Details
| Originalsprache | Englisch |
|---|---|
| Aufsatznummer | 5063 |
| Fachzeitschrift | Remote sensing |
| Jahrgang | 13 |
| Ausgabenummer | 24 |
| Frühes Online-Datum | 14 Dez. 2021 |
| Publikationsstatus | Veröffentlicht - Dez. 2021 |
Abstract
The current study presents a detailed assessment of risk zones related to karst collapse in Wuhan by analytical hierarchy process (AHP) and logistic regression (LR) models. The results showed that the LR model was more accurate with an area under the receiver operating characteristic (ROC) curve of 0.911 compared to 0.812 derived from the AHP model. Both models performed well in identifying high-risk zones with only a 3% discrepancy in area. However, for the medium-and low-risk classes, although the spatial distribution of risk zoning results were similar between two approaches, the spatial extent of the risk areas varied between final models. The reliability of both methods were reduced significantly by excluding the InSAR-based ground subsidence map from the analysis, with the karst collapse presence falling into the high-risk zone being reduced by approximately 14%, and karst collapse absence falling into the karst area being increased by approximately 6.5% on the training samples. To evaluate the practicality of using only results from ground subsidence maps for the risk zonation, the results of AHP and LR are compared with a weighted angular distortion (WAD) method for karst risk zoning in Wuhan. We find that the areas with relatively large subsidence horizontal gradient values within the karst belts are generally spatially consistent with high-risk class areas identified by the AHP-and LR-based approaches. However, the WAD-based approach cannot be used alone as an ideal karst collapse risk assessment model as it does not include geological and natural factors into the risk zonation.
ASJC Scopus Sachgebiete
- Erdkunde und Planetologie (insg.)
- Allgemeine Erdkunde und Planetologie
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
in: Remote sensing, Jahrgang 13, Nr. 24, 5063, 12.2021.
Publikation: Beitrag in Fachzeitschrift › Artikel › Forschung › Peer-Review
}
TY - JOUR
T1 - Karst collapse risk zonation and evaluation in Wuhan, China based on analytic hierarchy process, logistic regression, and insar angular distortion approaches
AU - Hu, Jiyuan
AU - Motagh, Mahdi
AU - Wang, Jiayao
AU - Qin, Fen
AU - Zhang, Jianchen
AU - Wu, Wenhao
AU - Han, Yakun
N1 - Funding Information: Funding: This research was sponsored by the National Natural Science Foundation of China (No. 42174037), the Key Laboratory of Geo-space Environment and Geodesy, the Ministry of Education, Wuhan University (No. 19-02-04), and Henan Provincial Key R&D and Promotion Special Project (Science and Technology Research) (No. 212102310414).
PY - 2021/12
Y1 - 2021/12
N2 - The current study presents a detailed assessment of risk zones related to karst collapse in Wuhan by analytical hierarchy process (AHP) and logistic regression (LR) models. The results showed that the LR model was more accurate with an area under the receiver operating characteristic (ROC) curve of 0.911 compared to 0.812 derived from the AHP model. Both models performed well in identifying high-risk zones with only a 3% discrepancy in area. However, for the medium-and low-risk classes, although the spatial distribution of risk zoning results were similar between two approaches, the spatial extent of the risk areas varied between final models. The reliability of both methods were reduced significantly by excluding the InSAR-based ground subsidence map from the analysis, with the karst collapse presence falling into the high-risk zone being reduced by approximately 14%, and karst collapse absence falling into the karst area being increased by approximately 6.5% on the training samples. To evaluate the practicality of using only results from ground subsidence maps for the risk zonation, the results of AHP and LR are compared with a weighted angular distortion (WAD) method for karst risk zoning in Wuhan. We find that the areas with relatively large subsidence horizontal gradient values within the karst belts are generally spatially consistent with high-risk class areas identified by the AHP-and LR-based approaches. However, the WAD-based approach cannot be used alone as an ideal karst collapse risk assessment model as it does not include geological and natural factors into the risk zonation.
AB - The current study presents a detailed assessment of risk zones related to karst collapse in Wuhan by analytical hierarchy process (AHP) and logistic regression (LR) models. The results showed that the LR model was more accurate with an area under the receiver operating characteristic (ROC) curve of 0.911 compared to 0.812 derived from the AHP model. Both models performed well in identifying high-risk zones with only a 3% discrepancy in area. However, for the medium-and low-risk classes, although the spatial distribution of risk zoning results were similar between two approaches, the spatial extent of the risk areas varied between final models. The reliability of both methods were reduced significantly by excluding the InSAR-based ground subsidence map from the analysis, with the karst collapse presence falling into the high-risk zone being reduced by approximately 14%, and karst collapse absence falling into the karst area being increased by approximately 6.5% on the training samples. To evaluate the practicality of using only results from ground subsidence maps for the risk zonation, the results of AHP and LR are compared with a weighted angular distortion (WAD) method for karst risk zoning in Wuhan. We find that the areas with relatively large subsidence horizontal gradient values within the karst belts are generally spatially consistent with high-risk class areas identified by the AHP-and LR-based approaches. However, the WAD-based approach cannot be used alone as an ideal karst collapse risk assessment model as it does not include geological and natural factors into the risk zonation.
KW - Analytical hierarchy process
KW - Karst collapse
KW - Logistic regression
KW - Risk zonation
KW - Weighted angular distortion method
UR - http://www.scopus.com/inward/record.url?scp=85121301162&partnerID=8YFLogxK
U2 - 10.3390/rs13245063
DO - 10.3390/rs13245063
M3 - Article
AN - SCOPUS:85121301162
VL - 13
JO - Remote sensing
JF - Remote sensing
SN - 2072-4292
IS - 24
M1 - 5063
ER -