High-resolution water budget estimates benefit from modeling of human water management and satellite data assimilation (DA) in river basins with a large human footprint. Utilizing the Noah-MP land surface model, in combination with an irrigation module, Sentinel-1 backscatter and snow depth observations, we produce a set of 0.7-km$^2$ digital water budget replicas of the Po river basin (Italy) for 2015-2023. The results demonstrate that irrigation modeling consistently improves the seasonal soil moisture variation and summer streamflow at all gauges in the valley after withdrawal of irrigation water from the streamflow (12\% error reduction relative to observed low summer streamflow), even if the basin-wide irrigation amount is underestimated. Sentinel-1 backscatter DA for soil moisture updating strongly interacts with irrigation modeling: when both are activated, the soil moisture updates are limited, and the simulated irrigation amounts are reduced. Backscatter DA systematically reduces soil moisture in the spring, which improves downstream spring streamflow. Assimilating Sentinel-1 snow depth retrievals over the surrounding Alps and Apennines further improves spring streamflow in a complementary way (2\% error reduction relative to observed high spring streamflow). Despite the seasonal improvements, irrigation modeling and Sentinel-1 backscatter DA cannot significantly improve short-term or interannual variations in soil moisture, irrigation results in a systematically prolonged high vegetation productivity, and snow depth DA only impacts the deep snowpacks. These findings help to advance the design and production of digital water budget replicas for river basins.

Irrigation is an important component of the terrestrial water cycle, but it is often poorly accounted for in models. Recent studies have attempted to integrate satellite data and land surface models via data assimilation (DA) to (1) detect and quantify irrigation, and (2) better model the related land surface variables such as soil moisture, vegetation, and evapotranspiration. In this study, different synthetic DA experiments are tested to advance satellite DA for the estimation of irrigation. We assimilate synthetic Sentinel-1 backscatter observations into the Noah-MP model coupled with an irrigation scheme. When updating soil moisture, we found that the DA sets better initial conditions to trigger irrigation in the model. However, large DA updates to wetter conditions can inhibit irrigation simulation. Building on this limitation, we propose an improved DA algorithm using a buddy check approach. The method still updates the land surface, but now the irrigation trigger is not based on the evolution of soil moisture, but on an adaptive innovation outlier detection. The new method was tested with different levels of model and observation error. For mild model and observation errors, the DA outperforms the model-only 14-day irrigation estimates by about 30% in terms of root-mean-squared differences, when frequent (daily or every other day) observations are available. The improvements can surpass 50% for high forcing errors. However, with longer observation intervals (7 days), the system strongly underestimates the irrigation amounts. The method is flexible and can be expanded to other DA systems and to a real world case.

Seasonal snow is an important water source and contributor to river discharge in mountainous regions. Therefore the amount of snow and its distribution are necessary inputs for hydrological modeling. However, the distribution of seasonal snow in mountains has long been uncertain, for lack of consistent, high resolution satellite retrievals over mountains. Recent research has shown the potential of the Sentinel-1 radar satellite to map snow depth at sub-kilometer resolution in mountainous regions. In this study we assimilate these new snow depth retrievals into the Noah-Multiparameterization land surface model using an ensemble Kalman filter for the western European Alps. The land surface model was coupled to the Hydrological Modeling and Analysis Platform to provide simulations of routed river discharge. The results show a reduction in the systematic underestimation of snow depth, going from 38 cm for the open loop (OL) to 11 cm for the data assimilation (DA) experiment. The mean absolute error similarly improves from 44 cm to 37 cm with DA, with an improvement at 59% of the in situ sites. The DA updates in snow depth results in enhanced snow water equivalent and discharge simulations. The systematic negative bias in the OL is mostly resolved, and the median temporal correlation between discharge simulations and measurements increases from 0.61 to 0.73 for the DA. Therefore, our study demonstrates the utility of the S1 snow depth retrievals to improve not only snow depth amounts, but also the snow melt contribution to river discharge, and hydrological modeling in general.

Tropical peatlands are characterized by highly organic, heterogeneous, and compressible peat soils. Without sampling and disturbing the soil, peat hydraulic and discharge parameters can be estimated from analyzing the in situ water level rise and recession. Such an analysis allows for the representation of the hydraulic behavior of a peatland from water level, precipitation, and topography data. Water level is measured in several remote tropical peatlands, whereas in situ precipitation is often not. Gridded satellite precipitation products provide an alternative, but are coarse and highly uncertain. Here, we introduce an algorithm for the hydrological parameterization of water level dynamics using satellite-based precipitation, and apply it to a tropical peatland in Brunei, while accounting for representativeness errors in the precipitation data. First, we adapt the rise and recession analysis developed by Cobb & Harvey (2019) for use with Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement mission (IMERG) precipitation estimates. The adapted rise analysis reduces the average error in the slope of the master rise curve with IMERG data from 21% to 3%. The average daily recession overestimation with IMERG data is reduced from 0.45 cm day -1 to 0.18 cm day -1. We also quantify the sensitivity of our rise analysis to precipitation errors using an ensemble of erroneous precipitation time series. Second, the adapted master rise and recession curves are used to fit soil hydraulic and discharge function parameters within the peatland-specific module of the NASA Catchment Land Surface Model. Our method enables the retrieval of accurate hydrological parameters for our case study, and should be tested in other peatland regions and with other satellite-based precipitation products.

This study assesses the potential of regional microwave backscatter data assimilation (DA) in AquaCrop for the first time, using NASA’s Land Information System (LIS). Soil moisture and crop biomass simulations from AquaCrop were updated using Sentinel-1 synthetic aperture radar (SAR) observations, over three regions in Europe. This was done in two separate DA experiments. The first experiment concerned updating soil moisture using VV-polarized backscatter and the corrections were propagated via the model to the biomass. In the second experiment, the DA setup was extended by also updating the biomass with VH-polarized backscatter. The assimilation showed mixed results, with slight overall improvements in the (anomaly) correlations of soil moisture simulations relative to independent satellite and in situ data. By contrast, the biomass estimates obtained with backscatter data assimilation did not agree better with reference datasets from independent optical and microwave satellite data. Given the complex backscatter response over agricultural areas, more research is required to optimally perform DA to improve crop growth estimation.

Current lightning predictions are uncertain because they either rely on empirical diagnostic relationships based on the present climate or use coarse-scale climate scenario simulations in which deep convection is parameterized. Previous studies demonstrated that simulations with convection-permitting resolutions (km-scale) improve lightning predictions compared to coarser-grid simulations using convection parameterization for different geographical locations but not over the boreal zone. In this study, lightning simulations with the NASA Unified-Weather Research and Forecasting (NU-WRF) model are evaluated over a 955x540 km2 domain including the Great Slave Lake in Canada for six lightning seasons. The simulations are performed at convection-parameterized (9 km) and convection-permitting (3 km) resolution using the Goddard 4ICE and the Thompson microphysics (MP) schemes. Four lightning indices are evaluated against observations from the Canadian Lightning Detection Network (CLDN), in terms of spatiotemporal frequency distribution, spatial pattern, daily climatology, and an event-based overall skill assessment. Concerning the model configuration, regardless of the spatial resolution, the Thompson scheme is superior to the Goddard 4ICE scheme in predicting the daily climatology but worse in predicting the spatial patterns of lightning occurrence. Several evaluation metrics indicate the benefit of working at a convection-permitting resolution. The relative performance of the different lightning indices depends on the evaluation criteria. Finally, this study demonstrates issues of the models to reproduce the observed spatial pattern of lightning well, which might be related to an insufficient representation of land surface heterogeneity in the study area.

Tropical peatlands are among the most carbon-dense ecosystems on Earth, and their water storage dynamics strongly control these carbon stocks. The hydrological functioning of tropical peatlands differs from that of northern peatlands, which has not yet been accounted for in global land surface models (LSMs). Here, we integrated tropical peat-specific hydrology modules into a global LSM for the first time, by utilizing the peatland-specific model structure adaptation (PEATCLSM) of the NASA Catchment Land Surface Model (CLSM). We developed literature-based parameter sets for natural (PEATCLSMTrop,Nat) and drained (PEATCLSMTrop,Drain) tropical peatlands. The operational CLSM version (which includes peat as a soil class) and PEATCLSMTrop,Nat were forced with global meteorological input data and evaluated over the major tropical peatland regions in Central and South America, the Congo Basin, and Southeast Asia. Evaluation against a unique and extensive data set of in situ water level and eddy covariance-derived evapotranspiration showed an overall improvement in bias and correlation over all three study regions. Over Southeast Asia, an additional simulation with PEATCLSMTrop,Drain was run to address the large fraction of drained tropical peatlands in this region. PEATCLSMTrop,Drain outperformed both CLSM and PEATCLSMTrop,Nat over drained sites. Despite the overall improvements of both tropical PEATCLSM modules, there are strong differences in performance between the three study regions. We attribute these performance differences to regional differences in accuracy of meteorological forcing data, and differences in peatland hydrologic response that are not yet captured by our model.