Knowledge transfer from simulation to reality via Long Short-Term Memory networks:  Estimating groundwater table depth anomalies over Europe

2021 ◽  
Author(s):  
Yueling Ma ◽  
Carsten Montzka ◽  
Bagher Bayat ◽  
Stefan Kollet
Keyword(s):  
Short Term Memory ◽  
Scientific Data ◽  
Groundwater Table ◽  
Short Term ◽  
Term Memory ◽  
Continental Scale ◽  
Groundwater Table Depth ◽  
Long Short Term Memory ◽  
Simulation Results

<p>Near real-time groundwater table depth measurements are scarce over Europe, leading to challenges in monitoring groundwater resources at the continental scale. In this study, we leveraged knowledge learned from simulation results by Long Short-Term Memory (LSTM) networks to estimate monthly groundwater table depth anomaly (<em>wtd<sub>a</sub></em>) data over Europe. The LSTM networks were trained, validated, and tested at individual pixels on anomaly data derived from daily integrated hydrologic simulation results over Europe from 1996 to 2016, with a spatial resolution of 0.11° (Furusho-Percot et al., 2019), to predict monthly <em>wtd<sub>a</sub></em> based on monthly precipitation anomalies (<em>pr<sub>a</sub></em>) and soil moisture anomalies (<em>θ<sub>a</sub></em>). Without additional training, we directly fed the networks with averaged monthly <em>pr<sub>a</sub></em> and <em>θ<sub>a</sub></em> data from 1996 to 2016 obtained from commonly available observational datasets and reanalysis products, and compared the network outputs with available borehole <em>in situ</em> measured <em>wtd<sub>a</sub></em>. The LSTM network estimates show good agreement with the <em>in situ</em> observations, resulting in Pearson correlation coefficients of regional averaged <em>wtd<sub>a</sub></em> data in seven PRUDENCE regions ranging from 42% to 76%, which are ~ 10% higher than the original simulation results except for the Iberian Peninsula. Our study demonstrates the potential of LSTM networks to transfer knowledge from simulation to reality for the estimation of <em>wtd<sub>a</sub></em> over Europe. The proposed method can be used to provide spatiotemporally continuous information at large spatial scales in case of sparse ground-based observations, which is common for groundwater table depth measurements. Moreover, the results highlight the advantage of combining physically-based models with machine learning techniques in data processing.</p><p> </p><p>Reference:</p><p>Furusho-Percot, C., Goergen, K., Hartick, C., Kulkarni, K., Keune, J. and Kollet, S. (2019). Pan-European groundwater to atmosphere terrestrial systems climatology from a physically consistent simulation. Scientific Data, 6(1).</p>

scholarly journals An Indirect Approach Based on Long Short-Term Memory Networks to Estimate Groundwater Table Depth Anomalies Across Europe With an Application for Drought Analysis

2021 ◽  
Vol 3 ◽  
Author(s):  
Yueling Ma ◽  
Carsten Montzka ◽  
Bagher Bayat ◽  
Stefan Kollet
Keyword(s):  
Soil Moisture ◽  
Indirect Method ◽  
Short Term Memory ◽  
Groundwater Table ◽  
Short Term ◽  
Data Set ◽  
Term Memory ◽  
Continental Scale ◽  
Groundwater Table Depth ◽  
Long Short Term Memory

The lack of high-quality continental-scale groundwater table depth observations necessitates developing an indirect method to produce reliable estimation for water table depth anomalies (wtda) over Europe to facilitate European groundwater management under drought conditions. Long Short-Term Memory (LSTM) networks are a deep learning technology to exploit long-short-term dependencies in the input-output relationship, which have been observed in the response of groundwater dynamics to atmospheric and land surface processes. Here, we introduced different input variables including precipitation anomalies (pra), which is the most common proxy of wtda, for the networks to arrive at improved wtda estimates at individual pixels over Europe in various experiments. All input and target data involved in this study were obtained from the simulated TSMP-G2A data set. We performed wavelet coherence analysis to gain a comprehensive understanding of the contributions of different input variable combinations to wtda estimates. Based on the different experiments, we derived an indirect method utilizing LSTM networks with pra and soil moisture anomaly (θa) as input, which achieved the optimal network performance. The regional medians of test R2 scores and RMSEs obtained by the method in the areas with wtd ≤ 3.0 m were 76–95% and 0.17–0.30, respectively, constituting a 20–66% increase in median R2 and a 0.19–0.30 decrease in median RMSEs compared to the LSTM networks only with pra as input. Our results show that introducing θa significantly improved the performance of the trained networks to predict wtda, indicating the substantial contribution of θa to explain groundwater anomalies. Also, the European wtda map reproduced by the method had good agreement with that derived from the TSMP-G2A data set with respect to drought severity, successfully detecting ~41% of strong drought events (wtda ≥ 1.5) and ~29% of extreme drought events (wtda ≥ 2) in August 2015. The study emphasizes the importance to combine soil moisture information with precipitation information in quantifying or predicting groundwater anomalies. In the future, the indirect method derived in this study can be transferred to real-time monitoring of groundwater drought at the continental scale using remotely sensed soil moisture and precipitation observations or respective information from weather prediction models.

Modeling of groundwater table depth anomalies using Long Short-Term Memory networks over Europe

2020 ◽  
Author(s):  
Yueling Ma ◽  
Carsten Montzka ◽  
Bagher Bayat ◽  
Stefan Kollet
Keyword(s):  
Real Time ◽  
Short Term Memory ◽  
Groundwater Resources ◽  
Scientific Data ◽  
Great Promise ◽  
Monthly Precipitation ◽  
Short Term ◽  
Term Memory ◽  
Continental Scale ◽  
Long Short Term Memory

<p>Groundwater is the dominant source of fresh water in many European countries. However, due to a lack of near-real-time water table depth (wtd) observations, monitoring of groundwater resources is not feasible at the continental scale. Thus, an alternative approach is required to produce wtd data from other available observations near-real-time. In this study, we propose Long Short-Term Memory (LSTM) networks to model monthly wtd anomalies over Europe utilizing monthly precipitation anomalies as input. LSTM networks are a special type of artificial neural networks, showing great promise in exploiting long-term dependencies between time series, which is expected in the response of groundwater to precipitation. To establish the methodology, spatially and temporally continuous data from terrestrial simulations at the continental scale were applied with a spatial resolution of 0.11°, ranging from the year 1996 to 2016 (Furusho-Percot et al., 2019). They were divided into a training set (1996 – 2012), a validation set (2012 – 2014) and a testing set (2015 -2016) to construct local models on selected pixels over eight PRUDENCE regions. The outputs of the LSTM networks showed good agreement with the simulation results in locations with a shallow wtd (~3m). It is important to note, the quality of the models was strongly affected by the amount of snow cover. Moreover, with the introduction of monthly evapotranspiration anomalies as additional input, pronounced improvements of the network performances were only obtained in more arid regions (i.e., Iberian Peninsula and Mediterranean). Our results demonstrate the potential of LSTM networks to produce high-quality wtd anomalies from hydrometeorological variables that are monitored at the large scale and part of operational forecasting systems potentially facilitating the implementation of an efficient groundwater monitoring system over Europe.</p><p>Reference:</p><p>Furusho-Percot, C., Goergen, K., Hartick, C., Kulkarni, K., Keune, J. and Kollet, S. (2019). Pan-European groundwater to atmosphere terrestrial systems climatology from a physically consistent simulation. Scientific Data, 6(1).</p>

Forecasting Net Groundwater Depletion in Well Irrigation Areas with Long Short-term Memory Networks

2020 ◽  
Author(s):  
Li Yuheng ◽  
Tang Lihua
Keyword(s):  
Hydrological Model ◽  
Short Term Memory ◽  
Irrigation Management ◽  
Groundwater Table ◽  
Groundwater Depletion ◽  
Short Term ◽  
Term Memory ◽  
Groundwater Use ◽  
Freshwater Budget ◽  
Long Short Term Memory

<p>Due to the scarcity of available surface water, many irrigated areas in North China Plain (NCP) heavily rely on groundwater, which has resulted in groundwater overexploitation and massive environmental impacts, such as groundwater depression core and land subsidence. The net groundwater depletion, one of the groundwater indicators, means the actual groundwater consumption for human impact. This indicator is quite essential for the evaluation of the effects of agricultural activities in well irrigation areas. However, net depletion forecasts, which can help inform the management of well irrigation areas, are generally unavailable with easy methods. Therefore, this study explored machine learning models, Long Short-term Memory (LSTM) networks, to forecast net groundwater depletion in well irrigation counties, Hebei Province. Firstly, Luancheng county was selected to construct the forecasting model. The training dataset was prepared by collecting the measured precipitation, remote sensing evaporation and groundwater table from 2006-2017. Besides, an agro-hydrological model (Soil-Water-Atmosphere-Plant, SWAP) with an optimization tool (Parameter ESTimation, PEST) was used to calculate the net depletion, and an unsaturated-saturated zone water balance conceptual hydrological model was constructed to calculate the net groundwater use. Secondly, to determine the effect of training data type on model accuracy, freshwater budget (evaporation minus precipitation), change of groundwater table and net groundwater use were chosen as training inputs by analyzing related temporal variable characteristics of net groundwater depletion. The response time of training inputs with net groundwater depletion were also approximated with highest cross-correlation value (CCF). Then, by circular bootstrapping methods to enlarge the Luancheng datasets from 2006-2016, the annual and monthly model for forecasting the net depletion were respectively trained with enlarged Luancheng datasets. Additionally, to test the model’s ability to predict the net groundwater depletion in other well irrigation areas with the similar rule of groundwater depletion, the annual and monthly forecasting scenarios were also carried out in the adjacent county, Zhaoxian. The results showed that both of the monthly and annual models estimating the groundwater net depletion had good performance in Zhaoxian from 2006-2017, with NSE of 0.91 and 0.81, respectively. According to the modelling results, further analysis showed that groundwater depletion in research counties mainly occurred in spring (March to May) and winter (December to February). In addition, the major factor leading to groundwater depletion in spring and winter was freshwater budget; while in summer and autumn, soil moisture determined the depletion activity. These results demonstrate the feasible use of LSTM networks to create annual and monthly forecasts of net groundwater depletion in well irrigation areas with similar depletion rule, which can provide valuable suggestion to well irrigation management in NCP within a challenging environment.</p><p><strong>Keywords: net groundwater depletion; long short-term memory; well irrigation areas</strong></p>

scholarly journals Developing and Testing a Long Short-Term Memory Stream Temperature Model in Daily and Continental Scale

2020 ◽  
Author(s):  
Farshid Rahmani ◽  
Samantha Oliver ◽  
Wenyu Ouyang ◽  
Alison Appling ◽  
Kathryn Lawson ◽  
...  
Keyword(s):  
Short Term Memory ◽  
Stream Temperature ◽  
Temperature Model ◽  
Short Term ◽  
Term Memory ◽  
Continental Scale ◽  
Long Short Term Memory

Predicting energy and carbon fluxes using LSTM networks

2021 ◽  
Author(s):  
Claire Brenner ◽  
Jonathan Frame ◽  
Grey Nearing ◽  
Karsten Schulz
Keyword(s):  
Short Term Memory ◽  
Model Performance ◽  
Carbon Fluxes ◽  
Reanalysis Data ◽  
Scientific Data ◽  
Short Term ◽  
Term Memory ◽  
Carbon Cycles ◽  
Wide Range ◽  
Long Short Term Memory

<p>Global land-atmosphere energy and carbon fluxes are key drivers of the Earth’s climate system. Their assessment over a wide range of climates and biomes is therefore essential (i) for a better understanding and characterization of land-atmosphere exchanges and feedbacks and (ii) for examining the effect of climate change on the global water, energy and carbon cycles. </p><p>Large-sample datasets such as the FLUXNET2015 dataset (Pastorello et al., 2020) foster the use of machine learning (ML) techniques as a powerful addition to existing physically-based modelling approaches. Several studies have investigated ML techniques for assessing energy and carbon fluxes, and while across-site variability and the mean seasonal cycle are typically well predicted, deviations from mean seasonal behaviour remains challenging (Tramontana et al., 2016). </p><p>In this study we examine the importance of memory effects for predicting energy and carbon fluxes at half-hourly and daily temporal resolutions. To this end, we train a Long Short-Term Memory (LSTM, Hochreiter and Schmidthuber, 1997), a recurrent neural network with explicit memory, that is particularly suited for time series predictions due to its capability to store information over longer (time) sequences. We train the LSTM on a large number of FLUXNET sites part of the FLUXNET2015 dataset using local meteorological forcings and static site attributes derived from remote sensing and reanalysis data. </p><p>We evaluate model performance out-of-sample (leaving out individual sites) in a 10-fold cross-validation. Additionally, we compare results from the LSTM with results from another ML technique, XGBoost (Chen and Guestrin, 2016), that does not contain system memory. By analysing the differences in model performances of both approaches across various biomes, we investigate under which conditions the inclusion of memory might be beneficial for modelling energy and carbon fluxes.</p><p> </p><p>References:</p><p>Chen, Tianqi, and Carlos Guestrin. "Xgboost: A scalable tree boosting system." Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. 2016.</p><p>Hochreiter, Sepp, and Jürgen Schmidhuber. "Long short-term memory." Neural computation 9.8 (1997): 1735-1780.</p><p>Pastorello, Gilberto, et al. "The FLUXNET2015 dataset and the ONEFlux processing pipeline for eddy covariance data." Scientific data 7.1 (2020): 1-27</p><p>Tramontana, Gianluca, et al. "Predicting carbon dioxide and energy fluxes across global   FLUXNET sites with regression algorithms." Biogeosciences 13.14 (2016): 4291-4313.</p>

scholarly journals Instrument Bias Correction With Machine Learning Algorithms: Application to Field-Portable Mass Spectrometry

2020 ◽  
Vol 8 ◽  
Author(s):  
B. Loose ◽  
R. T. Short ◽  
S. Toler
Keyword(s):  
Statistical Learning ◽  
Environmental Conditions ◽  
Bias Correction ◽  
Short Term Memory ◽  
Short Term ◽  
Learning Methods ◽  
Term Memory ◽  
Instrument Bias ◽  
Long Short Term Memory

In situ sensors for environmental chemistry promise more thorough observations, which are necessary for high confidence predictions in earth systems science. However, these can be a challenge to interpret because the sensors are strongly influenced by temperature, humidity, pressure, or other secondary environmental conditions that are not of direct interest. We present a comparison of two statistical learning methods—a generalized additive model and a long short-term memory neural network model for bias correction of in situ sensor data. We discuss their performance and tradeoffs when the two bias correction methods are applied to data from submersible and shipboard mass spectrometers. Both instruments measure the most abundant gases dissolved in water and can be used to reconstruct biochemical metabolisms, including those that regulate atmospheric carbon dioxide. Both models demonstrate a high degree of skill at correcting for instrument bias using correlated environmental measurements; the difference in their respective performance is less than 1% in terms of root mean squared error. Overall, the long short-term memory bias correction produced an error of 5% for O2 and 8.5% for CO2 when compared against independent membrane DO and laser spectrometer instruments. This represents a predictive accuracy of 92–95% for both gases. It is apparent that the most important factor in a skillful bias correction is the measurement of the secondary environmental conditions that are likely to correlate with the instrument bias. These statistical learning methods are extremely flexible and permit the inclusion of nearly an infinite number of correlates in finding the best bias correction solution.

Sign in / Sign up

Export Citation Format

Share Document