This session highlights advances in applications of AI/ML and process-based hydrological modeling to provide actionable information across a range of timescales (paleo, retro, nowcast, forecast, projection) and space scales (local-to-global). This includes theoretical developments in process modeling, machine learning, knowledge guided deep learning, hydroinformatics, etc., as well as research-to-operations (R2O) and application-specific innovation. The goal of this session is to help model and method developers and users understand the state-of-the-art in terms of outcome-oriented (rather than science-oriented) hydrological simulation, analysis and prediction. Papers describing current challenges for this community are also welcome, with a special interest in highlighting needs and opportunities for cross-disciplinary collaboration between industry, academia, and government.
Drought is a multi-faceted phenomenon that challenges our current prediction capabilities, yet its environmental and economic consequences are among the most serious of all natural disasters. It straddles the nexus of all five pillars of the Environmental Security theme for this year’s AMS Annual Meeting. In a warming climate, drought is expected to increase in frequency, duration, and intensity at both regional and global scales, and this will result in increasing environmental security risk.
Improving analysis and prediction of all drought types requires the use of multiple data sources, including in-situ and remote-sensing data, surface observations, and indicators of societal impact. Satellite hydrological variables and vegetation indices have contributed dramatically to our understanding of the mechanisms of drought occurrence and development, as well as facilitating the separation of the drought signal from normal hydrologic and vegetation conditions. Remotely sensed land observations are used to force or parameterize models, and the hydrological outputs provide the foundation for existing drought indicators. However, making significant improvements in monitoring and prediction will not only require advances in understanding drought mechanisms, but also of the societal impacts and how to better manage water resources. There are still many open scientific questions related to data fusion, integration of drought indicators, emerging social media data sources and the optimal combination of these data sets for providing insights to climate, environmental security, and societal changes with respect to drought events.
Specific topics addressed by presenters could include, but are not limited to: current drought prediction science and skill at various lead times; advances in our understanding of the causes and characteristics of drought, through climatological analyses, impacts of land-atmosphere interactions, and numerical simulations; innovative management uses of that science; and case studies illustrating advances in understanding, monitoring and prediction of drought and drought impacts. Further, papers addressing gaps and deficiencies in our current methods for predicting droughts and estimating its effects on vegetation, water and energy resources, environmental security, and the health and food security of human populations are also invited.
Significant advances in the estimation and application of evapotranspiration (ET) and atmospheric demand (Eo) have been made, with a broad range of methods used for estimation (in-situ to remote sensing and modeling) and applications spanning from ecology to agriculture to drought monitoring. Specific topics of interest for this session may include but are not limited to: (1) advances in the estimation of ET from various perspectives, including remote sensing platforms, machine learning or artificial intelligence (AI) techniques, ground-based point observations and parameterizations, plant-based experimentation, and water budgets; (2) operational ET estimation and/or application; (3) estimation of ET or Eo using ensemble modeling techniques; (4) forecasting of ET or Eo from point to regional spatial scales and subseasonal-to-seasonal timescales; and (5) ET and Eo for input or assimilation into operational land-surface or hydrologic modeling, and as a metric of hydroclimatic trends and variability. This year, in keeping with the AMS theme of weather, water and climate for a more secure world, we are particularly seeking submissions relating to how ET and Eo can be used to monitor, predict, and/or mitigate disruptions in ecosystems, agriculture (food production), drought, or other water related issues that affect human welfare and national security.
Predictions of water cycle extremes are limited by uncertainties in key ecohydrological model processes, such as transpiration and latent heat regulated by stomatal conductance, and relevant data products largely related to scale, heterogeneity, complexity, representativeness, and model structure. Recent advances in machine learning offer solutions to some of these issues. Thus, this session focuses on innovative methods for (1) data-driven Earth system modeling, including hybrid process-/machine-learning parameterizations and surrogate-based modeling, of terrestrial ecohydrological and hydrometeorological system processes, including but not limited to stomatal control on transpiration and latent heat with feedbacks to convection, precipitation, soil moisture dynamics, atmospheric rivers, and changes in these relationships under climate change; (2) assessing the fidelity of ecohydrological process representations in models through applications of statistical and machine learning approaches, including knowledge-guided, interpretable, and transfer learning methods; (3) developing, synthesizing, and evaluating high resolution data products, including but not limited to soil and nutrient properties, vegetation distribution, hydraulic properties and meteorological variables. Encouraged are contributions that focus on model benchmarking metrics, uncertainty quantification methods, and hierarchical modeling approaches employing novel model-data fusion and machine learning applications to improve the mechanistic understanding of physical and ecohydrological responses and feedbacks to the Earth system, particularly under climate change.
This session invites papers on all aspects of extreme precipitation, including snowfall, relationship to floods, and hydrologic impacts. Possible topics include observations, modeling, short-term and seasonal prediction, climate change, and risk assessment. Papers exploring the causes and consequences of individual extreme precipitation events that cause floods or terminate droughts, details of the relationship between extreme precipitation and flooding, extreme snowfall accumulation and melt, as well as the drivers and environmental security impacts of changing extreme precipitation and flood risk are particularly encouraged.
Flash droughts, characterized by unusually rapid onset and/or intensification, present significant challenges for drought monitoring, forecasting, and mitigation. Their rapid intensification reduces warning and preparation time, resulting in unique and magnified impacts compared to slower-developing droughts, especially to food and energy production, health outcomes, and water resource availability. While a physical understanding of flash droughts has improved significantly over the past decade, the effect of flash drought on environmental security remains less clear. In line with the theme of the 102nd AMS Annual Meeting, we are inviting submissions that advance our understanding of the linkages between flash droughts and energy, food, health, and water security, and how to improve monitoring, early warning, and prediction of these events to reduce vulnerabilities from global to community scales.
Flood prediction and management are global needs. Hydrologic modeling advancements have led to the development of new predictive tools that leverage the collection and synthesis of observational data into modeling frameworks. These advancements have produced a research and development environment exhibiting a rapidly expanding range of data and products. This session invites contributions from all sectors of the AMS and broader hydrologic community whose work focuses on: (1) the development of flood-prediction tools (models, assimilation of observations, and methodologies to create and improve hydrologic representation of flood processes), (2) the analysis of floods and (3) lessons learned from managing floods from scientific and societal perspectives.
We are particularly interested in submissions focused on the current state and future of continental and global scale hydrologic prediction. We hope to continue exploring issues raised during the 2017 Town Hall discussion at the 97th AMS Annual Meeting in Seattle, WA "Challenges, Opportunities, and Advances in Global Flood Forecasting”. The focus was on flood prediction systems and methods applied at the global scale. Since that time further advancements in computing power and improved methodologies have measurably improved flood monitoring and have made running distributed hydrologic models at small spatial scales, on the order of 1–10 km2 spatial resolution, more feasible, with forecast lead times of up to a week or more.
Our hope is that proposed papers on current continental and global scale hydrologic forecasting systems will emphasize analyses of model performance compared to ground-truthed observations.The importance of these systems are underscored by the emergence of the Global Flood Partnership (GFP), which is aimed at the “development of flood observational and modeling infrastructure, leveraging on existing initiatives for better predicting and managing flood disaster impacts and flood risk globally.” The success of the GFP can directly benefit many countries where flood risk is poorly managed, but is critical to humanitarian assistance efforts. Significant questions surround the practical limits of continental and global scale hydrologic predictions that are useful and meaningful to communities’ efforts in Disaster Risk Reduction (DRR).
Urban floods are flashy and can cause wide spread damage due to the high concentration of population and development. Flood warning plays a critical part in providing flood safety resilience to urban communities during intense storm events. Early and more accurate rain and wind intensity forecasts, measurement, flood monitoring and forecasting, and timely communication of flood threat prior and during storm events are example ways in which a flood warning system can help reduce flood risk to a community. This topical session wishes to bring together professionals working in urban flooding to discuss their experiences in the following areas:
How Can Soil Moisture Make a Difference? Using Soil Moisture Data for Improved Applications or Decision Making:
Soil moisture is an active area of hydrological and climatological research. Simultaneously, there is strong interest from natural resource managers, state climatologists, and other end users for better applications and tools that harness the predictive potential of soil moisture to improve decision making on the ground. The National Coordinated Soil Moisture Monitoring Network is a collaborative effort between federal and state agencies, research institutes, and other organizations to facilitate progress in measuring, analyzing, and applying soil moisture data, and to ensure coordination between research activities and application needs.
This session is focused on research that demonstrates the value of soil moisture data for improved conditions monitoring or decision making across the full range of application areas, including flood or drought forecasting, agricultural or ecological monitoring, fire danger assessment, water resource management, and so forth. Specific case studies, on-the-ground stories, and other specific demonstrations will be emphasized.
Recognizing that this is an emerging area, demonstration of potential (as opposed to actual) application value is also acceptable. Given increasing opportunities specific to in situ soil moisture data, priority will be given to research/applications that use in situ data in some capacity.
The world has water resources challenges that are being exacerbated by growing populations, climate change, land-use and landcover change, and evolving socioeconomic drivers that significantly challenge and impact the foresight of those charged with setting policy and making decisions. We must consider if our water monitoring and modeling capabilities are sufficient to address these challenges. The focus of this session is to identify observational and modeling gaps in support of continental to global scale water resource monitoring and prediction. We envision papers that evoke the legacy of, and build from the Global Energy and Water Exchanges (GEWEX) Continental-Scale International Project (GCIP) and the GEWEX America’s Prediction Project (GAPP) and similar past and current programs.
The session is a forum to bring the observational and modeling communities together to advance the state of the science and our predictive capabilities. As such submissions are invited that explore interdisciplinary and integrated approaches to water cycle science that lead to a better understanding of water availability now and in the changing future. Of particular relevance are papers that explore ways to evaluate, improve and integrate existing surface observational networks across continental to global scales, in the context of terrestrial-based and space-borne remote sensing, and in support of Earth system modeling of the water-cycle, in order to better support national and global operational weather and hydroclimate monitoring and modeling efforts. For example, approaches including observational verification, parameterization development, data assimilation, observing system design, employing in situ observations to provide physical process constraints not currently possible/available from satellite observing systems, are welcome. It is recognized that studies over more limited spatial and temporal scales do lead to an improved characterization of the water cycle at continental to global scales and at time periods encompassing climate, and such papers are also encouraged.
Reservoir evaporation is that hard-to-quantify yet important-to-get-right variable affecting reservoir water balances today, tomorrow, and at the long-range planning timescale. Recent studies show promise in improving reservoir evaporation estimates using floating on-water monitoring platforms. The in-situ datasets collected from these platforms provide invaluable over-water observations and estimates of reservoir evaporation that can be used to support calibration and validation of other technologies, including remotely sensed datasets and modeling approaches. These technologies provide promise for extrapolating from sparse in-situ data collection networks to develop a broader spatial coverage of reservoir evaporation. Improving the accuracy and coverage of existing reservoir evaporation datasets is just one piece of the equation. However, can such improvements be factored into day-to-day reservoir operations and into long-range water planning? How can potential trends in historical reservoir evaporation be quantified with confidence given the inherent uncertainties and known deficiencies of the existing monitoring infrastructure? How can enhancements to the existing monitoring infrastructure and estimation methods today generate confidence in future reservoir evaporative loss under climate change?
This session welcomes abstracts focusing on, but not limited to, on-going applications of new technologies for addressing current challenges related to reservoir evaporation monitoring, how reservoir evaporation data are currently used for operational water supply decisions and for long-range water supply planning, how reservoir evaporative loss can be factored in as a category of water use/water demand, how climate change affects reservoir evaporative loss, and how such loss can be accounted for in estimates of future surface water availability.
Land Data Assimilation Systems (LDASs) are widely used to enable remote sensing data infusion, and in turn to construct spatially and temporally consistent estimates of land-surface conditions. Since the initiation of the North American (NLDAS) and Global LDAS (GLDAS), LDASs have been developed in various nations and regions, and have extended from being offline, semi-coupled to fully coupled. Benefiting from land surface model evolution and the availability of high-quality in-situ and satellite observations, LDASs have become more comprehensive, representing feedbacks and interactions between biological, meteorological and hydrological processes. They can thus be used to provide optimal initial conditions for offline and/or coupled modeling to support drought/flood monitoring and forecasting, agricultural crop planning, and water resource management, in addition to their role in numerical weather prediction. This session will highlight the advances in the development of LDASs and their applications. We particularly encourage contributions in (1) Developing the regional and global LDASs; (2) Advancing LSMs for LDASs; (3) Refining or evaluating land data assimilation techniques in LDASs; (4) Use of LDASs for applications including numerical weather prediction, water resources, food security, agriculture management, among others.
Land-atmosphere and land-ocean interactions play a key role in climate variability and change, as well as climate/weather predictability across space and time. The land’s role in the Earth system – its impact on atmospheric and ocean climatology and variability across a broad range of timescales, ranging from hours to centuries, for past, present, and future climates – has been the subject of much recent exploratory research. The meteorological, hydrological, biophysical, biogeochemical, and ecosystem processes, as well as the boundary-layer processes, that underlie the connections between surface and atmosphere are not yet fully understood. The scarcity of relevant observations, the complexity of the underlying processes and feedbacks, and the wide range of scales involved necessitate coordinated and exceedingly interdisciplinary investigations. This session focuses on (1) interfaces between climate, ecosystems, and the land branches of the energy, water, and carbon cycles and the impact of associated land processes on climate variability and change, including extreme events (such as droughts and flooding); (2) the dynamic, physical, and biogeochemical mechanisms by which the land surface (e.g., soil moisture and temperature, albedo, snow, vegetation, and streamflow) influence land, atmosphere, and ocean processes and their interactions at subseasonal to decadal time scales; (3) predictability associated with land initialization (i.e., soil moisture, soil temperature, vegetation, snow, and aerosol in snow, etc.) and land–atmosphere/ocean interactions from subseasonal to decadal time scales; and (4) application and analyses of large scale field campaign data, national and international observational networks (e.g., FLUXNET), satellite remote sensing, and reanalysis for land model development and land-atmosphere/ocean interaction studies. We welcome papers addressing any of these topics.
Millions of people worldwide rely on snow accumulation and melt for water resources. Particularly in developing countries and very remote regions, in-situ snow data are rarely available due to financial constraints and safety concerns. Thus, assessing the volume of water contained in the snowpack can be especially difficult but is extremely important for water resources management. Without information about the snowpack, these regions are particularly susceptible to flooding or drought, which may have broad reaching political and economic implications on the security of the region. Accurate estimates of snow water equivalent (SWE), snow covered area (SCA), melt timing, and other properties of snow are critical in accurately predicting runoff response for water resource management and thus aspects of water, food, and societal stability. Remote sensing and modeling techniques provide methods for observing and detecting snow evolution, onset of snowmelt, and spatial extent. Existing and novel remote sensing techniques have been employed to observe snow characteristics. Local and regional snow models have shown the ability to estimate snow properties, such as snow volume, liquid water content, and melt. Observational, in-situ datasets that drive these models with meteorological inputs and modify the model through data assimilation techniques are critical in accurately portraying snow evolution.
This session invites research on existing and novel methods for remote sensing, modeling, and data assimilation of snow hydrology, particularly efforts that identify and overcome gaps in the current knowledge of snow observation and modeling. We encourage submissions that consider water security, connecting snow accumulation and melt to water resource availability around the globe.
NOAA’s National Water Model: Improving Hydrologic Consistency, Operational Forecast Utility, and Enhancing Impact-Based Decision Support Services
Topic Description: Continental-scale hydrologic prediction models provide a consistent means to predict floods, droughts, water supply, and water quality across broad spatial scales. NOAA’s National Water Model (NWM) is an operational example of such a tool. This session seeks submissions on topics aimed at improving hydrologic consistency, operational forecast utility, and enhancing Impact-Based Decision Support Services (IDSS). Topics of interest include improved process representation across differing landscape and anthropogenic conditions; coupled inland and coastal processes and forecasting total water level; QA/QC representation; representation of geomorphic and bathymetric conditions as applied to flood inundation forecasting; and services testing and evaluation.
Safe access to sustainable water supplies and nutritious food stores are key to society’s well being. Agricultural irrigation is responsible for over 70 percent of all freshwater appropriated for human use, contributing to 40% of crop production in 20% of the agricultural landscapes. Yet, crop production in rainfed and irrigated working lands is exposed to floods and droughts, compromising food security in a family farm as well as worldwide. Observed climate change is already affecting food and water security through increasing temperatures, changing precipitation patterns, and greater frequency of extreme events. As climate change continues to impact both water and food security, understanding how physical and social processes are affected by extreme events and future climate changes is critical. and the connections between them will be increasingly important. Water resources are becoming more volatile, and the food sectors’ dependence on water becomes more acute. This is reflected in and challenges the achievement of the United Nations Sustainability Goals, several of which include ensuring water and food security at their core to meet the overall plan of reducing inequity and eliminating extreme poverty. This session invites presentations exploring the physical and/or social processes leading to increases in water and food security risks in the past, present, and future, including land surface changes, impacts of extreme events, diversions and their impact on transboundary agreements, as well as presentations focused on approaches to enhance food and water security.
The precipitation session focuses on precipitation observation and process understanding, modeling, estimation across spatiotemporal scales, advanced statistical techniques including artificial intelligence (AI)-based methods to produce, analyze, and/or visualize precipitation data, and applications of in-situ and remotely sensed precipitation products. Process understanding and observational development topics include, but are not limited to (1) precipitation processes and modeling in coupled or uncoupled model systems; (2) improved assimilation algorithms leveraging precipitation process understanding; (3) recent development pertaining to fusion and downscaling of precipitation products; while applications could include: (4) assimilation of precipitation and precipitation-related variables in weather or water models; (5) impact of uncertainties associated with precipitation observations on hydrologic design and modeling, (6) assessment of precipitation variability, across spatiotemporal scales.
Probabilistic hydrometeorological forecasting in a comprehensive integration of physical modeling, probability and statistics, and numerical methods. Model ensembles, parameter optimization, data assimilation, and input data quality all can contribute to forecast uncertainties. Significant challenges exist in improving probabilistic forecasts and addressing associated uncertainties in hydrometeorological applications. This session solicits papers on theoretical, experimental, and applied studies focusing on ensemble forecasting and uncertainty analysis in both offline and coupled systems. The topics include but not limited to:
Global and regional quantitative precipitation estimates (QPEs) are critical for understanding climate variability and hydrometeorological cycles, detecting natural disasters, improving flash flood and weather forecasts, and effectively managing the use of earth's freshwater resources. However, obtaining accurate QPEs is a challenging task in many areas of the world due to sparse gauge networks, complex terrain, and global water cycle acceleration. Recent advances in radar and satellite remote sensing of precipitation are progressing rapidly, with the aims of providing accurate and high-resolution precipitation estimates, accurate flash flood forecasting, and understanding the causes and underlying processes of these natural hazards. This session invites high quality, original research contributions from radar and satellite meteorology and associated data sciences, including improved QPE retrievals from satellite and radar remote observations and novel methods to produce multi-sensor QPEs by merging multiple satellite/radar products and/or in-situ observations. Assessments of remotely sensed QPE product performance, as well as demonstrations of the applicability of remotely sensed QPEs for improved weather and hazards forecasting or understanding of hydrometeorological processes are also encouraged.
There have been significant recent advances in the measurement and prediction of urban rainfall, and in prediction of urban hydrology. In order to improve liveability of the urban landscape, more attention towards restoring natural water balance is needed. Apart from this, stormwater is increasingly regarded as a resource which could, if tapped, address current and future urban water needs. Despite these advances, many important challenges remain in urban hydrology. There is a need to further improve short-term rainfall prediction in urban regions. In addition, uncertainty associated with climate change imposes a requirement that stormwater management systems are adaptable and resilient to future changes. Improvements in urban hydrology will be crucial in addressing some of these challenges. In this session, we invite papers focused on, but not limited to, advances in urban modeling, precipitation estimation, and instrumentation, as well as studies involving impacts of climate change on urban hydrology.