Search Results
You are looking at 1 - 10 of 16 items for :
- Author or Editor: Bart Nijssen x
- Journal of Hydrometeorology x
- Refine by Access: All Content x
Abstract
Data assimilation (DA) techniques have been widely applied to assimilate satellite-based soil moisture (SM) measurements into hydrologic models to improve streamflow simulations. However, past studies have reached mixed conclusions regarding the degree of runoff improvement achieved via SM state updating. In this study, a synthetic diagnostic framework is designed to 1) decompose the random error components in a hydrologic simulation, 2) quantify the error terms that originate from SM states, and 3) assess the effectiveness of SM DA to correct these random errors. The general framework is illustrated through a case study in which surface Soil Moisture Active Passive (SMAP) data are assimilated into a large-scale land surface model in the Arkansas–Red River basin. The case study includes systematic error in the simulated streamflow that imposes a first-order limit on DA performance. In addition, about 60% of the random runoff error originates directly from rainfall and cannot be corrected by SM DA. In particular, fast-response runoff dominates in much of the basin but is relatively unresponsive to state updating. Slow-response runoff is strongly controlled by the bottom-layer SM and therefore only modestly improved via the assimilation of surface measurements. Combined, the total runoff improvement in the synthetic analysis is small (<10% over the basin). Improvements in the real SMAP-assimilated case are further limited due to systematic error and other factors such as inaccurate error assumptions and SMAP rescaling. Findings from the diagnostic framework suggest that SM DA alone is insufficient to substantially improve streamflow estimates in large basins.
Abstract
Data assimilation (DA) techniques have been widely applied to assimilate satellite-based soil moisture (SM) measurements into hydrologic models to improve streamflow simulations. However, past studies have reached mixed conclusions regarding the degree of runoff improvement achieved via SM state updating. In this study, a synthetic diagnostic framework is designed to 1) decompose the random error components in a hydrologic simulation, 2) quantify the error terms that originate from SM states, and 3) assess the effectiveness of SM DA to correct these random errors. The general framework is illustrated through a case study in which surface Soil Moisture Active Passive (SMAP) data are assimilated into a large-scale land surface model in the Arkansas–Red River basin. The case study includes systematic error in the simulated streamflow that imposes a first-order limit on DA performance. In addition, about 60% of the random runoff error originates directly from rainfall and cannot be corrected by SM DA. In particular, fast-response runoff dominates in much of the basin but is relatively unresponsive to state updating. Slow-response runoff is strongly controlled by the bottom-layer SM and therefore only modestly improved via the assimilation of surface measurements. Combined, the total runoff improvement in the synthetic analysis is small (<10% over the basin). Improvements in the real SMAP-assimilated case are further limited due to systematic error and other factors such as inaccurate error assumptions and SMAP rescaling. Findings from the diagnostic framework suggest that SM DA alone is insufficient to substantially improve streamflow estimates in large basins.
Abstract
The severity–area–duration (SAD) method is used in conjunction with the Variable Infiltration Capacity model (VIC) to identify the major historical total moisture (TM; soil moisture plus snow water equivalent) droughts over the Pacific Northwest region, defined as the Columbia River basin and the region’s coastal drainages, for the period 1920–2013. The motivation is to understand how droughts identified using TM (a measure similar to that used in the U.S. Drought Monitor) relate to sector-specific drought measures that are more relevant to users. It is found that most of the SAD space is dominated by an extended drought period during the 1930s, although the most severe shorter droughts are in the 1970s (1976–78) and early 2000s (2000–04). The impact of the three severe TM droughts that dominate most of the SAD space are explored in terms of sector-specific measures relevant to dryland and irrigated agriculture, hydropower generation, municipal water supply, and recreation. It is found that in many cases the most severe droughts identified using the SAD method also appear among the most severe sector-specific droughts; however, there are important exceptions. Two types of inconsistencies are examined and the nature of the conditions that give rise to them are explored.
Abstract
The severity–area–duration (SAD) method is used in conjunction with the Variable Infiltration Capacity model (VIC) to identify the major historical total moisture (TM; soil moisture plus snow water equivalent) droughts over the Pacific Northwest region, defined as the Columbia River basin and the region’s coastal drainages, for the period 1920–2013. The motivation is to understand how droughts identified using TM (a measure similar to that used in the U.S. Drought Monitor) relate to sector-specific drought measures that are more relevant to users. It is found that most of the SAD space is dominated by an extended drought period during the 1930s, although the most severe shorter droughts are in the 1970s (1976–78) and early 2000s (2000–04). The impact of the three severe TM droughts that dominate most of the SAD space are explored in terms of sector-specific measures relevant to dryland and irrigated agriculture, hydropower generation, municipal water supply, and recreation. It is found that in many cases the most severe droughts identified using the SAD method also appear among the most severe sector-specific droughts; however, there are important exceptions. Two types of inconsistencies are examined and the nature of the conditions that give rise to them are explored.
Abstract
Water resources planning often uses streamflow predictions made by hydrologic models. These simulated predictions have systematic errors that limit their usefulness as input to water management models. To account for these errors, streamflow predictions are bias corrected through statistical methods that adjust model predictions based on comparisons to reference datasets (such as observed streamflow). Existing bias correction methods have several shortcomings when used to correct spatially distributed streamflow predictions. First, existing bias correction methods destroy the spatiotemporal consistency of the streamflow predictions when these methods are applied independently at multiple sites across a river network. Second, bias correction techniques are usually built on time-invariant mappings between reference and simulated streamflow without accounting for the processes that underpin the systematic errors. We describe improved bias correction techniques that account for the river network topology and allow for corrections that account for other processes. Further, we present a workflow that allows the user to select whether to apply these techniques separately or in conjunction. We evaluate four different bias correction methods implemented with our workflow in the Yakima River basin in the northwestern United States. We find that all four methods reduce systematic bias in the simulated streamflow. The spatially consistent bias correction methods produce spatially distributed streamflow as well as bias-corrected incremental streamflow, which is suitable for input to water management models. We demonstrate how the spatially consistent method avoids creating flows that are inconsistent between upstream and downstream locations, while performing similar to existing methods. We also find that conditioning on daily minimum temperature, which we use as a proxy for snowmelt processes, improves the timing of the corrected streamflow.
Significance Statement
To make streamflow predictions from hydrologic models more informative and useful for water resources management they are often postprocessed by a statistical procedure known as bias correction. In this work we develop and demonstrate bias correction techniques that are specifically tailored to streamflow prediction. These new techniques will make modeled streamflow predictions more useful in complex river systems undergoing climate change.
Abstract
Water resources planning often uses streamflow predictions made by hydrologic models. These simulated predictions have systematic errors that limit their usefulness as input to water management models. To account for these errors, streamflow predictions are bias corrected through statistical methods that adjust model predictions based on comparisons to reference datasets (such as observed streamflow). Existing bias correction methods have several shortcomings when used to correct spatially distributed streamflow predictions. First, existing bias correction methods destroy the spatiotemporal consistency of the streamflow predictions when these methods are applied independently at multiple sites across a river network. Second, bias correction techniques are usually built on time-invariant mappings between reference and simulated streamflow without accounting for the processes that underpin the systematic errors. We describe improved bias correction techniques that account for the river network topology and allow for corrections that account for other processes. Further, we present a workflow that allows the user to select whether to apply these techniques separately or in conjunction. We evaluate four different bias correction methods implemented with our workflow in the Yakima River basin in the northwestern United States. We find that all four methods reduce systematic bias in the simulated streamflow. The spatially consistent bias correction methods produce spatially distributed streamflow as well as bias-corrected incremental streamflow, which is suitable for input to water management models. We demonstrate how the spatially consistent method avoids creating flows that are inconsistent between upstream and downstream locations, while performing similar to existing methods. We also find that conditioning on daily minimum temperature, which we use as a proxy for snowmelt processes, improves the timing of the corrected streamflow.
Significance Statement
To make streamflow predictions from hydrologic models more informative and useful for water resources management they are often postprocessed by a statistical procedure known as bias correction. In this work we develop and demonstrate bias correction techniques that are specifically tailored to streamflow prediction. These new techniques will make modeled streamflow predictions more useful in complex river systems undergoing climate change.
Abstract
The Tropical Rainfall Measuring Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA) near-real-time (RT) data are considered less accurate than the TMPA research quality (RP) data because of the simplified data processing algorithm and the lack of gauge adjustments. However, for near-real-time hydrological applications, such as drought nowcasting, the RT data must play a key role given latency considerations and consistency is essential with products like RP, which have a long-term climatology. The authors used a bivariate test to examine the consistency between the monthly RT and RP precipitation estimates for 12 yr (2000–12) and found that, for over 75% of land cells globally, RT and RP were statistically consistent at 0.05 significance level. The inconsistent grid cells are spatially clustered in western North America, northern South America, central Africa, and most of Australia. The authors also show that RT generally increases with time relative to RP in northern South America and western Australia, while in western North America and eastern Australia, RT decreases relative to RP. In other areas such as the eastern part of North America, Eurasia, and southern part of the South America, the RT data are statistically consistent with the RP data and are appropriate for global- or macroscale hydrological applications.
Abstract
The Tropical Rainfall Measuring Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA) near-real-time (RT) data are considered less accurate than the TMPA research quality (RP) data because of the simplified data processing algorithm and the lack of gauge adjustments. However, for near-real-time hydrological applications, such as drought nowcasting, the RT data must play a key role given latency considerations and consistency is essential with products like RP, which have a long-term climatology. The authors used a bivariate test to examine the consistency between the monthly RT and RP precipitation estimates for 12 yr (2000–12) and found that, for over 75% of land cells globally, RT and RP were statistically consistent at 0.05 significance level. The inconsistent grid cells are spatially clustered in western North America, northern South America, central Africa, and most of Australia. The authors also show that RT generally increases with time relative to RP in northern South America and western Australia, while in western North America and eastern Australia, RT decreases relative to RP. In other areas such as the eastern part of North America, Eurasia, and southern part of the South America, the RT data are statistically consistent with the RP data and are appropriate for global- or macroscale hydrological applications.
Abstract
Man-made reservoirs play a key role in the terrestrial water system. They alter water fluxes at the land surface and impact surface water storage through water management regulations for diverse purposes such as irrigation, municipal water supply, hydropower generation, and flood control. Although most developed countries have established sophisticated observing systems for many variables in the land surface water cycle, long-term and consistent records of reservoir storage are much more limited and not always shared. Furthermore, most land surface hydrological models do not represent the effects of water management activities. Here, the contribution of reservoirs to seasonal water storage variations is investigated using a large-scale water management model to simulate the effects of reservoir management at basin and continental scales. The model was run from 1948 to 2010 at a spatial resolution of 0.25° latitude–longitude. A total of 166 of the largest reservoirs in the world with a total capacity of about 3900 km3 (nearly 60% of the globally integrated reservoir capacity) were simulated. The global reservoir storage time series reflects the massive expansion of global reservoir capacity; over 30 000 reservoirs have been constructed during the past half century, with a mean absolute interannual storage variation of 89 km3. The results indicate that the average reservoir-induced seasonal storage variation is nearly 700 km3 or about 10% of the global reservoir storage. For some river basins, such as the Yellow River, seasonal reservoir storage variations can be as large as 72% of combined snow water equivalent and soil moisture storage.
Abstract
Man-made reservoirs play a key role in the terrestrial water system. They alter water fluxes at the land surface and impact surface water storage through water management regulations for diverse purposes such as irrigation, municipal water supply, hydropower generation, and flood control. Although most developed countries have established sophisticated observing systems for many variables in the land surface water cycle, long-term and consistent records of reservoir storage are much more limited and not always shared. Furthermore, most land surface hydrological models do not represent the effects of water management activities. Here, the contribution of reservoirs to seasonal water storage variations is investigated using a large-scale water management model to simulate the effects of reservoir management at basin and continental scales. The model was run from 1948 to 2010 at a spatial resolution of 0.25° latitude–longitude. A total of 166 of the largest reservoirs in the world with a total capacity of about 3900 km3 (nearly 60% of the globally integrated reservoir capacity) were simulated. The global reservoir storage time series reflects the massive expansion of global reservoir capacity; over 30 000 reservoirs have been constructed during the past half century, with a mean absolute interannual storage variation of 89 km3. The results indicate that the average reservoir-induced seasonal storage variation is nearly 700 km3 or about 10% of the global reservoir storage. For some river basins, such as the Yellow River, seasonal reservoir storage variations can be as large as 72% of combined snow water equivalent and soil moisture storage.
Abstract
Atmospheric rivers (ARs) are narrow, elongated corridors of high water vapor content transported from tropical and/or extratropical cyclones. We characterize precipitation and snow water equivalent associated with ARs intersecting the western U.S. coast during the cold season (November– March) of water years 1949–2015. For each AR landfalling date, we retrieved the precipitation and relevant hydrometeorological variables from dynamically downscaled atmospheric reanalyses (NCEP–NCAR) using the WRF mesoscale numerical weather prediction model. Landfalling ARs resulted in higher precipitation amounts throughout the domain than did non-AR storms. ARs contributed the most extreme precipitation events during January and February. Daily snow water equivalent (SWE) changes during ARs indicate that high positive net snow accumulation conditions accompany ARs in December, January, and February. We also assess the historical impact of AR storm duration and precipitation frequency by considering the precipitation depth estimated during a 72-h window bounding the AR landfall date. More extreme precipitation amounts are produced by ARs in the South Cascades and Sierra Nevada ranges compared with ARs with landfall farther north. Most AR extreme precipitation events (and lower SWE accumulations) are produced during warm AR dates, especially toward the northern end of our domain. Analysis of ARs during dry and wet years reveals that ARs during wet years are more frequent and produce heavier precipitation and snow accumulation as compared with dry years. Such conditions are evident in drought events that are associated with a reduced frequency of ARs.
Abstract
Atmospheric rivers (ARs) are narrow, elongated corridors of high water vapor content transported from tropical and/or extratropical cyclones. We characterize precipitation and snow water equivalent associated with ARs intersecting the western U.S. coast during the cold season (November– March) of water years 1949–2015. For each AR landfalling date, we retrieved the precipitation and relevant hydrometeorological variables from dynamically downscaled atmospheric reanalyses (NCEP–NCAR) using the WRF mesoscale numerical weather prediction model. Landfalling ARs resulted in higher precipitation amounts throughout the domain than did non-AR storms. ARs contributed the most extreme precipitation events during January and February. Daily snow water equivalent (SWE) changes during ARs indicate that high positive net snow accumulation conditions accompany ARs in December, January, and February. We also assess the historical impact of AR storm duration and precipitation frequency by considering the precipitation depth estimated during a 72-h window bounding the AR landfall date. More extreme precipitation amounts are produced by ARs in the South Cascades and Sierra Nevada ranges compared with ARs with landfall farther north. Most AR extreme precipitation events (and lower SWE accumulations) are produced during warm AR dates, especially toward the northern end of our domain. Analysis of ARs during dry and wet years reveals that ARs during wet years are more frequent and produce heavier precipitation and snow accumulation as compared with dry years. Such conditions are evident in drought events that are associated with a reduced frequency of ARs.
Abstract
The concepts of model benchmarking, model agility, and large-sample hydrology are becoming more prevalent in hydrologic and land surface modeling. As modeling systems become more sophisticated, these concepts have the ability to help improve modeling capabilities and understanding. In this paper, their utility is demonstrated with an application of the physically based Variable Infiltration Capacity model (VIC). The authors implement VIC for a sample of 531 basins across the contiguous United States, incrementally increase model agility, and perform comparisons to a benchmark. The use of a large-sample set allows for statistically robust comparisons and subcategorization across hydroclimate conditions. Our benchmark is a calibrated, time-stepping, conceptual hydrologic model. This model is constrained by physical relationships such as the water balance, and it complements purely statistical benchmarks due to the increased physical realism and permits physically motivated benchmarking using metrics that relate one variable to another (e.g., runoff ratio). The authors find that increasing model agility along the parameter dimension, as measured by the number of model parameters available for calibration, does increase model performance for calibration and validation periods relative to less agile implementations. However, as agility increases, transferability decreases, even for a complex model such as VIC. The benchmark outperforms VIC in even the most agile case when evaluated across the entire basin set. However, VIC meets or exceeds benchmark performance in basins with high runoff ratios (greater than ~0.8), highlighting the ability of large-sample comparative hydrology to identify hydroclimatic performance variations.
Abstract
The concepts of model benchmarking, model agility, and large-sample hydrology are becoming more prevalent in hydrologic and land surface modeling. As modeling systems become more sophisticated, these concepts have the ability to help improve modeling capabilities and understanding. In this paper, their utility is demonstrated with an application of the physically based Variable Infiltration Capacity model (VIC). The authors implement VIC for a sample of 531 basins across the contiguous United States, incrementally increase model agility, and perform comparisons to a benchmark. The use of a large-sample set allows for statistically robust comparisons and subcategorization across hydroclimate conditions. Our benchmark is a calibrated, time-stepping, conceptual hydrologic model. This model is constrained by physical relationships such as the water balance, and it complements purely statistical benchmarks due to the increased physical realism and permits physically motivated benchmarking using metrics that relate one variable to another (e.g., runoff ratio). The authors find that increasing model agility along the parameter dimension, as measured by the number of model parameters available for calibration, does increase model performance for calibration and validation periods relative to less agile implementations. However, as agility increases, transferability decreases, even for a complex model such as VIC. The benchmark outperforms VIC in even the most agile case when evaluated across the entire basin set. However, VIC meets or exceeds benchmark performance in basins with high runoff ratios (greater than ~0.8), highlighting the ability of large-sample comparative hydrology to identify hydroclimatic performance variations.
Abstract
We propose a conceptual and theoretical foundation for information-based model benchmarking and process diagnostics that provides diagnostic insight into model performance and model realism. We benchmark against a bounded estimate of the information contained in model inputs to obtain a bounded estimate of information lost due to model error, and we perform process-level diagnostics by taking differences between modeled versus observed transfer entropy networks. We use this methodology to reanalyze the recent Protocol for the Analysis of Land Surface Models (PALS) Land Surface Model Benchmarking Evaluation Project (PLUMBER) land model intercomparison project that includes the following models: CABLE, CH-TESSEL, COLA-SSiB, ISBA-SURFEX, JULES, Mosaic, Noah, and ORCHIDEE. We report that these models (i) use only roughly half of the information available from meteorological inputs about observed surface energy fluxes, (ii) do not use all information from meteorological inputs about long-term Budyko-type water balances, (iii) do not capture spatial heterogeneities in surface processes, and (iv) all suffer from similar patterns of process-level structural error. Because the PLUMBER intercomparison project did not report model parameter values, it is impossible to know whether process-level error patterns are due to model structural error or parameter error, although our proposed information-theoretic methodology could distinguish between these two issues if parameter values were reported. We conclude that there is room for significant improvement to the current generation of land models and their parameters. We also suggest two simple guidelines to make future community-wide model evaluation and intercomparison experiments more informative.
Abstract
We propose a conceptual and theoretical foundation for information-based model benchmarking and process diagnostics that provides diagnostic insight into model performance and model realism. We benchmark against a bounded estimate of the information contained in model inputs to obtain a bounded estimate of information lost due to model error, and we perform process-level diagnostics by taking differences between modeled versus observed transfer entropy networks. We use this methodology to reanalyze the recent Protocol for the Analysis of Land Surface Models (PALS) Land Surface Model Benchmarking Evaluation Project (PLUMBER) land model intercomparison project that includes the following models: CABLE, CH-TESSEL, COLA-SSiB, ISBA-SURFEX, JULES, Mosaic, Noah, and ORCHIDEE. We report that these models (i) use only roughly half of the information available from meteorological inputs about observed surface energy fluxes, (ii) do not use all information from meteorological inputs about long-term Budyko-type water balances, (iii) do not capture spatial heterogeneities in surface processes, and (iv) all suffer from similar patterns of process-level structural error. Because the PLUMBER intercomparison project did not report model parameter values, it is impossible to know whether process-level error patterns are due to model structural error or parameter error, although our proposed information-theoretic methodology could distinguish between these two issues if parameter values were reported. We conclude that there is room for significant improvement to the current generation of land models and their parameters. We also suggest two simple guidelines to make future community-wide model evaluation and intercomparison experiments more informative.
Abstract
The diurnal cycle of solar radiation represents the strongest energetic forcing and dominates the exchange of heat and mass of the land surface with the atmosphere. This diurnal heat redistribution represents a core of land–atmosphere coupling that should be accurately represented in land surface models (LSMs), which are critical parts of weather and climate models. We employ a diagnostic model evaluation approach using a signature-based metric that describes the diurnal variation of heat fluxes. The metric is obtained by decomposing the diurnal variation of surface heat fluxes into their direct response and the phase lag to incoming solar radiation. We employ the output of 13 different LSMs driven with meteorological forcing of 20 FLUXNET sites (PLUMBER dataset). All LSMs show a poor representation of the evaporative fraction and thus the diurnal magnitude of the sensible and latent heat flux under cloud-free conditions. In addition, we find that the diurnal phase of both heat fluxes is poorly represented. The best performing model only reproduces 33% of the evaluated evaporative conditions across the sites. The poor performance of the diurnal cycle of turbulent heat exchange appears to be linked to how models solve for the surface energy balance and redistribute heat into the subsurface. We conclude that a systematic evaluation of diurnal signatures is likely to help to improve the simulated diurnal cycle, better represent land–atmosphere interactions, and therefore improve simulations of the near-surface climate.
Abstract
The diurnal cycle of solar radiation represents the strongest energetic forcing and dominates the exchange of heat and mass of the land surface with the atmosphere. This diurnal heat redistribution represents a core of land–atmosphere coupling that should be accurately represented in land surface models (LSMs), which are critical parts of weather and climate models. We employ a diagnostic model evaluation approach using a signature-based metric that describes the diurnal variation of heat fluxes. The metric is obtained by decomposing the diurnal variation of surface heat fluxes into their direct response and the phase lag to incoming solar radiation. We employ the output of 13 different LSMs driven with meteorological forcing of 20 FLUXNET sites (PLUMBER dataset). All LSMs show a poor representation of the evaporative fraction and thus the diurnal magnitude of the sensible and latent heat flux under cloud-free conditions. In addition, we find that the diurnal phase of both heat fluxes is poorly represented. The best performing model only reproduces 33% of the evaluated evaporative conditions across the sites. The poor performance of the diurnal cycle of turbulent heat exchange appears to be linked to how models solve for the surface energy balance and redistribute heat into the subsurface. We conclude that a systematic evaluation of diurnal signatures is likely to help to improve the simulated diurnal cycle, better represent land–atmosphere interactions, and therefore improve simulations of the near-surface climate.
Abstract
A common term in the continental and oceanic components of the global water cycle is freshwater discharge to the oceans. Many estimates of the annual average global discharge have been made over the past 100 yr with a surprisingly wide range. As more observations have become available and continental-scale land surface model simulations of runoff have improved, these past estimates are cast in a somewhat different light. In this paper, a combination of observations from 839 river gauging stations near the outlets of large river basins is used in combination with simulated runoff fields from two implementations of the Variable Infiltration Capacity land surface model to estimate continental runoff into the world’s oceans from 1950 to 2008. The gauges used account for ~58% of continental areas draining to the ocean worldwide, excluding Greenland and Antarctica. This study estimates that flows to the world’s oceans globally are 44 200 (±2660) km3 yr−1 (9% from Africa, 37% from Eurasia, 30% from South America, 16% from North America, and 8% from Australia–Oceania). These estimates are generally higher than previous estimates, with the largest differences in South America and Australia–Oceania. Given that roughly 42% of ocean-draining continental areas are ungauged, it is not surprising that estimates are sensitive to the land surface and hydrologic model (LSM) used, even with a correction applied to adjust for model bias. The results show that more and better in situ streamflow measurements would be most useful in reducing uncertainties, in particular in the southern tip of South America, the islands of Oceania, and central Africa.
Abstract
A common term in the continental and oceanic components of the global water cycle is freshwater discharge to the oceans. Many estimates of the annual average global discharge have been made over the past 100 yr with a surprisingly wide range. As more observations have become available and continental-scale land surface model simulations of runoff have improved, these past estimates are cast in a somewhat different light. In this paper, a combination of observations from 839 river gauging stations near the outlets of large river basins is used in combination with simulated runoff fields from two implementations of the Variable Infiltration Capacity land surface model to estimate continental runoff into the world’s oceans from 1950 to 2008. The gauges used account for ~58% of continental areas draining to the ocean worldwide, excluding Greenland and Antarctica. This study estimates that flows to the world’s oceans globally are 44 200 (±2660) km3 yr−1 (9% from Africa, 37% from Eurasia, 30% from South America, 16% from North America, and 8% from Australia–Oceania). These estimates are generally higher than previous estimates, with the largest differences in South America and Australia–Oceania. Given that roughly 42% of ocean-draining continental areas are ungauged, it is not surprising that estimates are sensitive to the land surface and hydrologic model (LSM) used, even with a correction applied to adjust for model bias. The results show that more and better in situ streamflow measurements would be most useful in reducing uncertainties, in particular in the southern tip of South America, the islands of Oceania, and central Africa.
