State Hydrological Institute

Synthesis of river-monitoring data reveals that the average annual discharge of fresh water from the six largest Eurasian rivers to the Arctic Ocean increased by 7% from 1936 to 1999. The average annual rate of increase was 2.0 +/- 0.7 cubic kilometers per year. Consequently, average annual discharge from the six rivers is now about 128 cubic kilometers per year greater than it was when routine measurements of discharge began. Discharge was correlated with changes in both the North Atlantic Oscillation and global mean surface air temperature. The observed large-scale change in freshwater flux has potentially important implications for ocean circulation and climate.

Freeze and breakup dates of ice on lakes and rivers provide consistent evidence of later freezing and earlier breakup around the Northern Hemisphere from 1846 to 1995. Over these 150 years, changes in freeze dates averaged 5.8 days per 100 years later, and changes in breakup dates averaged 6.5 days per 100 years earlier; these translate to increasing air temperatures of about 1.2°C per 100 years. Interannual variability in both freeze and breakup dates has increased since 1950. A few longer time series reveal reduced ice cover (a warming trend) beginning as early as the 16th century, with increasing rates of change after about 1850.

This paper is the outcome of a community initiative to identify major unsolved scientific problems in hydrology motivated by a need for stronger harmonisation of research efforts. The procedure involved a public consultation through online media, followed by two workshops through which a large number of potential science questions were collated, prioritised, and synthesised. In spite of the diversity of the participants (230 scientists in total), the process revealed much about community priorities and the state of our science: a preference for continuity in research questions rather than radical departures or redirections from past and current work. Questions remain focused on the process-based understanding of hydrological variability and causality at all space and time scales. Increased attention to environmental change drives a new emphasis on understanding how change propagates across interfaces within the hydrological system and across disciplinary boundaries. In particular, the expansion of the human footprint raises a new set of questions related to human interactions with nature and water cycle feedbacks in the context of complex water management problems. We hope that this reflection and synthesis of the 23 unsolved problems in hydrology will help guide research efforts for some years to come.

Abstract A critical analysis of the present situation on the global water resources assessment is made. Basic data and methodological approaches used by the author for the assessment and prediction of water resources, water use and water availability on the global scale are briefly described. On the basis of data generalization of the world hydrological network new data are given on the dynamics of renewable water resources of the continents, physiographic and economic regions, selected countries as well as on the river water inflow to the world ocean. The results of the assessments for the 20th century and for the future before 2010–2025 on the water supply for municipal, industrial and agricultural needs as well as an additional evaporation from reservoirs are presented. Loads on water resources and water availability depending on socio-economic and phisiographic factors are analyzed; regions of water scarcity and water resources deficit are discovered. Possible ways of water supply improvement and elim...

(2008). The implications of projected climate change for freshwater resources and their management. Hydrological Sciences Journal: Vol. 53, No. 1, pp. 3-10.

Soil moisture is an important variable in the climate system. Understanding and predicting variations of surface temperature, drought, and flood depend critically on knowledge of soil moisture variations, as do impacts of climate change and weather forecasting. An observational dataset of actual in situ measurements is crucial for climatological analysis, for model development and evaluation, and as ground truth for remote sensing. To that end, the Global Soil Moisture Data Bank, a Web site (http://climate.envsci.rutgers.edu/soil_moisture) dedicated to collection, dissemination, and analysis of soil moisture data from around the globe, is described. The data bank currently has soil moisture observations for over 600 stations from a large variety of global climates, including the former Soviet Union, China, Mongolia, India, and the United States. Most of the data are in situ gravimetric observations of soil moisture; all extend for at least 6 years and most for more than 15 years. Most of the stations have grass vegetation, and some are agricultural. The observations have been used to examine the temporal and spatial scales of soil moisture variations, to evaluate Atmospheric Model Intercomparison Project, Project for Intercomparison of Land-Surface Parameterization Schemes, and Global Soil Wetness Project simulations of soil moisture, for remote sensing of soil moisture, for designing new soil moisture observational networks, and to examine soil moisture trends. For the top 1-m soil layers, the temporal scale of soil moisture variation at all midlatitude sites is 1.5 to 2 months and the spatial scale is about 500 km. Land surface models, in general, do not capture the observed soil moisture variations when forced with either model-generated or observed meteorology. In contrast to predictions of summer desiccation with increasing temperatures, for the stations with the longest records summer soil moisture in the top 1 m has increased while temperatures have risen. The increasing trend in precipitation more than compensated for the enhanced evaporation.

Abstract Terrestrial hydrology is central to the Arctic system and its freshwater circulation. Water transport and water constituents vary, however, across a very diverse geography. In this paper, which is a component of the Arctic Freshwater Synthesis, we review the central freshwater processes in the terrestrial Arctic drainage and how they function and change across seven hydrophysiographical regions (Arctic tundra, boreal plains, shield, mountains, grasslands, glaciers/ice caps, and wetlands). We also highlight links between terrestrial hydrology and other components of the Arctic freshwater system. In terms of key processes, snow cover extent and duration is generally decreasing on a pan‐Arctic scale, but snow depth is likely to increase in the Arctic tundra. Evapotranspiration will likely increase overall, but as it is coupled to shifts in landscape characteristics, regional changes are uncertain and may vary over time. Streamflow will generally increase with increasing precipitation, but high and low flows may decrease in some regions. Continued permafrost thaw will trigger hydrological change in multiple ways, particularly through increasing connectivity between groundwater and surface water and changing water storage in lakes and soils, which will influence exchange of moisture with the atmosphere. Other effects of hydrological change include increased risks to infrastructure and water resource planning, ecosystem shifts, and growing flows of water, nutrients, sediment, and carbon to the ocean. Coordinated efforts in monitoring, modeling, and processing studies at various scales are required to improve the understanding of change, in particular at the interfaces between hydrology, atmosphere, ecology, resources, and oceans.

Precipitation measurements in the United States (as well as all other countries) are adversely affected by the gauge undercatch bias of point precipitation measurements. When these measurements are used to obtain areal averages, particularly in mountainous terrain, additional biases may be introduced because most stations are at lower elevations in exposed sites.

The biases and large-scale inhomogeneities in the time series of measured precipitation and snowfall over the United States and Canada are discussed and analyzed. The spatial statistical characteristics of monthly and annual snowfall and total precipitation are investigated and parameterized. After adjustments and selection of the “best” network, reliable “first guess” estimates of North American snowfall and precipitation are obtained. Century-long time series of unbiased annual precipitation over the regions to the south of 55°N and 40-year time series of unbiased area-averaged annual precipitation and snowfall for all of North America are developed. The analysis of their trends shows the following. 1) During the last 100 years, annual precipitation has increased in southern Canada (south of 55°N) by 13% and in the contiguous United States by 4%; however, the main domain of this century-scale precipitation increase is eastern Canada and adjacent to it northern regions of the United States. 2) Up to a 20% increase has occurred in annual snowfall and rainfall during the last four decades in Canada north of 55°N. The relationships between century-long precipitation time series over North America with Northern Hemisphere surface air temperature and the South Oscillation index (SOI) are investigated. It is shown that ENSO (negative anomaly of SOI) is usually accompanied by an increase of precipitation whenever it affects the United States (especially in the southwestern region of the country).

The standard 8" nonrecording precipitation gauge has been used historically by the National Weather Service (NWS) as the official precipitation measurement instrument of the U.S. climate station network. From 1986 to 1992, the accuracy and performance of this gauge (unshielded or with an Alter shield) were evaluated during the WMO Solid Precipitation Measurement Intercomparison at three stations in the United States and Russia, representing a variety of climate, terrain, and exposure. The double-fence intercomparison reference (DFIR) was the reference standard used at all the intercomparison stations in the Intercomparison project. The Intercomparison data collected at different sites are compatible with respect to the catch ratio (gauge measured/DFIR) for the same gauges, when compared using wind speed at the height of gauge orifice during the observation period. The effects of environmental factors, such as wind speed and temperature, on the gauge catch were investigated. Wind speed was found to be the most important factor determining gauge catch when precipitation was classified into snow, mixed, and rain. The regression functions of the catch ratio versus wind speed at the gauge height on a daily time step were derived for various types of precipitation. Independent checks of the equations have been conducted at these intercomparison stations and good agreement was obtained. Application of the correction procedures for wind, wetting loss, and trace amounts was made on a daily basis at Barrow, Alaska, for 1982 and 1983, and, on average, the gauge-measured precipitation was increased by 20% for rain and 90% for snow.

Snow is a critically important and rapidly changing feature of the Arctic. However, snow-cover and snowpack conditions change through time pose challenges for measuring and prediction of snow. Plausible scenarios of how Arctic snow cover will respond to changing Arctic climate are important for impact assessments and adaptation strategies. Although much progress has been made in understanding and predicting snow-cover changes and their multiple consequences, many uncertainties remain. In this paper, we review advances in snow monitoring and modelling, and the impact of snow changes on ecosystems and society in Arctic regions. Interdisciplinary activities are required to resolve the current limitations on measuring and modelling snow characteristics through the cold season and at different spatial scales to assure human well-being, economic stability, and improve the ability to predict manage and adapt to natural hazards in the Arctic region.

New data are presented on the changes of mean global surface air temperature and annual precipitation over extratropical continents of the Northern Hemisphere. Global warming occurred during the last century with a mean trend of 0.5°C/100 years. It is shown that for the same period the annual precipitation over the land in the 35°–70°N zone increased by 6%. The observed variations of precipitation coincide with the results of general circulation modeling of doubled CO2 equilibrium climate change by sign but contradict by scale.

Observed decreases in pan evaporation over most of the United States and the former USSR during the post‐WWII period, if interpreted as a decrease in actual evaporation, are at odds with increases in temperature and precipitation over many regions of these two countries. Using parallel observations of actual and pan evaporation at six Russian, one Latvian, and one U.S. experimental sites, we recalibrate trends in pan evaporation to make them more representative of actual evaporation changes. After applying this transformation, pan evaporation time series over southern Russia and most of the United States reveal an increasing trend in actual evaporation during the past forty years.

Abstract There is a drastic geographic imbalance in available global streamflow gauge and catchment property data, with additional large variations in data characteristics. As a result, models calibrated in one region cannot normally be migrated to another without significant modifications. Currently in these regions, non‐transferable machine learning models are habitually trained over small local data sets. Here we show that transfer learning (TL), in the senses of weight initialization and weight freezing, allows long short‐term memory (LSTM) streamflow models that were pretrained over the conterminous United States (CONUS, the source data set) to be transferred to catchments on other continents (the target regions), without the need for extensive catchment attributes available at the target location. We demonstrate this possibility for regions where data are dense (664 basins in Great Britain), moderately dense (49 basins in central Chile), and scarce with only remotely sensed attributes available (5 basins in China). In both China and Chile, the TL models showed significantly elevated performance compared to locally trained models using all basins. The benefits of TL increased with the amount of available data in the source data set, and seemed to be more pronounced with greater physiographic diversity. The benefits from TL were greater than from pretraining LSTM using the outputs from an uncalibrated hydrologic model. These results suggest hydrologic data around the world have commonalities which could be leveraged by deep learning, and synergies can be had with a simple modification of the current workflows, greatly expanding the reach of existing big data. Finally, this work diversified existing global streamflow benchmarks.

Quantification of sediment fluxes from rivers is fundamental to understanding land‐ocean linkages in the Arctic. Numerous publications have focused on this subject over the past century, yet assessments of temporal trends are scarce and consensus on contemporary fluxes is lacking. Published estimates vary widely, but often provide little accessory information needed to interpret the differences. We present a pan‐arctic synthesis of sediment flux from 19 arctic rivers, primarily focusing on contributions from the eight largest ones. For this synthesis, historical records and recent unpublished data were compiled from Russian, Canadian, and United States sources. Evaluation of these data revealed no long‐term trends in sediment flux, but did show stepwise changes in the historical records of two of the rivers. In some cases, old values that do not reflect contemporary fluxes are still being reported, while in other cases, typographical errors have been propagated into the recent literature. Most of the discrepancy among published estimates, however, can be explained by differences in years of records examined and gauging stations used. Variations in sediment flux from year to year in arctic rivers are large, so estimates based on relatively few years can differ substantially. To determine best contemporary estimates of sediment flux for the eight largest arctic rivers, we used a combination of newly available data, historical records, and literature values. These estimates contribute to our understanding of carbon, nutrient, and contaminant transport to the Arctic Ocean and provide a baseline for detecting future anthropogenic or natural change in the Arctic.

Some early and current general circulation models (GCMs) use bucket models for soil hydrology calculations. More recently, the Simple Biosphere Model (SiB) was developed to incorporate the effects of vegetation on fluxes of moisture, momentum, and energy at the earth's surface into soil hydrology models. In this study, a Simplified SiB (SSiB) soil hydrology model and a 15-cm bucket model are forced by observed meteorological and actinometric data every 3 h for 6-yr simulations at the six stations. The model calculations of soil moisture are compared to observations of soil moisture, literally "ground truth', snow cover, surface albedo, and net radiation, and with each other. While producing similar soil moisture simulations, the models produce very different surface latent and sensible heat fluxes, which would have large effects on GCM simulations. -from Authors

The permafrost regions occupy about 25% of the Northern Hemisphere's terrestrial surface, and more than 60% of that of Russia. Warming, thawing, and degradation of permafrost have been observed in many locations in recent decades and are likely to accelerate in the future as a result of climatic change. Changes of permafrost have important implications for natural systems, humans, and the economy of the northern lands. Results from mathematical modeling indicate that by the mid-21st century, near-surface permafrost in the Northern Hemisphere may shrink by 15%-30%, leading to complete thawing of the frozen ground in the upper few meters, while elsewhere the depth of seasonal thawing may increase on average by 15%-25%, and by 50% or more in the northernmost locations. Such changes may shift the balance between the uptake and release of carbon in tundra and facilitate emission of greenhouse gases from the carbon-rich Arctic wetlands. Serious public concerns are associated with the effects that thawing permafrost may have on the infrastructure constructed on it. Climate-induced changes of permafrost properties are potentially detrimental to almost all structures in northern lands, and may render many of them unusable. Degradation of permafrost and ground settlement due to thermokarst may lead to dramatic distortions of terrain and to changes in hydrology and vegetation, and may lead ultimately to transformation of existing landforms. Recent studies indicate that nonclimatic factors, such as changes in vegetation and hydrology, may largely govern the response of permafrost to global warming. More studies are needed to better understand and quantify the effects of multiple factors in the changing northern environment.

Analyses of historical ocean temperature data at a depth of 125 meters in the North Atlantic Ocean indicate that from 1950 to 1990 the subtropical and subarctic gyres exhibited linear trends that were opposite in phase. In addition, multivariate analyses of yearly mean temperature anomaly fields between 20 degrees N and 70 degrees N in the North Atlantic show a characteristic space-time temperature oscillation from 1947 to 1990. A quasidecadal oscillation, first identified at Ocean Weather Station C, is part of a basin-wide feature. Gyre and basin-scale variations such as these provide the observational basis for climate diagnostic and modeling studies.

The Project for the Intercomparison of Land-Surface Parameterization Schemes (PILPS) aims to improve understanding and modeling of land surface processes. PILPS phase 2(d) uses a set of meteorological and hydrological data spanning 18 yr (1966-83) from a grassland catchment at the Valdai water-balance research site in Russia. A suite of stand-alone simulations is performed by 21 land surface schemes (LSSs) to explore the LSSs' sensitivity to downward longwave radiative forcing, timescales of simulated hydrologic variability, and biases resulting from single-year simulations that use recursive spinup. These simulations are the first in PILPS to investigate the performance of LSSs at a site with a well-defined seasonal snow cover and frozen soil. Considerable model scatter for the control simulations exists. However, nearly all the LSS scatter in simulated root-zone soil moisture is contained within the spatial variability observed inside the catchment. In addition, all models show a considerable sensitivity to longwave forcing for the simulation of the snowpack, which during the spring melt affects runoff, meltwater infiltration, and subsequent evapotranspiration. A greater sensitivity of the ablation, compared to the accumulation, of the winter snowpack to the choice of snow parameterization is found. Sensitivity simulations starting at prescribed conditions with no spinup demonstrate that the treatment of frozen soil (moisture) processes can affect the long-term variability of the models. The single-year recursive runs show large biases, compared to the corresponding year of the control run, that can persist through the entire year and underscore the importance of performing multiyear simulations.

A unique dataset of soil moisture in the upper 1-m soil layer at sites with natural plant cover in the Soviet Union is compared to simulations of soil moisture for the present climate by the Geophysical Fluid Dynamics Laboratory, Oregon State University, and United Kingdom Meteorological Office general circulation models. It is found that the present-day soil moisture regime is not well simulated by these models. Delworth and Manabe's hypothesis that the spectrum of time variations in soil moisture in the upper 1-m layer corresponds to a first-order Markov process with a decay time of the correlation function equal to the ratio of field capacity to potential evapotranspiration is empirically confirmed with this dataset. Analysis of measurement data over the 1972–1985 period reveals that the long-term trends of soil moisture north of 50°N are mainly due to increasing precipitation during this period of the same scale (1–3 cm/10 yr). The seasonal structure does not correspond to the “summer continental desiccation” scenario predicted by some climate models.