Search

The effect of machine learning and other enhancements on statistical–dynamical forecasts of soil moisture (0–10 and 0–100 cm) and a reference evapotranspiration fraction [evaporative stress index (ESI)] on subseasonal time scales (15–28 days) are explored. The predictors include the current and past land surface conditions and dynamical model hindcasts from the Subseasonal to Seasonal Prediction project (S2S). When the methods are enhanced with machine learning and other improvements, the increases in skill are almost exclusively coming from predictors drawn from observations of current and past land surface states. This suggests that operational S2S flash drought forecasts should focus on optimizing use of information on current conditions rather than on integrating dynamically based forecasts, given the current state of knowledge. Nonlinear machine learning methods lead to improved skill over linear methods for soil moisture but not for ESI. Improvements for both soil moisture and ESI are realized by increasing the sample size by including surrounding grid points in training and increasing the number of predictors. In addition, all the improvements in the soil moisture forecasts predominantly impact soil moistening rather than soil drying—i.e., prediction of conditions moving away from drought rather than into drought—especially when the initial soil state is drier than normal. The physical reasons for the nonlinear machine learning improvements are also explored. Significance Statement: Rapidly intensifying droughts pose extra challenges for predictability. Here, dynamical forecast model output is combined with nonlinear machine learning methods to improve forecasts of rapid changes in soil moisture and the evaporative stress index (ESI). [ABSTRACT FROM AUTHOR]

The Middle East and North Africa (MENA) region has experienced more frequent and severe drought events in recent decades, leading to increasingly pressing concerns over already strained food and water security. An effective drought monitoring and early warning system is thus critical to support risk mitigation and management by countries in the region. Here we investigate the potential for assimilation of leaf area index (LAI) and soil moisture observations to improve the representation of the overall hydrological and carbon cycles and drought by an advanced land surface model. The results reveal that assimilating soil moisture does not meaningfully improve model representation of the hydrological and biospheric processes for this region, but instead it degrades the simulation of the interannual variation in evapotranspiration (ET) and carbon fluxes, mainly due to model weaknesses in representing prognostic phenology. However, assimilating LAI leads to greater improvement, especially for transpiration and carbon fluxes, by constraining the timing of simulated vegetation growth response to evolving climate conditions. LAI assimilation also helps to correct for the erroneous interaction between the prognostic phenology and irrigation during summertime, effectively reducing a large positive bias in ET and carbon fluxes. Independently assimilating LAI or soil moisture alters the categorization of drought, with the differences being greater for more severe drought categories. We highlight the vegetation representation in response to changing land use and hydroclimate as one of the key processes to be captured for building a successful drought early warning system for the MENA region. [ABSTRACT FROM AUTHOR]

Recent years have seen growing appreciation that rapidly intensifying flash droughts are significant climate hazards with major economic and ecological impacts. This has motivated efforts to inventory, monitor, and forecast flash drought events. Here we consider the question of whether the term "flash drought" comprises multiple distinct classes of event, which would imply that understanding and forecasting flash droughts might require more than one framework. To do this, we first extend and evaluate a soil moisture volatility–based flash drought definition that we introduced in previous work and use it to inventory the onset dates and severity of flash droughts across the contiguous United States (CONUS) for the period 1979–2018. Using this inventory, we examine meteorological and land surface conditions associated with flash drought onset and recovery. These same meteorological and land surface conditions are then used to classify the flash droughts based on precursor conditions that may represent predictable drivers of the event. We find that distinct classes of flash drought can be diagnosed in the event inventory. Specifically, we describe three classes of flash drought: "dry and demanding" events for which antecedent evaporative demand is high and soil moisture is low, "evaporative" events with more modest antecedent evaporative demand and soil moisture anomalies, but positive antecedent evaporative anomalies, and "stealth" flash droughts, which are different from the other two classes in that precursor meteorological anomalies are modest relative to the other classes. The three classes exhibit somewhat different geographic and seasonal distributions. We conclude that soil moisture flash droughts are indeed a composite of distinct types of rapidly intensifying droughts, and that flash drought analyses and forecasts would benefit from approaches that recognize the existence of multiple phenomenological pathways. [ABSTRACT FROM AUTHOR]

The term "flash drought" is frequently invoked to describe droughts that develop rapidly over a relatively short timescale. Despite extensive and growing research on flash drought processes, predictability, and trends, there is still no standard quantitative definition that encompasses all flash drought characteristics and pathways. Instead, diverse definitions have been proposed, supporting wide-ranging studies of flash drought but creating the potential for confusion as to what the term means and how to characterize it. Use of different definitions might also lead to different conclusions regarding flash drought frequency, predictability, and trends under climate change. In this study, we compared five previously published definitions, a newly proposed definition, and an operational satellite-based drought monitoring product to clarify conceptual differences and to investigate the sensitivity of flash drought inventories and trends to the choice of definition. Our analyses indicate that the newly introduced Soil Moisture Volatility Index definition effectively captures flash drought onset in both humid and semi-arid regions. Analyses also showed that estimates of flash drought frequency, spatial distribution, and seasonality vary across the contiguous United States depending upon which definition is used. Definitions differ in their representation of some of the largest and most widely studied flash droughts of recent years. Trend analysis indicates that definitions that include air temperature show significant increases in flash droughts over the past 40 years, but few trends are evident for definitions based on other surface conditions or fluxes. These results indicate that "flash drought" is a composite term that includes several types of events and that clarity in definition is critical when monitoring, forecasting, or projecting the drought phenomenon. [ABSTRACT FROM AUTHOR]

South and Southeast Asia is subject to significant hydrometeorological extremes, including drought. Under rising temperatures, growing populations, and an apparent weakening of the South Asian monsoon in recent decades, concerns regarding drought and its potential impacts on water and food security are on the rise. Reliable sub-seasonal to seasonal (S2S) hydrological forecasts could, in principle, help governments and international organizations to better assess risk and act in the face of an oncoming drought. Here, we leverage recent improvements in S2S meteorological forecasts and the growing power of Earth observations to provide more accurate monitoring of hydrological states for forecast initialization. Information from both sources is merged in a South and Southeast Asia sub-seasonal to seasonal hydrological forecasting system (SAHFS-S2S), developed collaboratively with the NASA SERVIR program and end users across the region. This system applies the Noah-Multiparameterization (NoahMP) Land Surface Model (LSM) in the NASA Land Information System (LIS), driven by downscaled meteorological fields from the Global Data Assimilation System (GDAS) and Climate Hazards InfraRed Precipitation products (CHIRP and CHIRPS) to optimize initial conditions. The NASA Goddard Earth Observing System Model sub-seasonal to seasonal (GEOS-S2S) forecasts, downscaled using the National Center for Atmospheric Research (NCAR) General Analog Regression Downscaling (GARD) tool and quantile mapping, are then applied to drive 5 km resolution hydrological forecasts to a 9-month forecast time horizon. Results show that the skillful predictions of root zone soil moisture can be made 1 to 2 months in advance for forecasts initialized in rainy seasons and up to 8 months when initialized in dry seasons. The memory of accurate initial conditions can positively contribute to forecast skills throughout the entire 9-month prediction period in areas with limited precipitation. This SAHFS-S2S has been operationalized at the International Centre for Integrated Mountain Development (ICIMOD) to support drought monitoring and warning needs in the region. [ABSTRACT FROM AUTHOR]

Many regions in Africa and the Middle East are vulnerable to drought and to water and food insecurity, motivating agency efforts such as the U.S. Agency for International Development's (USAID) Famine Early Warning Systems Network (FEWS NET) to provide early warning of drought events in the region. Each year these warnings guide life-saving assistance that reaches millions of people. A new NASA multimodel, remote sensing-based hydrological forecasting and analysis system, NHyFAS, has been developed to support such efforts by improving the FEWS NET's current early warning capabilities. NHyFAS derives its skill from two sources: (i) accurate initial conditions, as produced by an offline land modeling system through the application and/or assimilation of various satellite data (precipitation, soil moisture, and terrestrial water storage), and (ii) meteorological forcing data during the forecast period as produced by a state-of-the-art ocean-land-atmosphere forecast system. The land modeling framework used is the Land Information System (LIS), which employs a suite of land surface models, allowing multimodel ensembles and multiple data assimilation strategies to better estimate land surface conditions. An evaluation of NHyFAS shows that its 1-5-month hindcasts successfully capture known historic drought events, and it has improved skill over benchmark-type hindcasts. The system also benefits from strong collaboration with end-user partners in Africa and the Middle East, who provide insights on strategies to formulate and communicate early warning indicators to water and food security communities. The additional lead time provided by this system will increase the speed, accuracy, and efficacy of humanitarian disaster relief, helping to save lives and livelihoods. [ABSTRACT FROM AUTHOR]

The objective of the North American Land Data Assimilation System (NLDAS) is to provide best-available estimates of near-surface meteorological conditions and soil hydrological status for the continental United States. To support the ongoing efforts to develop data assimilation (DA) capabilities for NLDAS, the results of Gravity Recovery and Climate Experiment (GRACE) DA implemented in a manner consistent with NLDAS development are presented. Following previous work, GRACE terrestrial water storage (TWS) anomaly estimates are assimilated into the NASA Catchment land surface model using an ensemble smoother. In contrast to many earlier GRACE DA studies, a gridded GRACE TWS product is assimilated, spatially distributed GRACE error estimates are accounted for, and the impact that GRACE scaling factors have on assimilation is evaluated. Comparisons with quality-controlled in situ observations indicate that GRACE DA has a positive impact on the simulation of unconfined groundwater variability across the majority of the eastern United States and on the simulation of surface and root zone soil moisture across the country. Smaller improvements are seen in the simulation of snow depth, and the impact of GRACE DA on simulated river discharge and evapotranspiration is regionally variable. The use of GRACE scaling factors during assimilation improved DA results in the western United States but led to small degradations in the eastern United States. The study also found comparable performance between the use of gridded and basin-averaged GRACE observations in assimilation. Finally, the evaluations presented in the paper indicate that GRACE DA can be helpful in improving the representation of droughts. [ABSTRACT FROM AUTHOR]

In the United States, irrigation represents the largest consumptive use of freshwater and accounts for approximately one-third of total water usage. Irrigation impacts soil moisture and can ultimately influence clouds and precipitation through land-planetary boundary layer (PBL) coupling processes. This study utilizes NASA's Land Information System (LIS) and the NASA Unified Weather Research and Forecasting Model (NU-WRF) framework to investigate the effects of drip, flood, and sprinkler irrigation methods on land-atmosphere interactions, including land-PBL coupling and feedbacks at the local scale. To initialize 2-day, 1-km WRF forecasts over the central Great Plains in a drier-than-normal (2006) and a wetter-than-normal year (2008), 5-yr irrigated LIS spinups were used. The offline and coupled simulation results show that regional irrigation impacts are sensitive to time, space, and method and that irrigation cools and moistens the surface over and downwind of irrigated areas, ultimately resulting in both positive and negative feedbacks on the PBL depending on the time of day and background climate conditions. Furthermore, the results portray the importance of both irrigation method physics and correct representation of several key components of land surface models, including accurate and timely land-cover and crop-type classification, phenology (greenness), and soil moisture anomalies (through a land surface model spinup) in coupled prediction models. [ABSTRACT FROM AUTHOR]

Positive soil moisture-precipitation feedbacks can intensify heat and prolong drought under conditions of precipitation deficit. Adequate representation of these processes in regional climate models is, therefore, important for extended weather forecasts, seasonal drought analysis, and downscaled climate change projections. This paper presents the first application of the NASA Unified Weather Research and Forecasting Model (NU-WRF) to simulation of seasonal drought. Simulations of the 2006 southern Great Plains drought performed with and without soil moisture memory indicate that local soil moisture feedbacks had the potential to concentrate precipitation in wet areas relative to dry areas in summer drought months. Introduction of a simple dynamic surface albedo scheme that models albedo as a function of soil moisture intensified the simulated feedback pattern at local scale-dry, brighter areas received even less precipitation while wet, whereas darker areas received more-but did not significantly change the total amount of precipitation simulated across the drought-affected region. This soil-moisture-mediated albedo land-atmosphere coupling pathway is structurally excluded from standard versions of WRF. [ABSTRACT FROM AUTHOR]

Advanced land surface models (LSMs) offer detailed estimates of distributed hydrological fluxes and storages. These estimates are extremely valuable for studies of climate and water resources, but they are difficult to verify as field measurements of soil moisture, evapotranspiration, and surface and subsurface runoff are sparse in most regions. In contrast, river discharge is a hydrologic flux that is recorded regularly and with good accuracy for many of the world's major rivers. These measurements of discharge spatially integrate all upstream hydrological processes. As such, they can be used to evaluate distributed LSMs, but only if the simulated runoff is properly routed through the river basins. In this study, a rapid, computationally efficient source-to-sink (STS) routing scheme is presented that generates estimates of river discharge at gauge locations based on gridded runoff output. We applied the scheme as a postprocessor to archived output of the Global Land Data Assimilation System (GLDAS). GLDAS integrates satellite and ground-based data within multiple offline LSMs to produce fields of land surface states and fluxes. The application of the STS routing scheme allows for evaluation of GLDAS products in regions that lack distributed in situ hydrological measurements. We found that the four LSMs included in GLDAS yield very different estimates of river discharge and that there are distinct geographic patterns in the accuracy of each model as evaluated against gauged discharge. The choice of atmospheric forcing data set also had a significant influence on the accuracy of simulated discharge. [ABSTRACT FROM AUTHOR]

Snow cover over land has a significant impact on the surface radiation budget, turbulent energy fluxes to the atmosphere, and local hydrological fluxes. For this reason, inaccuracies in the representation of snow-covered area (SCA) within a land surface model (LSM) can lead to substantial errors in both offline and coupled simulations. Data assimilation algorithms have the potential to address this problem. However, the assimilation of SCA observations is complicated by an information deficit in the observation—SCA indicates only the presence or absence of snow, not snow water equivalent—and by the fact that assimilated SCA observations can introduce inconsistencies with atmospheric forcing data, leading to nonphysical artifacts in the local water balance. In this paper, a novel assimilation algorithm is presented that introduces Moderate Resolution Imaging Spectroradiometer (MODIS) SCA observations to the Noah LSM in global, uncoupled simulations. The algorithm uses observations from up to 72 h ahead of the model simulation to correct against emerging errors in the simulation of snow cover while preserving the local hydrologic balance. This is accomplished by using future snow observations to adjust air temperature and, when necessary, precipitation within the LSM. In global, offline integrations, this new assimilation algorithm provided improved simulation of SCA and snow water equivalent relative to open loop integrations and integrations that used an earlier SCA assimilation algorithm. These improvements, in turn, influenced the simulation of surface water and energy fluxes during the snow season and, in some regions, on into the following spring. [ABSTRACT FROM AUTHOR]

The Euphrates Plain (EP) experiences large interannual variability in vegetation cover, especially in areas of marginal rain-fed agriculture. Vegetation in this region is primarily limited by available soil moisture, as determined by winter precipitation, spring precipitation, and air temperature. Satellite analyses indicate that the springtime normalized difference vegetation index (NDVI) is negatively correlated with surface albedo, and that interannual variability in albedo in the EP produces an estimated forcing on the radiation balance that peaks at 16.0 W m-2 in May. Simulations with a regional climate model indicate that surface energy fluxes during a drought year (1999) differed substantially from those during a year with normal precipitation (2003). These differences were geographically specific, with the EP exhibiting increased albedo and decreased sensible heat flux while the neighboring Zagros Plateau region showed no albedo effect, a large increase in sensible heat flux, and an offsetting reduction in latent heat flux. In both the EP and the Zagros there was a potential for positive feedbacks on temperature and drought in late spring, though the most likely feedback mechanisms differed between the two regions: in the EP surface brightening leads to cooling and reduced turbulent heat flux, while in the Zagros region reduced latent heat flux leads to warming and a deepening of the planetary boundary layer. [ABSTRACT FROM AUTHOR]

Summary: A land surface model’s ability to simulate states (e.g., soil moisture) and fluxes (e.g., runoff) is limited by uncertainties in meteorological forcing and parameter inputs as well as inadequacies in model physics. In this study, anomalies of terrestrial water storage (TWS) observed by the Gravity Recovery and Climate Experiment (GRACE) satellite mission were assimilated into the NASA Catchment land surface model in western and central Europe for a 7-year period, using a previously developed ensemble Kalman smoother. GRACE data assimilation led to improved runoff estimates (in temporal correlation and root mean square error) in 17 out of 18 hydrological basins, even in basins smaller than the effective resolution of GRACE. Improvements in root zone soil moisture were less conclusive, partly due to the shortness of the in situ data record. GRACE data assimilation also had significant impacts in groundwater estimates including trend and seasonality. In addition to improving temporal correlations, GRACE data assimilation also reduced increasing trends in simulated monthly TWS and runoff associated with increasing rates of precipitation. The assimilation downscaled (in space and time) and disaggregated GRACE data into finer scale components of TWS which exhibited significant changes in their dryness rankings relative to those without data assimilation, suggesting that GRACE data assimilation could have a substantial impact on drought monitoring. Signals of drought in GRACE TWS correlated well with MODIS Normalized Difference Vegetation Index (NDVI) data in most areas. Although they detected the same droughts during warm seasons, drought signatures in GRACE derived TWS exhibited greater persistence than those in NDVI throughout all seasons, in part due to limitations associated with the seasonality of vegetation. Mass imbalances associated with GRACE data assimilation and challenges of using GRACE data for drought monitoring are discussed. [ABSTRACT FROM AUTHOR]

A nine-year (2010–2018) field study in the Debre Mawi watershed was conducted to understand the effect of governmentally-imposed and farmer-initiated conservation practices. The watershed is in the sub-humid Ethiopian Highlands which experience high and increasing erosion rates despite years of conservation efforts. Consequently, reservoirs are filling up with sediment and soil degradation is enhanced, calling for the evaluation of conservation practices currently in use. The few past long-term experimental studies on structural practices are inconclusive. In addition, only anecdotal information is available for streamflow and sediment loss. Precipitation, stream discharge, and suspended sediment concentrations were recorded manually in the Debre Mawi watershed during the nine-year period. Groundwater depth and total saturated area measurements were taken for selected periods. From 2012 to 2014, government-mandated conservation practices were constructed, which consisted of 50-cm-deep infiltration furrows with bunds downslope. These furrows were filled in with sediment by 2018. At the same time, the acreage of eucalyptus trees planted by farmers on the most vulnerable lands tripled to 5% of the total area with most trees fully grown in 2018. Runoff coefficients and sediment concentrations decreased steadily throughout the nine years. In the saturated bottomlands, the observations suggested that government-sponsored infiltration furrows in the saturated bottomlands were ineffective and may concentrate flows and enhance gully erosion, while eucalyptus trees appear effective. The results of this observational study point to both the potential benefits of conservation practices in this sub-humid tropical highland region and to emerging long-term risks. If structural conservation is to be pursued in watersheds like Debre Mawi, due attention must be given to the safe removal of excess water from the valley bottoms. The vegetative farmer-initiated practice of planting eucalyptus trees effectively reduced streamflow and erosion, but at the same time, might dry up wells during the dry monsoon phase which should be investigated further. Image 1 • Acreage of eucalyptus planted by farmers increased three-fold in nine years. • Observed runoff and sediment yield declined with increased acreage of eucalytus. • Dry phase eucalyptus evaporation delayed saturation excess runoff in rain phase. • Infiltration furrows were not maintained, filled up, and appeared ineffective. • Groundwater irrigation is feasible when lava dikes block subsurface downhill flow. [ABSTRACT FROM AUTHOR]