ALBERT

Description: The need for the development of a method for generating an ensemble of rainfall scenarios, which are conditioned on the observed rainfall, and its place in the HYREX programme is discussed. A review of stochastic models for rainfall, and rainfall forecasting techniques, is followed by a justification for the choice of the Modified Turning Bands (MTB) model in this context. This is a stochastic model of rainfall which is continuous over space and time, and which reproduces features of real rainfall fields at four distinct scales: raincells, cluster potential regions, rainbands and the overall outline of a storm at the synoptic scale. The model can be used to produce synthetic data sets, in the same format as data from a radar. An inversion procedure for inferring a construction of the MTB model which generates a given sequence of radar images is described. This procedure is used to generate an ensemble of future rainfall scenarios which are consistent with a currently observed storm. The combination of deterministic modelling at the large scales and stochastic modelling at smaller scales, within the MTB model, makes the system particularly suitable for short-term forecasts. As the lead time increases, so too does the variability across the set of generated scenarios. Keywords: MTB model, space-time rainfall field model, rainfall radar, HYREX, real-time flow forecasting

Description: Global seasonal hydrologic prediction is crucial to mitigating the impacts of droughts and floods, especially in the developing world. Hydrologic predictability at seasonal lead times (i.e., 1–6 months) comes from knowledge of initial hydrologic conditions (IHCs) and seasonal climate forecast skill (FS). In this study we quantify the contributions of two primary components of IHCs – soil moisture and snow water content – and FS (of precipitation and temperature) to seasonal hydrologic predictability globally on a relative basis throughout the year. We do so by conducting two model-based experiments using the variable infiltration capacity (VIC) macroscale hydrology model, one based on ensemble streamflow prediction (ESP) and another based on Reverse-ESP (Rev-ESP), both for a 47 yr re-forecast period (1961–2007). We compare cumulative runoff (CR), soil moisture (SM) and snow water equivalent (SWE) forecasts from each experiment with a VIC model-based reference data set (generated using observed atmospheric forcings) and estimate the ratio of root mean square error (RMSE) of both experiments for each forecast initialization date and lead time, to determine the relative contribution of IHCs and FS to the seasonal hydrologic predictability. We find that in general, the contributions of IHCs to seasonal hydrologic predictability is highest in the arid and snow-dominated climate (high latitude) regions of the Northern Hemisphere during forecast periods starting on 1 January and 1 October. In mid-latitude regions, such as the Western US, the influence of IHCs is greatest during the forecast period starting on 1 April. In the arid and warm temperate dry winter regions of the Southern Hemisphere, the IHCs dominate during forecast periods starting on 1 April and 1 July. In equatorial humid and monsoonal climate regions, the contribution of FS is generally higher than IHCs through most of the year. Based on our findings, we argue that despite the limited FS (mainly for precipitation) better estimates of the IHCs could lead to improvement in the current level of seasonal hydrologic forecast skill over many regions of the globe at least during some parts of the year.

Description: Land evapotranspiration (ET) estimates are available from several global data sets. Here, monthly global land ET synthesis products, merged from these individual data sets over the time periods 1989–1995 (7 yr) and 1989–2005 (17 yr), are presented. The merged synthesis products over the shorter period are based on a total of 40 distinct data sets while those over the longer period are based on a total of 14 data sets. In the individual data sets, ET is derived from satellite and/or in situ observations (diagnostic data sets) or calculated via land-surface models (LSMs) driven with observations-based forcing or output from atmospheric reanalyses. Statistics for four merged synthesis products are provided, one including all data sets and three including only data sets from one category each (diagnostic, LSMs, and reanalyses). The multi-annual variations of ET in the merged synthesis products display realistic responses. They are also consistent with previous findings of a global increase in ET between 1989 and 1997 (0.13 mm yr−2 in our merged product) followed by a significant decrease in this trend (−0.18 mm yr−2), although these trends are relatively small compared to the uncertainty of absolute ET values. The global mean ET from the merged synthesis products (based on all data sets) is 493 mm yr−1 (1.35 mm d−1) for both the 1989–1995 and 1989–2005 products, which is relatively low compared to previously published estimates. We estimate global runoff (precipitation minus ET) to 263 mm yr−1 (34 406 km3 yr−1) for a total land area of 130 922 000 km2. Precipitation, being an important driving factor and input to most simulated ET data sets, presents uncertainties between single data sets as large as those in the ET estimates. In order to reduce uncertainties in current ET products, improving the accuracy of the input variables, especially precipitation, as well as the parameterizations of ET, are crucial.

Description: Key issues involved in converting MTB ensemble forecasts of rainfall into ensemble forecasts of runoff are addressed. The physically-based distributed modelling system, SHETRAN, is parameterised for the Brue catchment, and used to assess the impact of averaging spatially variable MTB rainfall inputs on the accuracy of simulated runoff response. Averaging is found to have little impact for wet antecedent conditions and to lead to some underestimation of peak discharge under dry catchment conditions. The simpler ARNO modelling system is also parameterised for the Brue and SHETRAN and ARNO calibration and validation results are found to be similar. Ensemble forecasts of runoff generated using both SHETRAN and the simpler ARNO modelling system are compared. The ensemble is more spread out with the SHETRAN model, and a likely explanation is that the ARNO model introduces too much smoothing. Nevertheless, the forecasting performance of the simpler model could be adequate for flood warning purposes. Keywords: SHETRAN, ARNO, HYREX, rainfall-runoff model, Brue, real-time flow forecasting

Description: In the present work, we developed a new formulation for the estimation of the soil moisture in the root zone based on the measured value of soil moisture at the surface. The method sheds lights on the relationship between surface and root zone soil moisture and has applications in the use of satellite remote sensing retrievals of soil moisture. It derives from a simplified form of the soil water balance equation and provides a closed form of the relationship between the root zone and the surface soil moisture with a limited number of physically consistent parameters. The approach was first used to interpret soil moisture dynamics at the point scale using soil moisture measurements taken from the African Monsoon Multidisciplinary Analysis (AMMA) database. Thereafter it was also tested over an extended domain using modeled soil moisture data obtained from the North American Land Data Assimilation System (NLDAS). The NLDAS database provides modeled soil moisture data averaged over different depths for the conterminous US covering different climatic and physical conditions. In general, the method performed better than a traditional low pass filter and its results are found to be influenced by rainfall dynamics and also by the observed variance of soil moisture in the lower layer. The limited number of the parameters and their physical interpretation allows a direct application of the procedure to other regions.

Description: Land evapotranspiration (ET) estimates are available from several global datasets. Here, monthly global land ET synthesis products, merged from these individual datasets over the time periods 1989–1995 (7 yr) and 1989–2005 (17 yr), are presented. The merged synthesis products over the shorter period are based on a total of 40 distinct datasets while those over the longer period are based on a total of 14 datasets. In the individual datasets, ET is derived from satellite and/or in-situ observations (diagnostic datasets) or calculated via land-surface models (LSMs) driven with observations-based forcing and atmospheric reanalyses. Statistics for four merged synthesis products are provided, one including all datasets and three including only datasets from one category each (diagnostic, LSMs, and reanalyses). The multi-annual variations of ET in the merged synthesis products display realistic responses. They are also consistent with previous findings of a global increase in ET between 1989 and 1997 (1.15 mm yr−2 in our merged product) followed by a decrease in this trend (−1.40 mm yr−2), although these trends are relatively small compared to the uncertainty of absolute ET values. The global mean ET from the merged synthesis products (based on all datasets) is 1.35 mm per day for both the 1989–1995 and 1989–2005 products, which is relatively low compared to previously published estimates. We estimate global runoff (precipitation minus ET) to 34 406 km3 per year for a total land area of 130 922 km2. Precipitation, being an important driving factor and input to most simulated ET datasets, presents uncertainties between single datasets as large as those in the ET estimates. In order to reduce uncertainties in current ET products, improving the accuracy of the input variables, especially precipitation, as well as the parameterizations of ET are crucial.

Description: The analysis of the spatial and temporal patterns of low flows as well as their generation mechanisms over large geographic regions can provide valuable insights and understanding for climate change impacts, regional frequency analysis, risk assessment of extreme events, and decision-making regarding allowable withdrawals. We use nonparametric tests to identify abrupt and gradual changes in time series of low flows and their timing for 508 USGS streamflow gauging sites in the eastern US with more than 50 years of daily data, to systematically distinguish the effects of human intervention from those of climate variability. A time series decomposition algorithm was applied to 1 day, 7 day, 30 day, and 90 day annual low flow time series that combines the Box–Ljung test for detection of autocorrelation, the Pettitt test for abrupt step changes and the Mann–Kendall test for monotonic trends. Examination of the USGS notes for each site confirmed that many of the step changes and around half of the sites with an increasing trend were associated with regulation. Around a third of the sites with a decreasing trend were associated with a change of gauge datum. Overall, a general pattern of increasing low flows in the northeast and decreasing low flows in the southeast is evident over a common time period (1951–2005), even when discarding sites with significant autocorrelation, documented regulation or other human impacts. The north–south pattern of trends is consistent with changes in antecedent precipitation. The main exception is along the mid-Atlantic coastal aquifer system from eastern Virginia northwards, where low flows have decreased despite increasing precipitation, and suggests that declining groundwater levels due to pumping may have contributed to decreased low flows. For most sites, the majority of low flows occur in one season in the late summer to autumn, as driven by the lower precipitation and higher evaporative demand in this season, but this is complicated in many regions because of the presence of a secondary low flow season in the winter for sites in the extreme northeast and in the spring for sites in Florida. Trends in low flow timing are generally undetectable, although abrupt step changes appear to be associated with regulation.

Description: We present protocols and input data for Phase 1 of the Global Gridded Crop Model Intercomparison, a project of the Agricultural Model Intercomparison and Improvement Project (AgMIP). The project includes global simulations of yields, phenologies, and many land-surface fluxes using 12–15 modeling groups for many crops, climate forcing data sets, and scenarios over the historical period from 1948 to 2012. The primary outcomes of the project include (1) a detailed comparison of the major differences and similarities among global models commonly used for large-scale climate impact assessment, (2) an evaluation of model and ensemble hindcasting skill, (3) quantification of key uncertainties from climate input data, model choice, and other sources, and (4) a multi-model analysis of the agricultural impacts of large-scale climate extremes from the historical record.

Description: The analysis of the spatial and temporal patterns of low flows as well as their generation mechanisms over large geographic regions can provide valuable insights and understanding for climate change impacts, regional frequency analysis, risk assessment of extreme events, and decision-making regarding allowable withdrawals. The goal of this paper is to examine nonstationarity in low flow generation across the eastern US and explore the potential anthropogenic influences or climate drivers. We use nonparametric tests to identify abrupt and gradual changes in time series of low flows and their timing for 508 USGS streamflow gauging sites in the eastern US with more than 50 years of daily data, to systematically distinguish the effects of human intervention from those of climate variability. A time series decomposition algorithm was applied to 1-day, 7-day, 30-day, and 90-day annual low flow time series that combines the Box–Ljung test for detection of autocorrelation, the Pettitt test for abrupt step changes and the Mann–Kendall test for monotonic trends. Examination of the USGS notes for each site showed that many of the sites with step changes and around half of the sites with an increasing trend have been documented as having some kind of regulation. Sites with decreasing or no trend are less likely to have documented influences on flows. Overall, a general pattern of increasing low flows in the northeast and decreasing low flows in the southeast is evident over a common time period (1951–2005), even when discarding sites with significant autocorrelation, documented regulation or other human impacts. The north–south pattern of trends is consistent with changes in antecedent precipitation. The main exception is along the mid-Atlantic coastal aquifer system from eastern Virginia northwards, where low flows have decreased despite increasing precipitation, and suggests that declining groundwater levels due to pumping may have contributed to decreased low flows. For most sites, the majority of low flows occur in one season in the late summer to fall, as driven by the lower precipitation and higher evaporative demand in this season, but this is complicated in many regions because of the presence of a secondary low flow season in the winter for sites in the extreme northeast and in the spring for sites in Florida. Trends in low flow timing are generally undetectable, although abrupt step changes appear to be associated with regulation.

Description: In the present work, we developed a new formulation for the estimation of the soil moisture in the root zone based on the measured value of soil moisture at the surface. It was derived from a simplified soil water balance equation for semiarid environments that provides a closed form of the relationship between the root zone and the surface soil moisture with a limited number of physically consistent parameters. The method sheds lights on the mentioned relationship with possible applications in the use of satellite remote sensing retrievals of soil moisture. The proposed approach was used on soil moisture measurements taken from the African Monsoon Multidisciplinary Analysis (AMMA) and the Soil Climate Analysis Network (SCAN) databases. The AMMA network was designed with the aim to monitor three so-called mesoscale sites (super sites) located in Benin, Mali, and Niger using point measurements at different locations. Thereafter the new formulation was tested on three additional stations of SCAN in the state of New Mexico (US). Both databases are ideal for the application of such method, because they provide a good description of the soil moisture dynamics at the surface and the root zone using probes installed at different depths. The model was first applied with parameters assigned based on the physical characteristics of several sites. These results highlighted the potential of the methodology, providing a good description of the root-zone soil moisture. In the second part of the paper, the model performances were compared with those of the well-known exponential filter. Results show that this new approach provides good performances after calibration with a set of parameters consistent with the physical characteristics of the investigated areas. The limited number of parameters and their physical interpretation makes the procedure appealing for further applications to other regions.