Spatio-temporal modeling of Louisiana land subsidence using high resolution geo-spatial data

Problems caused by subsidence are very common in many areas of the world, and this kind of problems may be serious and threatening to living people in Louisiana. Adverse subsidence in Louisiana will cause serious problems, such as excessive wetland formation or land loss, if we can’t make appropriate treatments, and this topic will also be what we focus on in this research (Kent and Dokka 2012). For subsidence survey, we can use three kinds of common techniques, leveling, InSAR (Interferometric Synthetic Aperture Radar) and GPS observation (Lu, C. et al. 2012). In this research, high accuracy of subsidence data in Louisiana has been collected by GPS, and Kriged-Kalman Filter (KKF) has been used to process such subsidence data (Mardia et al. 1998). Results by KKF have shown spatio-temporal distributions of subsidence rates from 2011 to 2013, and these results have also been validated by the Bayou Corne Sinkhole knowledge in this research (Mardia et al. 1998; Cusanza 2013; Jones and Blom 2014; Jones and Blom 2015). Based on the validated KKF results in this research, we have used some geo-statistics models, such as Geographically Weighted Regression (GWR), the spatial-lag model and the spatial-error model, so as to find which main factors have caused adverse subsidence in the study site in 2013 (Mardia et al. 1998; Fotheringham et al. 2002; Baller et al. 2001; Wang 2006; Wang et al. 2014; Abdollahzadeh et al. 2013). Modeling results have shown that, either GWR or the spatial-error model may be suitable in this research, and Bayou Corne Sinkhole, sediment compaction, groundwater withdrawal and mass loading of buildings may be the significant and explainable factors causing subsidence in the study site (Fotheringham et al. 2002; Hayashi and Fumio 2000; Abdollahzadeh et al. 2013; Xu and Wang 2015; Kim et al. 2006; Kim et al. 2009; Oh and Lee 2010; Oh et al. 2011; Cusanza 2013; Jones and Blom 2014; Jones and Blom 2015; Anselin et al. 2006; Baller et al. 2001; Wang 2006; Wang et al. 2014; Shang et al. 2011; Sclater and Christie 1980). Thus, in this research, we have concluded that KKF is a valid model to generate spatio-temporal distributions of subsidence rates, by GWR the spatial heterogeneity for subsidence will be clearly found and by the spatial-lag model the main factors causing subsidence in Louisiana will also be clearly found (Mardia et al. 1998; Fotheringham et al. 2002; Hayashi and Fumio 2000; Abdollahzadeh et al. 2013; Xu and Wang 2015; Kim et al. 2006; Kim et al. 2009; Oh and Lee 2010; Oh et al. 2011; Cusanza 2013; Jones and Blom 2014; Jones and Blom 2015; Anselin et al. 2006; Baller et al. 2001; Wang 2006; Wang et al. 2014; Shang et al. 2011; Sclater and Christie 1980).

Name thesis:degree_name

Doctor of Philosophy (PhD)

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Social and Behavioral Sciences

Grantor

Geography and Anthropology

Year dc:date.available

2016

Author dc:creator

• Xiang, Hanyu

• subsidence
• Louisiana
• factor

Statement dc:rights

• unrestricted
• Release the entire work immediately for access worldwide.

Identifier

etd-06012016-135948
https://repository.lsu.edu/gradschool_dissertations/229

OAI identifier oai:identifier

oai:repository.lsu.edu:gradschool_dissertations-1228