Abstract. Public wind resource datasets are central to wind energy planning, particularly for distributed wind installations where it may be infeasible to collect on-site measurements or run bespoke simulations. Yet despite their broad use, the site-level accuracy at hub height remains only partially quantified. This work addresses this gap in two steps: (i) we develop a unified, observation-based benchmark to evaluate the performance of the most common models used in industry, and (ii) we propose a new machine-learned ensemble approach that leverages multiple models to synthesize improved estimates that address the shortcomings of individual models. For each dataset and observation series we form long-term empirical wind speed quantiles. This quantile representation allows us to compare products with different periods of record without requiring temporal overlap and evaluate both wind speed distribution errors and site-level mean biases. Results show that the ensemble method reduces quantile-dependent mean bias to near zero across the distribution and lowers mean absolute bias in long-term mean wind speed by roughly one-third relative to the best-performing individual dataset. Finally, we use the trained model to produce a national, gridded set of wind speed quantiles for the publicly accessible WindWatts platform. Together, the benchmark, ensemble model, and deployment dataset demonstrate that machine learning can meaningfully correct and combine existing public datasets, providing more reliable, distributional wind resource information for early-stage assessment and planning.

Distributed wind projects, particularly those involving small wind turbines, are more subject to financial and temporal limitations than utility-scale wind energy. Onsite measurements are often not feasible or economically viable investments, leading to developers, analysts, and customers in the distributed wind community relying on wind resource models to establish generation estimates. One popular wind product used by the distributed wind community in the United States is the global, high-resolution Global Wind Atlas from the Technical University of Denmark and the World Bank Group. Wind resource models are valuable tools for siting and establishing generation expectations but are not entirely accurate, which can lead to distributed wind customer dissatisfaction when actual energy generation does not meet pre-construction expectations. To enhance the understanding of the performance and limitations of utilizing Global Wind Atlas for wind resource assessment, this work presents the validation of the model wind speeds using meteorological towers across the diverse geography of the United States with measurement heights relevant to distributed wind hub heights (20 m – 100 m). The analysis expands to quantify the performance of Global Wind Atlas in representation of seasonal, diurnal, and interannual variability in the wind resource along with an assessment of wind shear accuracy at locations with measurements at multiple heights.

Abstract. This study investigates how distributed wind (DW) energy could be strategically deployed in areas with elevated energy burdens by analyzing spatial, economic, and demographic factors. We use a set of metrics that incorporate residential and macroeconomic variables, including algebraic transformations of energy burden to better capture affordability across different income levels. These metrics are correlated with demand-adjusted annual energy production, which reflects DW potential across residential, commercial, and industrial sectors. Using mixed-effects modeling and state-level fixed-effects regressions, we identify key covariates associated with high energy burden. Our results reveal significant geographic variability both across and within states, with stronger correlations between DW potential and residential energy burden in regions where burden is closely tied to poverty rates and agricultural activities. Based on these patterns, we group states into two categories and special cases reflecting correlation strength and DW potential, highlighting potential opportunities to improve energy affordability through targeted siting of distributed wind projects.

Distributed wind projects, which are connected at the distribution level of an electricity system or in off-grid applications to serve specific or local energy needs, often rely solely on wind resource models to establish wind speed and energy generation expectations. Historically, anemometer loan programs have provided an affordable avenue for more accurate onsite wind resource assessment, and the lowering cost of lidar systems has shown similar advantages for more recent assessments. While a full 12 months of onsite wind measurement is the standard for correcting model-based long-term wind speed estimates for utility-scale wind farms, the time and capital investment involved in gathering onsite measurements must be reconciled with the energy needs and funding opportunities that drive expedient deployment of distributed wind projects. Much literature exists to quantify the performance of correcting long-term wind speed estimates with 1 or more years of observational data, but few studies explore the impacts of correcting with months-long observational periods. This study aims to answer the question of how short you can go in terms of the observational time period needed to make impactful improvements to model-based long-term wind speed estimates. Three algorithms, multivariable linear regression, adaptive regression splines, and regression trees, are evaluated for their skill at correcting long-term wind resource estimates from the European Centre for Medium-Range Weather Forecasts Reanalysis version 5 (ERA5) using months-long periods of observational data from 66 locations across the US. On average, correction with even 1 month of observations provides significant improvement over the baseline ERA5 wind speed estimates and produces median bias magnitudes and relative errors within 0.22 m s−1 and 4 percentage points of the median bias magnitudes and relative errors achieved using the standard 12 months of data for correction. However, in cases when the shortest observational periods (1 to 2 months) used for correction are not well correlated with the overlapping ERA5 reference, the resultant long-term wind speed errors are worse than those produced using ERA5 without correction. Summer months, which are characterized by weaker relative wind speeds and standard deviations for most of the evaluation sites, tend to produce the worst results for long-term correction using months-long observations. The three tested algorithms perform similarly for long-term wind speed bias; however, regression trees perform notably worse than multivariable linear regression and adaptive regression splines in terms of correlation when using 6 months or less of observational data for correction. Translating the analysis to wind energy, median relative errors in the capacity factor are on average within 10 % using 1 month of training. If the observation period used for correction is not well correlated with the reference data, however, misrepresentation of the observed capacity factor can be substantial. The risk associated with poor correlation between the observed and reference datasets decreases with increasing training period length. In the worst-correlation scenarios, the median capacity factor relative errors from using 1, 3, and 6 months are within 47 %, 26 %, and 16 %, respectively.

The atmospheric dynamics that occur near the intersection of land and water offer exciting and challenging opportunities for wind energy deployment in coastal locations. New models and tools are continually being developed in support of wind resource assessment, and three recent products are explored in this work for their performance in representing characteristics of the wind resource at coastal locations: the Global Wind Atlas 3 (GWA3), the 2023 National Offshore Wind dataset (NOW-23), and the wind climate simulations that are a component of the Wind Integration National Dataset (WIND) Toolkit Long-Term Ensemble Dataset (WTK-LED Climate). These relatively new products are freely available and user-friendly so that anyone – from a utility-scale developer to a resident or business owner – can evaluate the potential for wind energy generation at their location of interest. The validations in this work provide guidance on the accuracy of wind resource assessments for coastal customers interested in installing small or midsize wind turbines (≤ 1 MW in capacity) to support energy needs at the residential, business, or community scale, such as the island and remotely located participants of the U.S. Department of Energy's Energy Transitions Initiative Partnership Project. At 23 coastal locations across the United States, dataset performance varies according to different evaluation metrics. All three recent datasets tend to overestimate the observed coastal wind resource. GWA3 produces the smallest annual average wind speed relative errors, whereas WTK-LED Climate is in best agreement in terms of representing diurnal wind speed cycles. NOW-23 is the highest performing of the datasets for representing seasonal and interannual trends in the coastal wind resource. While GWA3 and WTK-LED Climate are relatively insensitive to the dataset output heights selected for wind resource assessment at small and midsize wind turbine hub heights (20–60 m), significant variation in the NOW-23 representation of wind shear across the wind profile in the lowest 100 m of the atmosphere leads to notable differences in wind speed estimates according to the dataset output heights selected for evaluation. GWA3 exhibits challenges in the representation of observed wind speed diurnal cycles at small and midsize turbine hub heights, likely due to the dataset's consistent treatment of hourly wind speed trends regardless of altitude.

BACKGROUND:Acute mountain sickness (AMS) is a debilitating condition that may occur on ascent to high altitude, with limited options for chemoprophylaxis. The pathophysiology of AMS is poorly understood, though it may be similar to migraine. This study aimed to determine the efficacy of prochlorperazine, a first-line agent for acute migraine, for AMS prophylaxis. METHODS:We performed a randomized, double-blind, placebo-controlled trial involving healthy, unacclimatized adult participants, primarily from the Denver area (1609 m), who received either oral prochlorperazine or placebo three times daily for 24 hours during rapid ascent to Mount Blue Sky, Colorado (4348 m). We evaluated individuals who received at least the first dose of placebo or intervention following a modified intent-to-treat approach. Participants travelled by vehicle to 3910 m, then hiked to the summit of Mount Blue Sky, where they slept overnight. The primary outcome was AMS incidence as defined by the 2018 Lake Louise Questionnaire, which was assessed on the evening of ascent and the following morning. RESULTS:We analysed 56 participants (25 women), with a mean age of 39 [IQR 28-49], with 28 participants in each study arm. Key baseline characteristics were equally distributed and well-matched between study arms. The incidence of AMS was 28 (50%), with 18 (64%) in the placebo arm and 10 (36%) in the prochlorperazine arm (P = 0.06). The absolute risk reduction was 28.6%, the number needed to treat was 4 and the odds ratio was 0.28 (95% C.I. 0.11-0.94). There were no serious adverse events, and there were no significant differences in the side effects between arms, including for drowsiness (P = 0.47). CONCLUSION:Our results suggest that prochlorperazine is effective in preventing AMS. Larger studies are warranted to validate our findings.

Acute mountain sickness (AMS) is a debilitating condition that individuals may develop on ascent to high altitude. It is characterized by headache, nausea, vomiting, dizziness, and fatigue with the potential to progress to fatal disease. Although the pathophysiology of AMS remains unclear, proposed mechanisms are hypothesized to be similar to migraine. Prochlorperazine, a first-line treatment for acute migraine, has been shown to abort migraine early and thus may be effective in preventing AMS. Its action as a respiratory stimulant additionally makes it a promising novel agent for AMS prevention. In this randomized double-blinded trial, participants will be randomized to receive oral prochlorperazine maleate or placebo for 24 h of three times daily dosing on a rapid ascent to 4348 m. Participants will be adults, aged 18, and older who are unacclimatized. Participants will remain at this elevation overnight. The Lake Louise Questionnaire will be utilized to define the primary outcome and presence of AMS and will be assessed the evening of and morning after ascent to peak altitude. Currently, acetazolamide is the preferred option for the chemoprophylaxis of AMS, which has been studied and utilized since the 1970s and involves potential prohibitive side effects. Other more efficacious options with more tolerable side effects are needed. Preventing AMS has the potential to limit both the morbidity and mortality associated with developing AMS and more serious diseases (notably high-altitude cerebral edema). Additionally, there is a substantial economic and environmental impact of AMS that could be prevented. Clinicaltrial.gov, NCT06450899. Registered on June 2024.

Gardner, Laurel, Linda E. Keyes, Caleb Phillips, Elan Small, Tejaswi Adhikari, Nathan Barott, Ken Zafren, Rony Maharjan, and James Marvel. Women at altitude: Menstrual-cycle phase, menopause, and exogenous progesterone are not associated with acute mountain sickness. High Alt Med Biol. 00:000-000, 2024. Background: Elevated progesterone levels in women may protect against acute mountain sickness (AMS). The impact of hormonal contraception (HC) on AMS is unknown. We examined the effect of natural and exogenous progesterone on the occurrence of AMS. Methods: We conducted a prospective observational convenience study of female trekkers in Lobuche (4,940 m) and Manang (3,519 m). We collected data on last menstrual period, use of exogenous hormones, and development of AMS. Results: There were 1,161 trekkers who met inclusion criteria, of whom 307 (26%) had AMS. There was no significant difference in occurrence of AMS between women in the follicular (28%) and the luteal (25%) phases of menstruation (p = 0.48). The proportion of premenopausal (25%) versus postmenopausal women (30%) with AMS did not differ (p = 0.33). The use of HC did not influence the occurrence of AMS (HC 23% vs. no HC 26%, p = 0.47), nor did hormonal replacement therapy (HRT) (HRT 11% vs. no HRT 31%, p = 0.13). Conclusion: We found no relationship between menstrual-cycle phase, menopausal status, or use of exogenous progesterone and the occurrence of AMS in trekkers and conclude that hormonal status is not a risk factor for AMS. Furthermore, women should not be excluded from future AMS studies based on hormonal status.

Current wind resources within the United States (US) indicate a potential to profitably install nearly 1,400 gigawatts of distributed wind (DW) capacity. This amount is equivalent to over half of the United States’ current energy demand from electricity, making it enough to power millions of homes and businesses and replace countless fossil fuel-based generating plants. Despite the potential growth of DW in the US, deployments are presently hindered by a lack of confidence in resource estimation methods. One potential challenge is that smaller-scale turbines, with hub heights of 40 meters or less, are disproportionately impacted by obstacles such as buildings and vegetation. These obstacles may produce complex wake effects, best modeled with high-fidelity complex fluid dynamics (CFD) models that are too computationally expensive to use for routine siting and resource assessment. Thus, installers today make use of heuristics and simple equations to approximate the impact of obstacles while also leveraging long-term resource data from commercial or publicly available atmospheric models. This study evaluates these historical and commonly used methods alongside new lower-order obstacle models produced from CFD simulations and measurement-based bias correction. The preliminary results from this study show the importance of taking care in the choice and application of mesoscale atmospheric models and the significant value of bias correction using measurements from nearby meteorological towers. Detailed obstacle modeling provides only modest additional gains in performance and, in some cases, can add error, especially at sites where turbines have already been located to avoid obvious impact from upwind obstacles. These findings reinforce the importance of collecting in situ measurements and suggest that obstacle models may be better applied in practice to automated or computer-aided siting, rather than in economic wind resource assessments.

Advances in deep learning have enabled accurate language-based search and retrieval, e.g., over user photos, in the cloud. Many users prefer to store their photos in the home due to privacy concerns. As such, a need arises for models that can perform cross-modal search on resource-limited devices. State-of-the-art cross-modal retrieval models achieve high accuracy through learning entangled representations that enable fine-grained similarity calculation between a language query and an image, but at the expense of having a prohibitively high retrieval latency. Alternatively, there is a new class of methods that exhibits good performance with low latency, but requires a lot more computational resources, and an order of magnitude more training data (i.e. large web-scraped datasets consisting of millions of image-caption pairs) making them infeasible to use in a commercial context. From a pragmatic perspective, none of the existing methods are suitable for developing commercial applications for low-latency cross-modal retrieval on low-resource devices. We propose CrispSearch, a cascaded approach that greatly reduces the retrieval latency with minimal loss in ranking accuracy for on-device language-based image retrieval. The idea behind our approach is to combine a light-weight and runtime-efficient coarse model with a fine re-ranking stage. Given a language query, the coarse model effectively filters out many of the irrelevant image candidates. After this filtering, only a handful of strong candidates will be selected and sent to a fine model for re-ranking. Extensive experimental results with two SOTA models for the fine re-ranking stage, on standard benchmark datasets show that CrispSearch results in a speedup of up to 38 times over the SOTA fine methods with negligible performance degradation. Moreover, our method does not require millions of training instances, making it a pragmatic solution to on-device search and retrieval.

Each year, millions of people travel to high altitude regions and are at risk for high altitude illness. Advances in transportation to these locations, including the construction of direct roadways and increased popularity of mountain sports, have contributed to the growing number of individuals at risk. Much of our current knowledge of high altitude illness is from studies conducted in the Himalayas, specifically in the Khumbu and Annapurna regions, and altitude mountain sickness (AMS) rates vary from as low as 16% to approaching 50%. We present the largest contemporary cross-sectional survey study involving trekkers in 2 trekking regions of the Nepal Himalaya.

Small, Elan, Caleb Phillips, William Bunzel, Lakota Cleaver, Nishant Joshi, Laurel Gardner, Rony Maharjan, and James Marvel. Prior ambulatory mild coronavirus disease 2019 does not increase risk of acute mountain sickness. High Alt Med Biol. 24:201-208, 2023. Background: Given its long-term morbidity, understanding how prior coronavirus disease 2019 (COVID-19) may affect acute mountain sickness (AMS) susceptibility is important for preascent risk stratification. The objective of this study was to examine if prior COVID-19 impacts risk of AMS. Materials and Methods: This was a prospective observational study conducted in Lobuje (4,940 m) and Manang (3,519 m), Nepal, from April to May 2022. AMS was defined by the 2018 Lake Louise Questionnaire criteria. COVID-19 severity was defined using the World Health Organization-developed criteria. Results: In the Lobuje cohort of 2,027, 46.2% of surveyed individuals reported history of COVID-19, with 25.7% AMS point-prevalence. There was no significant relationship between prior ambulatory mild COVID-19 and AMS (p = 0.6) or moderate AMS (p = 1.0). In the Manang cohort of 908, 42.8% reported history of COVID-19, with 14.7% AMS point-prevalence. There was no significant relationship between prior ambulatory mild COVID-19 and AMS (p = 0.3) or moderate AMS (p = 0.4). Average months since COVID-19 was 7.4 (interquartile range [IQR] 3-10) for Lobuje, 6.2 (IQR 3-6) for Manang. Both cohorts rarely exhibited moderate COVID-19 history. Conclusions: Prior ambulatory mild COVID-19 was not associated with increased risk of AMS and should not preclude high-altitude travel.

Many airports are adopting battery electric buses in their shuttle fleets due to concerns over air quality and regulations. This study proposes a simulation-based optimization modeling framework to help airport shuttle operators effectively deploy electric buses. We evaluated a planned airport electric shuttle system with an event-driven simulator. Empirical data collected from existing systems were used to drive the simulations. We then proposed a simulation-based optimization model to determine the battery capacity, charging power, and number of chargers so that predefined objective(s) (e.g., minimizing total capital cost, minimizing emissions) are optimized. Compared to existing studies, the primary contribution of the proposed method is that it can model the real-world stochastic nature of operations in an electric bus system with much higher fidelity. To demonstrate the proposed modeling framework, we study a real-world shuttle system at the Dallas-Fort Worth International Airport, and present extensive numerical studies. When considering partial fleet electrification, the model can provide a set of Pareto optimal solutions. When considering full fleet electrification, the optimal solution requires a 50-kWh battery capacity and four 210-kW chargers, resulting in a total capital cost of $26,744,000. The results demonstrate that the proposed modeling framework can effectively optimize the planning of electric airport shuttle systems with partial or full fleet electrification.

Most of the existing traffic simulation packages require significant calibration work to be able to reflect reality. To evaluate special operations including emerging technologies, a microscopic simulation that tracks detailed interactions of all the elements of the traffic systems is usually needed. This type of simulation is usually computationally demanding. This work developed an Airport Shuttle Planning and Improved Routing Event-driven Simulation (ASPIRES) package to simulate and evaluate current, potential, and future airport shuttle operations. The simulation was driven by data and thus did not require much calibration effort. The discrete-event simulation nature of ASPIRES makes the simulation computationally efficient. Simulating 1 day of shuttle operations takes less than 2 s. The study site of this work is the Dallas/Fort Worth International Airport in the U.S. The shuttle service that connects the five terminals of the airport and the rental car center was studied. Travel times, dwell times, and passenger arrivals were simulated using empirical distributions derived mainly from real data to capture the stochastic nature of the rental car center shuttle bus operations. Data on bus miles traveled, bus energy consumption, passenger wait times, and passengers left behind at stops were collected to study the trade-off between energy use and passenger experience. Electric bus and on-demand bus operations were also included. The simulation outputs can show passengers statistics at terminals, shuttles statistics, and charging station statistics. ASPIRES cannot be used to model a generic traffic system but is well-suited for fleet systems.

Background:The COVID-19 pandemic saw governmentimposed health mandates that contributed to profound work, exercise, and lifestyle changes.There was an observed increase in weight gain and sedentary behavior during the pandemic in United States, as well as increased outdoor exercise and running in the United States.While road race times have consistently slowed over the last few decades, it is unknown how running performance changed during the pandemic.Methods: This was a retrospective analysis of finishers of the world's largest 10 km road race in 2019 (pre-pandemic) and 2021 (pandemic).Times were gathered using implicit JavaScript Object Notation API and novel web-scraping and post-processing.Participants' resident state and pace were correlated with open sourced publicly available durations of shelter-in-place, mask mandates, and gym closures in addition to COVID-19 hospitalization incidence and per capita mortality from March 2020 through March 2021.Runners' pace and participation differences were correlated with per state mitigation stipulations and percapita COVID-19 hospitalizations and mortality from March 2020 to March 2021, controlling for age and sex.All data was open access.Results: There were 60,525 pre-pandemic and 24,265 pandemic runners.The pace average was 1.4 min / mile faster (95% confidence interval [C] 1.34 to 1.45, p < 0.001) during the pandemic (12 versus 13.4 min / mile).Finishing times decreased during the pandemic in both women (12.9 vs. 14.3 min / mile, 95% CI 1.31 to 1.47, p < 0.001) and men (11.3 vs. 12.5 min / mile, 95% CI 1.2 to 1.3, p < 0.001).These performance trends were consistent in the 10,475 repeat runners (p < 0.001).These performance trends were consistent in the 10,475 repeat runners (p < 0.001) in both women (-0.4 min / mile, 95% CI -0.4 to -0.3, p < 0.01) and men (-0.3 min / mile, 95% CI -0.4 to -0.3, p < 0.01), with greatest improvement in those younger than 20 years old (-1 min / mile, 95% CI -1.2 to -0.7, p < 0.01).Mitigation had negligible impact on performance when weighted by resident state and duration of mask mandates (r = 0.17, 95% CI -0.1 to 0.5), shelter-in-place (r = 0.24, 95% CI -0.1 to 0.5), and gym closures (r = 0.14, 95% CI -0.2 to 0.4).There was no correlation between performance and COVID-19 hospitalization incidence (r = < 0.01, 95% CI -0.4 to 0.4) or per capita mortality (r < 0.01, 95% CI -0.3 to 0.3). Conclusion:Pandemic-era 10 km runners saw improved performance over pre-pandemic times, consistent across age groups and sex, with the biggest benefit in the youngest people.COVID-19 mitigation restrictions and disease severity did not affect performance, suggesting unmeasured lifestyle changes may have contributed to improved fitness.

The paper addresses in detail one of the most important barriers for small wind development, firstly the common lack of observational data and secondly the lack of accurate, cost-competitive, and friendly wind resource assessment tools for small wind applications. The paper contains a comparative analysis of a comprehensive number of methods for wind resource assessment of sites with obstacles based on different data sources. All the methods analyzed are quite well explained but they are obviously very sensitive to the quality of obstacles description and input data.

Abstract. Growth in adoption of distributed wind turbines for energy generation is significantly impacted by challenges associated with siting and accurate estimation of the wind resource. Small turbines, at hub heights of 40 m or less, are greatly impacted by terrestrial obstacles such as built structures and vegetation that can cause complex wake effects. While some progress in high-fidelity complex fluid dynamics (CFD) models has increased the potential accuracy for modelling the impacts of obstacles on turbulent wind flow, these models are too computationally expensive for practical siting and resource assessment applications. To understand the efficacy of available models in situ, this study evaluates classical and commonly used methods alongside new state-of-the-art lower-order models derived from CFD simulations and machine learning approaches. The evaluation is conducted using a subset of an extensive original dataset of measurements from more than 300 operational wind turbines in the northern Netherlands. We find that data driven methods (e.g., machine learning and statistical modelling) are most effective at predicting production at real sites with average error in annual energy production of 2.5 %. When sufficient data may not be available de novo to support these data-driven approaches, models derived from high fidelity simulations show promise and reliably outperform classical methods. On average these models have 6.3–11.5 % error compared to 26 % for classical methods and 27 % baseline error for reanalysis data without obstacle correction. While more performant on average, these methods are also sensitive to the quality of obstacle descriptions and reanalysis inputs.