AI or Artificial Intelligence is a pioneering technique that has enabled the creation of intelligent machines.or smart machines which has the power to self adapt based on the situation presented to it. It requires situations whose response is known and based on this training data set it learns the problems which it has to solve when it is ready. Due to the alarming success with AI in robotics, electronics etc fields the same technique is now used to solve the problems of water resource management.. This ppt shows seven most notable use of AI in water resources-based problems where satisfactory improvement has encouraged further application of the technique.
2. WHAT IS AI ?
• Artificial intelligence (AI) is a wide-ranging branch of
computer science concerned with building smart
machines capable of performing tasks that typically
require human intelligence.
• Artificial intelligence is widely used to provide
personalised recommendations to people, based for
example on their previous searches and purchases or
other online behaviour.
• AI is hugely important in commerce: optimising products,
planning inventory, logistics etc.
3. WHAT IS AI ?
CONTD.
• The term is frequently applied to the project of
developing systems endowed with the intellectual
processes characteristic of humans, such as the
ability to reason, discover meaning, generalize, or
learn from past experience.
• Basic logic behind AI is the information processing
by a neural network which can adapt itself based on
responses, just like human brain.
• AI can also be used for solving water resource
related problems as water resource problems are
highly adaptive and changes with both time and
space.
• Some of the example application of AI is delineated
in this presentation
4. AI ON WATER QUALITY
• Artificial intelligence is recently used to enhance remote
monitoring of water bodies by University of Stirling :–
• A new algorithm was developed
• Known as the 'meta-learning' method was developed to
analyse the data directly from satellite sensors thus
making it easier for coastal zone, environmental and
industry managers to monitor issues such as harmful
algal blooms (HABs) and possible toxicity in shellfish and
finfish.
Source : University of Stirling. "Artificial intelligence to monitor water quality more
effectively." ScienceDaily. www.sciencedaily.com/releases/2021/05/210504112514.htm
(accessed May 8, 2021).
5. FLOOD
SUSCEPTIBILITY
• Flood susceptibility maps are crucial steps
for decision-makers to prevent and
manage disasters which helps to identify
the vulnerable regions.
• Plenty of studies have used machine
learning models to produce reliable
susceptibility maps.
• A novel method was developed to use the
deep learning technique of LSTM to
predict flood susceptibility
• Appropriate feature engineering method
with LSTM was integrated to predict flood
susceptibility.
• As a result not only temporal but also
spatial accuracy of prediction get
improved.
Source : Fang, Zhice, Yi Wang, Ling Peng, and Haoyuan
Hong. "Predicting flood susceptibility using LSTM neural
networks." Journal of Hydrology 594 (2021): 125734.
6. ESTIMATION OF EVAPOTRANSPIRATION
AND ITS COMPONENTS
• Evapotranspiration (ET) and its components of soil
evaporation (E) and vegetation transpiration (T), as key
variables for the water-energy exchange between the
land surface and the atmosphere, are widely used in
hydrological and agricultural applications.
• The land surface temperature based two-source
energy balance (TSEB) model can provide high
accuracy E, T and ET, which are spatio-temporally
discontinuous, but spatio-temporally continuous daily
ET is more helpful in water resources management.
• This study attempted to develop a new combined
model coupling the TSEB model and deep neural
network (DNN) (TSEB_DNN) to improve the continuity
of the prediction.
• The TSEB_DNN model was found to be well consistent
with the in situ measurements and had the overall
correlation coefficient (R), root-mean-square-error
(RMSE), and bias values of 0.88, 0.88 mm d-1, and 0.37
mm d-1, respectively.
• The introduction of DNN improved the continuity of
the prediction both in spatial and temporal scale.
Source : Cui, Yaokui, Lisheng Song, and Wenjie Fan. "Generation of spatio-temporally
continuous evapotranspiration and its components by coupling a two-source energy balance
model and a deep neural network over the Heihe River Basin." Journal of Hydrology 597 (2021):
126176.
7. DETERMINATION OF GROUND
WATER LEVEL
• Precise estimation of physical hydrology components including groundwater levels
(GWLs) is a challenging task, especially in relatively non-contiguous watersheds.
• This study estimates GWLs with deep learning and artificial neural networks (ANNs),
namely a multilayer perceptron (MLP), long short term memory (LSTM), and a
convolutional neural network (CNN) with the help of the variables :stream level,
stream flow, precipitation, relative humidity and mean temperature.
• The stream level was found to be the major contributor to GWL fluctuation for the
Baltic River and Long Creek watersheds
• “The deep learning techniques introduced in this study to estimate GWL fluctuations
are convenient and accurate as compared to collection of periodic dips based on the
groundwater monitoring wells for groundwater inventory control and
management.”…Afzaal et.al.(2020).
Source : Afzaal, Hassan, Aitazaz A. Farooque, Farhat Abbas, Bishnu Acharya, and Travis
Esau. "Groundwater estimation from major physical hydrology components using
artificial neural networks and deep learning." Water 12, no. 1 (2020): 5.
8. SOIL MOISTURE
• A new multi-model ensemble committee framework called
ANN-CoM was developed.
• ANN-CoM was evaluated in forecasting upper- and lower-
layer soil moistures.
• Volterra, M5 tree, random forest, and ELM served as
comparison models.
• Superior performance of ANN-CoM was recorded at all
four study sites.
• Seasonally, ANN-CoM generated better forecasts of lower
layer soil moisture.
• Source : Prasad, Ramendra, Ravinesh C. Deo, Yan Li, and
Tek Maraseni. "Ensemble committee-based data intelligent
approach for generating soil moisture forecasts with
multivariate hydro-meteorological predictors." Soil and
Tillage Research 181 (2018): 63-81.
9. MONITORING
SUSPENDED
SEDIMENT
CONCENTRATIONS
• Purpose : The high costs of monitoring suspended
sediment concentration (SSC) in rivers calls for
development of indirect estimation methods, based
on relationships with other variables which are easier
and cheaper to obtain.
• Input : drainage area, soil type, land use and cover
and mean catchment slope turbidity, flow,
precipitation and exponentially weighted moving
average of past rainfall
• Model used : Artificial Neural network (ANN) based
models
• Location : Brazilian part of the Upper Paraguay River
Basin,
• Conclusion : The proposed methodology allows the
regional extrapolation of SSC to ungauged basins
with very good performance, even in heterogeneous
regions.
• Source : Campos, Juliana Andrade, and Olavo Correa
Pedrollo. "A regional ANN-based model to estimate
suspended sediment concentrations in ungauged
heterogeneous basins." Hydrological Sciences Journal
just-accepted (2021).
10. SNOW DEPTH
• Purpose : “Accurate high spatial resolution snow depth mapping in arid and
semi-arid regions is of great importance for snow disaster assessment and
hydrological modelling. “. But due to the complex topography and low
spatial-resolution microwave remote-sensing data, the existing snow depth
datasets have large errors and uncertainty, and actual spatiotemporal
heterogeneity of snow depth cannot be effectively detected.
• Inputs : The (Fengyun-3 Microwave Radiation Imager) FY-3 MWRI data were
downscaled to 500 m resolution to match Moderate-resolution Imaging
Spectroradiometer (MODIS) snow cover, meteorological and geographic data.
• Output : Downscaled snow depth data
• Model Used : Deep Learning
• Conclusion : “Downscaled snow depth could provide more detailed
information in spatial distribution, which has been used to analyze the
decrease of retrieval accuracy by various topography factors.”… Zhu
et.al.(2021).
• Source : Zhu, Linglong, Yonghong Zhang, Jiangeng Wang, Wei Tian, Qi Liu,
Guangyi Ma, Xi Kan, and Ya Chu. "Downscaling Snow Depth Mapping by
Fusion of Microwave and Optical Remote-Sensing Data Based on Deep
Learning." Remote Sensing 13, no. 4 (2021): 584.
11. CONCLUSION
• Six tenable examples of the application of Artificial Intelligence on
water resource problems were discussed.
• In all the studies, the most common observation was the
noticeable improvement of the accuracy in the estimations from
the deep learning model compared to the conventional numerical
frameworks.
• The capacity to adapt from the input data is the main advantage
of using AI
• But data requirement is also high.
• Model developed with fewer data is as accurate as linear models
• If very few data or examples are provided to the model for
learning the inherent relationships that exist in between the input
and output variables of the model AI based models will fail
miserably.
• In case of water resource problems data availability is the main
constraint.
• So intelligent application of AI based models hybridizing with
another model may solve the problems, but it may also
compromise the adaptable capacity of AI models.