Search

The accurate prediction of surface soil moisture (SM) is crucial for understanding hydrological processes. Deep learning (DL) models such as the long short-term memory model (LSTM) provide a powerful method and have been widely used in SM prediction. However, few studies have notably high success rates due to lacking prior knowledge in forms such as causality. Here we present a new causality-structure-based LSTM model (CLSTM), which could learn time interdependency and causality information for hydrometeorological applications. We applied and compared LSTM and CLSTM methods for forecasting SM across 64 FLUXNET sites globally. The results showed that CLSTM dramatically increased the predictive performance compared with LSTM. The Nash–Sutcliffe efficiency (NSE) suggested that more than 67% of sites witnessed an improvement of SM simulation larger than 10%. It is highlighted that CLSTM had a much better generalization ability that can adapt to extreme soil conditions, such as SM response to drought and precipitation events. By incorporating causal relations, CLSTM increased predictive ability across different lead times compared to LSTM. We also highlighted the critical role of physical information in the form of causality structure to improve drought prediction. At the same time, CLSTM has the potential to improve predictions of other hydrometeorological variables.

Long-term continuous monitoring of precipitation isotopes has great potential to advance our understanding of hydrometeorological processes that determine stable isotope variability in the monsoon regions. This study presents 4-yr daily precipitation isotopes from Yungui Plateau in southwestern China that are influenced by Indian summer monsoon and East Asian monsoon. The local meteoric water line (LMWL;δ2H = 8.12δ18O + 11.2) was first established at the Tengchong (TC) site, which was close to the global meteoric water line (GMWL;δ2H = 8δ18O + 10), indicating little secondary subcloud evaporation in the falling rain. Precipitationδ18O values exhibited significant inverse relationships with precipitation amount (r= −0.42), air temperature (r= −0.43), and relative humidity (r= −0.41) with lower correlation coefficients throughout the entire period, which indicated that precipitation isotopic variability in TC could not be well explained by the local meteorological factors but influenced by other combined factors of regional precipitation amount and upstream rainout. Precipitationδ18O values showed a clear V-shaped trend throughout the observation period, characterized by higherδ18O values during the premonsoon period whereas lower values during the postmonsoon period. This seasonal variation of precipitationδ18O values was associated with the seasonal movement of the intertropical convergence zone and seasonal changes in moisture transport. Combined with backward trajectory analysis, precipitationδ18O values were estimated by a Rayleigh distillation model showing that upstream rainout processes from the Bay of Bengal (BoB) toward land (Myanmar) and recycling moisture over land were key factors affecting the isotopic compositions of the TC precipitation. These findings could enhance our understanding of atmospheric dynamics and moisture source in the monsoon regions and will potentially facilitate the interpretation of numerous isotopic proxy records from this region.SIGNIFICANCE STATEMENTThe variability of the summer monsoon and its onset, duration, and failure directly determine the strong rainfall and drought in a given region and have great impacts on regional societies and agriculture. To better understand this variability, this study presented a 4-yr daily dataset of precipitation isotopes on the Yungui Plateau of southwestern China to explore atmospheric processes and moisture sources that drive isotopic variability in this region. Precipitationδ18O exhibited remarkably seasonal variability, with higher values in premonsoon period and lower values in the postmonsoon period. During the Indian summer monsoon period, moisture sources primarily originated from the BoB toward the TC site, experiencing rainout processes and local moisture recycling over land using a Rayleigh fractionation model. These findings shed new light on the temporal variations of precipitation stable isotopes and facilitate our understanding of hydrological cycle in the monsoon regions.

Downward shortwave radiation Rsd determines the surface energy balance, alters evapotranspiration and hydrological conditions, and feeds back to the regional and global climate. Large-scale Rsd estimates are usually retrieved from satellite-based top-of-atmosphere radiation and cloud parameters. These estimates are subject to biases and temporal inhomogeneity due to errors in atmospheric parameters, algorithms, and sensor changes. We found that three satellite products overestimate Rsd by 8%–10% over Asia for 1984–2006, particularly in high latitudes. We used the model tree ensemble (MTE) machine-learning algorithm and commonly used ensemble averaging methods to integrate ground observations and satellite products. Validations based on test stations and independent networks showed that the MTE approach reduces the median relative biases from 8%–10% to 2%, which is more effective than the ensemble averaging methods. We further evaluated the impacts of uncertainty in radiation forcing on surface energy and water balances using the land surface model Noah-MP. The uncertainty of radiation data affects the prediction of sensible heat the most, and also largely affects latent heat prediction in humid regions. Holding the other variables constant, a 10% positive bias in Rsd can lead to a 20%–60% positive bias in the monthly median sensible heat. The simulated hydrological responses to changing radiation forcing are nonlinear as a result of the interactions among evapotranspiration, snowpack, and soil moisture. Our analysis concludes that reducing uncertainty of radiation data is beneficial for predicting regional energy and water balances, which requires more high-quality ground observations and improved satellite retrieval algorithms.

Accurate spatiotemporal predictions of surface soil moisture (SM) are important for many critical applications. Machine learning models provide a powerful method for building an accurate and reliable predictive model of SM. However, the models used in recent studies have some limitations, including lack of spatial autocorrelation (SAC), vague representation of important features, and primarily focused on the one-step forecast. Thus, we proposed an attention based convolutional long-short term memory model (AttConvLSTM) for multistep forecasting. The model includes three layers; spatial compression, axial attention, and encoder-decoder prediction, which are used for compressing spatial information, feature extraction, and multistep prediction, respectively. The model was trained using surface SM from Soil Moisture Active Passive L4 product at 18km spatial resolution over the United States. The results show that AttConvLSTM predicts 24 hours ahead SM with mean R2 and RMSE is equal to 0.82 and 0.02, respectively. Compared with LSTM, AttConvLSTM improves the model performance over 73.6% of regions, with an improvement of 8.4% and 17.4% in R2 and RMSE, respectively. The performance of the model is mainly influenced by temporal autocorrelation (TAC). Moreover, we also highlight the importance of SAC on model performance, especially over regions with high SAC and low TAC. Moreover, our model is also competent for SM predictions from several hours to several days, which could be a useful tool for predicting all meteorological variables and forecasting extremes.

Hydrological extremes, in the form of droughts and floods, have impacts on a wide range of sectors including water availability, food security, and energy production. Given continuing large impacts of droughts and floods and the expectation for significant regional changes projected in the future, there is an urgent need to provide estimates of past events and their future risk, globally. However, current estimates of hydrological extremes are not robust and accurate enough, due to lack of long-term data records, standardized methods for event identification, geographical inconsistencies, and data uncertainties. To tackle these challenges, this article presents the development of the first Global Drought and Flood Catalogue (GDFC) for 1950–2016 by merging the latest in situ and remote sensing datasets with state-of-the-art land surface and hydrodynamic modeling to provide a continuous and consistent estimate of the terrestrial water cycle and its extremes. This GDFC also includes an unprecedented level of detailed analysis of drought and large-scale flood events using univariate and multivariate risk assessment frameworks, which incorporates regional spatial–temporal characteristics (i.e., duration, spatial extent, severity) and global hazard maps for different return periods. This Catalogue forms a basis for analyzing the changing risk of droughts and floods and can underscore national and international climate change assessments and provide a key reference for climate change studies and climate model evaluations. It also contributes to the growing interests in multivariate and compounding risk analysis.

Open-path eddy covariance systems are widely used for measuring the CO2 flux between land and atmosphere. A common problem is that they often yield negative fluxes or physiologically unreasonable CO2 uptake fluxes in the nongrowing season under cold conditions. In this study, a meta-analysis was performed on the eddy flux data from 64 FLUXNET sites and the relationship between the observed CO2 flux and the sensible heat flux was analyzed. In theory, these two fluxes should be independent of each other in cold conditions (air temperature lower than 0°C) when photosynthesis is suppressed. However, the results show that a significant and negative linear relationship existed between these two fluxes at 37 of the sites. The mean linear slope value is −0.008 ± 0.001 µmol m−2 s−1 per W m−2 among the 64 sites analyzed. The slope value was not significantly different among the three gas analyzer models (LI-7500, LI-7500A, IRGASON/EC150) used at these sites, indicating that self-heating may not be the only reason for the apparent wintertime net CO2 uptake. These results suggest a systematic bias toward larger carbon uptakes in the FLUXNET sites that deploy open-path eddy covariance systems.