, more definitively, whether it's reliable to evaluate model accuracy or performance based solely on sea-ice area. Nevertheless, sea-ice thickness simulations in CMIP6 can offer a basis for initial analyses of absolute sea-ice changes in the Southern Ocean, despite the need for more reliable observational thickness.