PSL Publication Report

Abstract

Saturation vapor pressure deficit (VPD), a measure of the difference between how much moisture the atmosphere can hold versus how much is present, is highly correlated with the annual mean area burned by wildland fires in the western United States. The present analysis uses linear inverse models (LIMs) to forecast seasonal VPD and decompose skill into contributions from a nonlinear trend, coupled sea surface temperature (SST)-VPD variability, and VPD-only variability. Subregions of the western US are considered using Geographic Area Coordination Centers which have different times of year and lead times for which VPD forecast skill is greatest. However, the sources of skill are similar among the subregions. In LIM forecasts, particularly those made for summer and early fall, the trend contributes to VPD skill up to 18 months in advance, with a secondary contribution from internal VPD variability at lead times of one to two months. Positive SST and VPD anomalies and negative soil moisture anomalies are associated with the positive sign of the trend time series, which has been observed without interruption since the late 1990s. Coupled SST-VPD variability contributes to VPD skill mainly for forecasts verifying between December to May for lead times up to 12 months in some subregions. Forecasts that are especially skillful and display high confidence, seasonal forecasts of opportunity (SFOs), are associated with SSTs that produce high VPD skill over California, the Southwest, and Texas, while internal VPD anomalies contribute to skill over the Great Basin and western Northern Plains. SFOs are initialized during periods of El Niño-Southern Oscillation development, with La Niña SSTs associated with positive western US VPD anomalies and consequently, enhanced wildland fire risk.

Abstract

Continuous measurements of carbon dioxide (CO₂) flux were collected from a 10 m eddy covariance tower in a coastal-marine environment in the Canadian Arctic Archipelago over the course of a 17-month period. The extended length of data collection resulted in a unique dataset that includes measurements from two spring melt and summer seasons and one autumn freeze-up. These field observations were used to verify findings from previous theoretical and laboratory experiments investigating air-sea gas exchange in connection with sea ice. The results corroborated previous findings showing that thick ice cover under winter conditions acts as a barrier to gas exchange. In the spring, CO₂ fluxes were downward (uptake) in both the presence of melt ponds and during ice break-up. However, diurnal cycles were present throughout the early spring melt period, corresponding to the opposing influences of freezing and melting at the ice surface. Fluxes measured during melt periods confirmed previous laboratory tank measurements that showed a gas transfer coefficient of melting ice of 0.4 mol m^-2 d^-1 atm^-1. Open water CO2 fluxes showed outgassing in early summer and uptake in mid-to-late summer, tied closely to trends in surface water temperature and its effect on the partial pressure of CO₂ in the water. The autumn period of the field campaign represents the first eddy covariance CO₂ fluxes measured over naturally forming sea ice. Our measurements showed mean upward fluxes (outgassing) of 1.1±1.5 mmol m^-2 d^-1 associated with the freezing of ice – the same order of magnitude found by previous laboratory tank experiments. However, peak flux periods during ice formation had measured fluxes that were a factor of 3 higher than the tank experiments, suggesting the importance of natural conditions (e.g., wind) on air-ice gas exchange. Conducting an Arctic-wide extrapolation we estimate CO₂ outgassing from the freezing period to be a counterbalance equivalent to 5 to 15 % of the magnitude of the estimated Arctic CO₂ sink. Overall, there was no evidence of dramatically enhanced gas exchange in marginal ice conditions as proposed by previous studies. Although the different seasons showed active CO₂ exchange, there was a balance between upward and downward fluxes at this specific location, resulting in a small net CO₂ uptake over the annual cycle of −0.3 g C m⁻².

Abstract

In the past decades, the California Current Ecosystem has experienced intense marine heatwaves, which have induced significant disruptions to local phytoplankton communities. Here, using 30-year cruise observations, we identify a previously undocumented vertical structure in chlorophyll-a concentration response to marine heatwaves, characterized by reductions in the surface layer coupled with increases in the subsurface. By integrating observations and coupled physical-biogeochemical model products, we demonstrate that declines of surface chlorophyll-a are primarily attributed to suppressed nutrient upwelled to the upper ocean. Although surface irradiance increased modestly (+3.5%), light availability in the subsurface layer improved substantially (+21.7%) due to reduced phytoplankton shading at the surface. Concurrent with enhanced lateral nutrient transport, phytoplankton growth at depth was promoted during heatwave events. This study highlights the pivotal role of subsurface phytoplankton dynamics in shaping the vertical chlorophyll-a concentration structure and its variability under extreme events.

Abstract

This paper evaluates seasonal forecasts of weather types (WTs), i.e., recurring large-scale atmospheric patterns, which have been developed using a clustering method using large-scale predictors to represent precipitation variability across the United States. Forecast quality is assessed using two seasonal hindcast products: from the operational weather community, hindcasts from the European Centre for Medium-Range Weather Forecasts (ECMWF), as well as hindcasts from the Community Earth System Model, version 2 (CESM2), a community resource developed by the National Science Foundation (NSF) National Center for Atmospheric Research, a climate research center. WTs are described in terms of their associated precipitation anomalies and examined in light of their relationship with established climate teleconnections. The spatial precipitation patterns associated with each WT are less well captured in the forecasting systems than the large-scale variables from which the WTs are derived. The WT patterns themselves are well represented in both seasonal forecasting systems, though, on average, ECMWF is slightly closer to observations. Forecasted WT frequency results show that both prediction systems have similar skill, with most differences depending on season and WT. Winter WT frequencies are generally more predictable than summer. Both forecast systems capture well the frequency rank order but underestimate the interannual frequency spread, which could be partially due to ensemble averaging. Analysis shows that forecasting of climate teleconnection indices alone would not be sufficient to represent the precipitation variability associated with the WTs. Comparable results from two initialized Earth system prediction models that originate from different sides of the weather–climate and operations–research spectrum are encouraging and contribute to WT forecasting and multimodel initialized prediction efforts.

Significance Statement

The purpose of this study is to evaluate how well seasonal forecasts capture large-scale weather patterns that are associated with precipitation variability across the United States. Our results show that large-scale patterns are better captured than precipitation patterns. Given the societal importance of precipitation, forecasting based on associated weather types could complement existing seasonal prediction systems.

Abstract

Profiles of thermodynamic and cloud properties and their transformations during Arctic Warm Air Intrusions (WAIs) and Cold Air Outbreaks (CAOs) were observed during an aircraft campaign, and simulated using the ICON weather prediction model. The data were collected along flight patterns aimed at sampling the same air parcels multiple times, enabling Eulerian and quasi-Lagrangian measurement-model comparisons and model process studies. Within the Eulerian framework, the temperature profiles agreed well with the ICON output although a small model bias of −0.9 K was detected over sea ice during CAOs. Also, the air parcels did not adjust to the changing surface skin temperature quickly enough. The specific humidity profiles were reproduced by ICON with mean deviations of 6.0 % and 19.5 % for WAIs and CAOs, respectively. Radar reflectivities based on ICON output captured the vertical cloud distributions during the airmass transformations. The simulated process rates of temperature and humidity along the trajectories showed that adiabatic processes dominated the heating and cooling of the air parcels over diabatic effects during both WAIs and CAOs. Of the diabatic processes, latent heating and turbulence had a stronger impact on the temperature process rates than terrestrial radiative effects, especially over the warm ocean surface during CAOs. Finally, a quasi-Lagrangian observation-model comparison was performed. For WAIs, the observed change rates of temperature and humidity were not perfectly captured in the simulations. For the CAOs, the calculated heating and moistening change rates of the airmasses were well represented by ICON with remaining deviations close to the surface.

Abstract

Previous studies of air-sea interactions over sharp oceanic fronts have suggested that it is the ocean that drives the atmosphere across sub-mesoscale ocean fronts, but it is the atmosphere that drives the ocean at synoptic scales; the responsible mechanism, however, is still a matter of debate. This paper examines direct sea surface temperature (SST) measurements of the skin (SSTskin) and near-surface SST (SSTdepth), and wind speeds measured during the Sub-Mesoscale Ocean Dynamics Experiment (S-MODE) along with derived bulk fluxes. We evaluate the modulation of the net heat flux, wind speed, and skin cooling across SST fronts and the ability of the COARE bulk flux algorithm to reproduce this variability. Bulk flux computations can be performed directly from a radiometric SSTskin, or more commonly, from the SSTdepth provided that the depth of the SST measurement is corrected for cool skin and diurnal warming effects. Both types of SST were measured during S-MODE allowing for (a) an assessment of the importance of having a SSTskin for a direct flux evaluation in frontal regions, and (b) an evaluation of the accuracy of the cool skin and diurnal warming corrections within COARE for the indirect bulk flux computation. The ocean-atmosphere feedback over the sampled S-MODE submesoscale front suggested that the ocean was indeed forcing the atmosphere, mainly through the surface net heat losses, while the wind response to changes in SSTskin was irregular. Testing of the COARE algorithm suggested that indirect bulk fluxes had sufficient accuracy to close the heat budget over the front.

Plain Language Summary

While previous studies of air-sea interactions over oceanic fronts have suggested that it is the ocean that drives the atmosphere across sub-mesoscale ocean fronts, but the atmosphere that drives the ocean at basin scales, the ultimate mechanism is not absolutely certain. In this paper, we look at the response of the heat flux, wind speed, and cooling at the ocean surface to changes in sea surface temperature across a strong thermal front. The observed feedback suggests that it is the heat flux from the ocean to the atmosphere that dominates the air-sea feedback over the sub-mesoscale front. We then use one of the most popular models to compute heat fluxes over the ocean, the so-called COARE bulk flux algorithm, to evaluate the skill of the model in estimating fluxes over the front. Because the ocean temperature changes with depth, the COARE model can be run using radiometric temperatures, which are more direct but challenging/expensive or deeper temperatures that are easier/cheaper to measure but require additional corrections to account for the shape of the ocean surface temperature profile and can introduce errors. Results indicate that the simpler measurements are good enough to evaluate the fluxes over the front.

Abstract

The Year of Polar Prediction (YOPP), an international research initiative organized by the World Meteorological Organization’s (WMO) World Weather Research Program from 2013 to 2022, aimed to markedly enhance environmental prediction capabilities in the polar regions and beyond, particularly in the context of a rapidly changing climate. YOPP achieved this through a concerted effort in observation, modeling, verification, user engagement, and educational activities. This article offers a comprehensive overview of YOPP’s key outcomes and impacts, using a dual approach that merges qualitative success stories with quantitative metrics. Scientifically, the focus is on the role of polar observations in improving prediction accuracy, enhanced understanding of processes to support model development, advancements in forecast verification, particularly in sea ice prediction, an improved understanding of the interconnections between polar and midlatitude regions, and effective user engagement. This paper also discusses how these scientific discoveries have been converted into practical applications, emphasizing the route from science to services. Additionally, it summarizes the education, communication, outreach, and coordination efforts employed to maximize YOPP’s impact. Finally, the article provides a series of recommendations for future research, informed by the insights gained from YOPP’s experiences and recent radical developments in technology.

Significance Statement

The Year of Polar Prediction (YOPP) was a landmark initiative aimed at enhancing our ability to predict environmental changes in the polar regions, areas that are increasingly affected by climate change. By integrating global efforts in observation, modeling, and data analysis, YOPP has significantly contributed to improve the accuracy of weather and climate forecasts in these critical zones, and beyond. These advancements matter because they provide crucial insights into polar processes along with their remote impacts, enhance global prediction models, and inform stakeholders about predictive capabilities. The project’s focus on user engagement and education ensures that these scientific achievements translate into practical benefits. The collaborative spirit of YOPP exemplifies how international scientific cooperation can address some of the most pressing environmental challenges of our time.

Abstract

Linear Inverse Models (LIMs) are widely used data-driven tools for studying El Niño Southern Oscillation (ENSO). However, standard LIMs struggle to simulate the observed asymmetry and diversity of ENSO events. Observations reveal that strong Central Pacific (CP) La Niñas and extreme Eastern Pacific (EP) El Niños occur more frequently than their counterparts, a feature standard LIMs fail to capture. We introduce a modified model, the Non-Gaussian LIM (NG-LIM), which transforms the LIM variables to better simulate ENSO asymmetry and diversity. Specifically, the NG-LIM reproduces the spatial pattern of sea surface temperature (SST) skewness and the inverted U-shaped relationship between the first two principal components of Tropical Pacific SST anomalies, reflecting more frequent strong CP La Niñas and extreme EP El Niños. NG-LIM simulations also show El Niños that are stronger and evolve more rapidly than La Niñas. This improved inverse model generates synthetic events to supplement the limited observational record.

Plain Language Summary

El Niño and La Niña are dominant patterns of climate variability that can have wide-reaching impacts on weather and ecosystems worldwide. Scientists often use mathematical models to study these events, including Linear Inverse Models (LIMs), which analyze past data to make predictions. However, standard LIMs struggle to capture certain asymmetric features of El Niño and La Niña events, like their uneven strength and their spatial footprint. For instance, intense El Niños tend to develop quickly and decay rapidly, while La Niñas often linger longer but are not as extreme. In this study, we introduce a modified model, the Non-Gaussian LIM (NG-LIM), which better represents these asymmetries between El Niño and La Niña. This modified model generates a broader range of synthetic events, providing a valuable tool for understanding these climate patterns.

Abstract

Much of the predictability of seasonal climate anomalies around the globe is due to the predictability of tropical El Niño–Southern Oscillation (ENSO) events and their global teleconnections. Despite decades of research, however, the relative roles of the dominant positive and negative feedbacks on predictable ENSO event magnitudes, patterns, and durations remain unclear. Here, an attempt is made to estimate these feedbacks directly from observational data. A 15-component linear inverse model (LIM) of the coupled Indo-Pacific atmosphere–ocean climate system is constructed for this purpose using reanalysis data for 1979–2017, and its predictable dynamics are clarified through a series of feedback-denial experiments. This approach yields, for the first time from data, a clear picture of the dominant competition between the destabilization of ENSO by positive near-surface zonal wind and subsurface oceanic feedbacks and its stabilization by a negative surface shortwave flux feedback associated with cloud shielding in cloudy areas. The results suggest that it is primarily this negative feedback that renders ENSO asymptotically stable. They also suggest that its underrepresentation is likely behind the tendency of climate models to extend the equatorial Pacific warming during El Niño (and cooling during La Niña) too far west into the western Pacific, compromising seasonal and longer-term predictions around the globe. An underrepresentation of this negative feedback over the Maritime Continent is also consistent with the mean sea surface temperature (SST) cold tongue and easterly trade wind biases of many climate models over the western equatorial Pacific.

Abstract

Most operational weather prediction centers cycle their global data assimilation systems at a cadence of two to four times per day for global weather prediction. Increases in computing power, observation availability, and horizontal model resolution motivate the development of a more rapidly updated global assimilation system. Experiments using hybrid four-dimensional ensemble-variational data assimilation (4DEnVar) with 1-h assimilation windows are performed in a version of the NOAA Global Data Assimilation System, and results are compared to those from traditional 6-hourly cycling. Skill is evaluated via comparisons with in situ observations, satellite winds, and regional analyses. Ensemble mean forecasts initialized from the hourly cycled fields show significantly better skill for some variables and regions than those initialized from traditional 6-hourly cycled fields, though they are informed by the same observations. These benefits appear only when dense, accurate observations are assimilated (e.g., aircraft wind). When aircraft data are withheld from assimilation, hourly cycling degrades the fit to in situ observations. Investigations into kinetic energy spectra demonstrate that fields from the hourly cycling system retain more energy at synoptic and subsynoptic scales than fields from the 6-hourly cycling system. These results suggest that a global hourly cycling assimilation system can use temporally and spatially dense observations more effectively than a 6-hourly cycling system.

Significance Statement

Global weather forecasts in the United States are currently initialized by assimilating in situ and satellite observations every 6 h. This work aims to determine the impact of shortening this assimilation window to one hour and increasing the frequency of updates. Results suggest that the shortened 1-h update cadence can improve the performance of the data assimilation by using dense observations, particularly of wind from aircraft and satellites, more effectively. While computational constraints remain a challenge, this suggests that modifying the operational data assimilation cadence could provide better global and regional weather forecasts.