Hydrosolutions (Switzerland)

Abstract The American Meteorological Society’s Committee on Ecological Forecasting, established in 2014, addresses the growing necessity for interdisciplinary collaboration between ecological and physical sciences to enhance forecasting capabilities amid escalating environmental challenges. Weather and climate events (e.g., hurricanes, floods, wildfires) produce significant ecological disturbances that disrupt ecosystems, ultimately impacting human populations and other living organisms. This article expands upon the Statement on The Future of Ecological Forecasting adopted by the American Meteorological Society (AMS) in January 2024 by providing essential context, references, and motivation. It underscores the integral relationship between human well-being and ecosystem services, which rely on stable ecological functions. By elucidating the potential of ecological forecasts (“ecoforecasts”) to predict significant changes in ecosystems that affect human populations, economies, and health, this article advocates for a coordinated approach among researchers, policymakers, and practitioners. Enhanced operational systems for ecological forecasting promise increased lead times and analytics to facilitate proactive planning and funding for shared use and mitigation efforts. The transdisciplinary nature of ecological forecasting aims to unify insights across various scientific domains, ultimately leading to improved decision-making processes that support sustainable adaptation to rapid environmental change. Significance Statement The AMS Statement on The Future of Ecological Forecasting (2024) marks a key milestone in linking meteorological and ecological forecasting, offering a framework for interdisciplinary collaboration among AMS, professional societies, agencies, stakeholders, and policy groups. This paper aims to preserve the Statement as a lasting reference, augment it with recent scientific and technical developments, and promote its use as a widely accessible planning tool to reinforce strategic advancements in ecological forecasting.

At Eurocrypt’25, Belkheyar et al. introduced a new non-linear layer called ChiChi or χχ. ChiChi is built based on Daemen’s χ function, but crucially gives a permutation in even dimension divisible by four.In this work, we generalize their construction in multiple ways, prove their open conjecture regarding the algebraic degree of the inverse of ChiChi, and investigate the properties of it against key recovery attacks.

Accurate and rapid flood inundation mapping is essential for early warning systems, risk assessment, and resource allocation. Flood mapping models vary in complexity—from fully hydrodynamic solvers (e.g., TRITON) to simpler, terrain‐based methods (e.g

Pharmacoepidemiologic studies, which promote rational drug use and improve health outcomes, often require Anatomical Therapeutic Chemical Classification System (ATC) drug classification within real-world data (RWD) sources. Existing classification tools are expensive, brittle, or have restrictive terms of service, and lack context that may inform classification itself. This study sought to establish large language models (LLMs) as an assisting technology in the drug classification task. This included developing artificial intelligence prompts that reason about drugs using RWD and showing that the resulting accuracy, efficiency, and effectiveness are favorable to alternative methods. A prompt was constructed to classify aspirin as either an analgesic or antithrombotic and evaluated within 12,294 anonymized daily dose strings from a polychronic population residing in the United States and Canada. The patients used a smart medication dispenser called "spencer" and consented to the use of their data for research. The LLM prompt requested that the best and next-best second-level ATC code be returned, and grading was performed on a 3-point scale. After success in a pilot sample of 20, an inference sample of 200 was taken without replacement. Finite population inference was carried out on the proportion of outputs receiving 1 of the top 2 grades. As a benchmark, Google's Programmable Search Engine was used to query the drug name plus "ATC code" followed by regex-based extraction of ATC codes. All imperfect results were reviewed. The population consisted of 12,294 daily dose strings from 86.26% (2908/3371) patients residing in Canada and 13.73% (463/3371) residing in the United States. A prompt using the chain-of-thought reasoning was able to distinguish between aspirin's analgesic versus antithrombotic therapeutic uses and performed well in the pilot sample. In the inferential sample, 87.5% (175/200) were graded as perfect, 5% (10/200) had a minor issue, and 7.5% (15/200) had a major issue. The estimate of the proportion of at least mostly correct classification was 92.5% (185/200, 80% CI 90.1%-94.9%). For the search-based algorithm, 82.5% (165/200) were deemed acceptable. The chain-of-thought reasoning was most helpful with supplements (eg, folic acid) when high doses indicated antianemic preparations. The problem formulation of daily dose inputs and multiple ATC outputs was sometimes incompatible with the drug (eg, pregabalin, calcitriol, and methotrexate). GPT-4o offers cost-effective drug classification from RWD without violating any terms of service. Using a chain-of-thought prompting technique, GPT-4o can reason about drug dosages that affect the class. The wide accessibility of LLMs gives every research team the ability to classify drugs at scale, a key prerequisite of pharmacoepidemiologic research.

In Australia,and elsewhere, one of the most visible manifestations of a changing climate are wildfires that appear to be becoming larger and more frequent.  In 1988 my CSIRO colleagues and I produced the first scientific examination of bushfire danger under climate change. The paper was based on a scenario of the expected climate in 2030 and states that the fire danger every year on average would be larger than the fire danger during the year in which the terrible Ash Wednesday fires occurred. This already seems to have occurred.  Humidity plays a very important role in determining which years will and will not have bad bushfires. The occurrence of large, out-of-control bushfires is (or at least used to be) a rare occurrence that is generally associated with a previous period of drought – most often a long, multi-year drought.  There is a tendency to give particularly memorable bushfires a name and it is instructive to examine such eponymous bushfires.  For Victoria, Australia the sequence of intervals, in years, between such named fires is:                      47, 28, 13, 44, 26, 10.It will be another couple of centuries before we will be able to determine whether the repetition of the 40, 20, 10 sequence is, or is not, a co-incidence. 

The hydrogen storage properties of ball-milled AB-type alloys containing La (a rare earth element) and Ni, Sn were investigated in this study.The experimental alloys were prepared using a compact arc melter under a high-purity argon atmosphere (99.9999%).The influence of microstructure and phase composition on the hydrogen absorption and desorption behavior was examined.The addition of Sn to the LaNi intermetallic significantly reduced the equilibrium pressure without decreasing the maximum hydrogen absorption capacity, up to a Sn content of 10 wt%.Furthermore, increasing the Sn content enhanced the hydrogen absorption rate and slightly stabilized the hydride phase HLaNi.

<strong class="journal-contentHeaderColor">Abstract.</strong> We use a new set of data available to compute 21^st century climate impacts on the hydrology of 221 catchments in high-mountain Central Asia. For each of these subcatchments, a parsimonious steady state stochastic soil moisture water balance model was set up and the partitioning of available water from precipitation into runoff and evaporation computed for different climate futures using the Budyko framework. Climate change sensitivity coefficients are analytically derived for the first time using the total differential method. Relative changes in discharge for three future periods 2011&ndash;2040, 2041&ndash;2070, and 2071&ndash;2100 were computed in relation to the baseline period from 1979&ndash;2011. For the baseline observation period, climate data from a global high-resolution climatology data set (CHELSA V21) were used to extract mean daily subcatchment-specific temperature and precipitation values. Data from the coupled model intercomparison project phase 6 (CMIP6) were used to compute catchment mean future climate data using 4 GCM models with 4 scenario runs each. CMIP6 data were bias corrected with CHELSA V21 observation data. For the spatial distribution of soil parameters, different global products were utilized. The robustness of the soil water balance model results was assessed using a comprehensive sensitivity analysis in relation to variations of these soil parameters over typically observed ranges for each subcatchment. The analysis of climate change suggests increasing precipitation over the three periods (+4.44 %, +5.89 %, and +8.51 % relative increases in median total precipitation averaged over subcatchment and scenarios). Median values of temperatures changes between periods relative to the baseline are +1.33 &deg;C, +2.44 &deg;C, and +3.55 &deg;C. Results of the hydrological soil water balance model runs suggest a median increase of discharge of +4.71 %, +7.44 % and +10.87 % for the corresponding periods. This is a strong indication of a wetter and hotter future in Central Asia, relative to today&rsquo;s hydroclimate. Modelling results suggest that decreasing contributions from glacier melt over the course of the 21^st century will be offset by increases in discharge consistently throughout the region, despite increasing potential evapotranspiration. Increases in relative discharge will be most pronounced in the Afghan Murghab-Harirud basin and in the Amu Darya. Changes in precipitation characteristics in terms of frequency and event depth also indicate possible impacts on hydrological extremes which remains a heavily under researched topic in Central Asia.

The Hydrosat Fusion Hub generates daily 20 m resolution Land Surface Temperature (LST) anywhere on Earth from spatiotemporal data fusion of Landsat, MODIS, and Sentinel-2. The fused data, source data, and intermediate products are served through a spatiotemporal asset catalog (STAC) API (https://fusion-stac.hydrosat.com). Spatiotemporal data fusion is used to first improve the temporal resolution of the LST data and then to predict high spatial resolution LST values for dates where no higher spatial resolution data are available. The resulting fusion data product and source data are then cataloged in the API for easy access by users. RMSE compared to ground sites ranged from 2.89K to 8.34K at individual sites and is less than 5.4K when considering all sites. These data provide the requisite resolutions and information needed for monitoring and managing critical temperature-related changes such as agricultural crop heat and water stress, and water use.