Remote sensing-AI based prediction of subsidence in Ararat Valley, Armenia Modeling Ground Subsidence Using InSAR, LSTM Networks, and Remote Sensing Data
(2024) In TVVR 5000 VVRM01 20241Division of Water Resources Engineering
- Abstract
- The Ararat Valley, being the cereal basket of Armenia and Turkey, is in danger
with over-abstracted groundwater, which already induces colossal water loss
and land subsidence. In the current work, advanced remote sensing techniques
and machine learning models are applied to predict subsidence and give datadriven
recommendations to effect sustainable water management. The tracking
of terrain changes and analysis and prediction of the patterns of subsidence are
carried out with the help of Synthetic Aperture Radar Interferometry and Long
Short-Term Memory networks, respectively. The integration of InSAR data
with LSTM models allows an in-depth analysis of subsidence dynamics,
forming a basis for forecasting and management... (More) - The Ararat Valley, being the cereal basket of Armenia and Turkey, is in danger
with over-abstracted groundwater, which already induces colossal water loss
and land subsidence. In the current work, advanced remote sensing techniques
and machine learning models are applied to predict subsidence and give datadriven
recommendations to effect sustainable water management. The tracking
of terrain changes and analysis and prediction of the patterns of subsidence are
carried out with the help of Synthetic Aperture Radar Interferometry and Long
Short-Term Memory networks, respectively. The integration of InSAR data
with LSTM models allows an in-depth analysis of subsidence dynamics,
forming a basis for forecasting and management strategies. In that regard, it is
early results confirmed the expediency of using the combination of remote
sensing data and machine learning techniques in stating that it is possible to
monitor subsidence accurately for all zones in the Ararat Valley. Further study
investigates the socio-economic effects that subsidence will have and will give
practical recommendations in relation to water management practice
improvements which include enforcement of water reuse regulations, enhance
the public awareness on the same, and adjustments in water pricing to its real
value. By addressing both technical and socio-economic aspects of water
management, the current research opens new horizons in providing a truly
integrative system for coping with both land subsidence and securing the
sustainable use of water resources in the Ararat Valley. It is considered a sign
that new and very promising opportunities are related to uniting innovative
technologies with large, comprehensive, and far-reaching policy measures to
effectively address environmental challenges. (Less) - Popular Abstract
- Ararat Valley is the cereal basket of Armenia and Turkey but suffers from maximal threats caused by
extreme groundwater abstraction, resulting in huge water losses and land subsidence. Advanced
remote sensing techniques and machine learning models are here applied to predict subsidence and
give data-driven recommendations on sustainable water management. The landform changes are
continuously monitored, and subsidence patterns are analyzed and predicted by Synthetic Aperture
Radar Interferometry (InSAR) and Long Short-Term Memory networks. InSAR data merged with
LSTM models allows for the comprehensive monitoring of the dynamics of subsidence, forming the
foundation for forecasting and management strategies. Preliminary results... (More) - Ararat Valley is the cereal basket of Armenia and Turkey but suffers from maximal threats caused by
extreme groundwater abstraction, resulting in huge water losses and land subsidence. Advanced
remote sensing techniques and machine learning models are here applied to predict subsidence and
give data-driven recommendations on sustainable water management. The landform changes are
continuously monitored, and subsidence patterns are analyzed and predicted by Synthetic Aperture
Radar Interferometry (InSAR) and Long Short-Term Memory networks. InSAR data merged with
LSTM models allows for the comprehensive monitoring of the dynamics of subsidence, forming the
foundation for forecasting and management strategies. Preliminary results demonstrated success in
the use of remote sensing data integrated with machine learning algorithms for accurate monitoring
of subsidence in all zones in the Ararat Valley. Socio-economic impacts of the subsidence are further
elaborated on, and practical recommendations for improving water management practices are
proposed: enforcement of water reuse regulations, increasing public awareness, and adjusting water
pricing to reflect its true value. By handling both the technical and socio-economic aspects of water
management, this research introduces an integrative system to minimize land subsidence and ensure
a sustainable water use regime in the Ararat Valley. It presents that there is huge hope for successfully
dealing with environmental challenges through an appropriate association of innovative technologies
with major policy actions. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9170815
- author
- Romano, Giuseppe LU
- supervisor
- organization
- alternative title
- Remote sensing-AI based prediction of subsidence in Ararat Valley, Armenia Modeling Ground Subsidence Using InSAR, LSTM Networks, and Remote Sensing Data
- course
- VVRM01 20241
- year
- 2024
- type
- H2 - Master's Degree (Two Years)
- subject
- publication/series
- TVVR 5000
- report number
- TVVR24/5012
- ISSN
- 1101-9824
- language
- English
- additional info
- Examiner; Kenneth M. Persson
- id
- 9170815
- date added to LUP
- 2024-08-22 09:06:43
- date last changed
- 2024-08-22 09:06:43
@misc{9170815,
abstract = {{The Ararat Valley, being the cereal basket of Armenia and Turkey, is in danger
with over-abstracted groundwater, which already induces colossal water loss
and land subsidence. In the current work, advanced remote sensing techniques
and machine learning models are applied to predict subsidence and give datadriven
recommendations to effect sustainable water management. The tracking
of terrain changes and analysis and prediction of the patterns of subsidence are
carried out with the help of Synthetic Aperture Radar Interferometry and Long
Short-Term Memory networks, respectively. The integration of InSAR data
with LSTM models allows an in-depth analysis of subsidence dynamics,
forming a basis for forecasting and management strategies. In that regard, it is
early results confirmed the expediency of using the combination of remote
sensing data and machine learning techniques in stating that it is possible to
monitor subsidence accurately for all zones in the Ararat Valley. Further study
investigates the socio-economic effects that subsidence will have and will give
practical recommendations in relation to water management practice
improvements which include enforcement of water reuse regulations, enhance
the public awareness on the same, and adjustments in water pricing to its real
value. By addressing both technical and socio-economic aspects of water
management, the current research opens new horizons in providing a truly
integrative system for coping with both land subsidence and securing the
sustainable use of water resources in the Ararat Valley. It is considered a sign
that new and very promising opportunities are related to uniting innovative
technologies with large, comprehensive, and far-reaching policy measures to
effectively address environmental challenges.}},
author = {{Romano, Giuseppe}},
issn = {{1101-9824}},
language = {{eng}},
note = {{Student Paper}},
series = {{TVVR 5000}},
title = {{Remote sensing-AI based prediction of subsidence in Ararat Valley, Armenia Modeling Ground Subsidence Using InSAR, LSTM Networks, and Remote Sensing Data}},
year = {{2024}},
}