Degree

Doctor of Philosophy (PhD)

Department

Civil Engineering

Document Type

Dissertation

Abstract

Hydrostratigraphy model is an essential component of building valid groundwater models. Many challenges are associated with constructing hydrostratigraphy models which include geological complexities such as faults, domes, and angular unconformities. Developing a method with an emphasis on capturing big data to thoroughly inform large-scale models is one of the challenges addressed in the first part of this study. The method is predicated upon discretization of the study domain into tiles based on the geological dip direction and faults. The application of the method in the state of Louisiana with the utilization of more than 114000 well logs demonstrates promising results including identification of hydrostratigraphic characteristics for different aquifers, connections between the Mississippi River and the Red River and their alluviums, connections between state's surface waters and aquifers, and identification of recharge zones. The Louisiana model also demonstrated two different sand patterns in southeast Louisiana which might have been caused by two depositional environments. Employment of the method in a groundwater flow modeling framework to build a flow model for the Chicot aquifer system in southwest Louisiana revealed the complexity of the aquifer system that contains highly interconnected aquifer sands. The groundwater flow analysis of the Chicot aquifer is of great importance because it is the most heavily pumped aquifer in the state as a part of the Coastal Lowland Aquifer System. The modeling results show that the storage loss due to groundwater pumping is offset by inflows from surficial recharge, rivers, and boundaries. The two large cones of depression created by the agricultural pumping in the east and by the industrial pumping in the west represent the key feature in the Chicot aquifer system. As the final goal of this study, an aquifer storage and recovery operation in south of the Chicot aquifer was studied. The focus of this part was on optimal scheduling of an aquifer storage and recovery (ASR) operation while addressing parameter uncertainty for one cycle where an injection season is followed by a pumping season. This end was achieved via utilization of a supervised learning method for surrogated modeling and use of an evolutionary optimization method. The results indicate that artificial neural network (ANN) is a promising tool for evaluation of ASR efficiency. The hydraulic conductivity and longitudinal dispersivity were found to be the most significant parameters which affect ASR.

Date

7-6-2021

Committee Chair

Tsai, Frank T.-C.

Share

COinS