Next Article in Journal
Simulation Optimization of Station-Level Control of Large-Scale Passenger Flow Based on Queueing Network and Surrogate Model
Previous Article in Journal
The Impact of Digitalization on Industrial Engineering Students’ Training from the Perspective of Their Insertion in the Labor Market in a Sustainable Economy: A Students’ Opinions Survey
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 16
Issue 17
10.3390/su16177501
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Urban Geothermal Resource Potential Mapping Using Data-Driven Models—A Case Study of Zhuhai City

by
Yu Bian
1,
Yong Ni
2,
Ya Guo
1,
Jing Wen
1,
Jie Chen
1,
Ling Chen
1,* and
Yongpeng Yang
1,*
1
China Aero Geophysical Survey and Remote Sensing Center for Natural Resources, Beijing 100083, China
2
The First Geological Brigade of Guangdong Geological Bureau, Zhuhai 519000, China
*
Authors to whom correspondence should be addressed.
Sustainability 2024, 16(17), 7501; https://doi.org/10.3390/su16177501
Submission received: 4 August 2024 / Revised: 22 August 2024 / Accepted: 27 August 2024 / Published: 29 August 2024
Browse Figures
Versions Notes

Abstract

:
Geothermal energy, with its promise of sustainability and a minimal environmental impact, offers a viable alternative to fossil fuels that can allow us to meet the increasing energy demands while mitigating concerns over climate change. Urban areas, with their large energy consumption, stand to benefit significantly from the integration of geothermal systems. With the growing need to harness renewable energy sources efficiently, the detection of urban subsurface resources represents a critical frontier in the pursuit of sustainability. The Guangdong Bay area, known for its abundant geothermal resources, stands at the forefront of this green energy revolution, so, in our study, we chose to evaluate Zhuhai City, which is a city representative of the resource-rich area of Guangdong. With the progress of geographic information system (GIS) technology, the land surface temperature (LST) has been used to monitor the spatial distribution characteristics of geothermal anomalies. However, relatively few studies have been conducted in the field of urban geothermal resources. In this study, we calculated the LST of Zhuhai City using Landsat 8 remote sensing data and then investigated the distributions of geothermal hot springs. Spatial data layers were constructed, including the geological structure, DEM and derivatives, lithology, and urban regions, and, based on technology with the integration of machine learning, their spatial correlations with geothermal anomalies were analyzed. The support vector machine (SVM) and the multilayer perceptron (MLP) were employed to produce maps of potential geothermal resources, and their susceptibility levels were divided into five classes: very low, low, moderate, high, and very high. Through model interpretation, we found the moderate-susceptibility class to dominate at 26.90% (SVM) and 46.27% (MLP) according to the two models. Considering the influence of artificial areas, we also corrected the original LST by identifying urban areas of thermal anomalies via the urban thermal anomaly leapfrog fusion extraction (UTALFE) method; following this augmentation, the results shifted to 24.16% (SVM) and 28.67% (MLP). Meanwhile, the area under the curve (AUC) values of all results were greater than 0.65, showing the superior performance and the high applicability of the chosen study area. This study demonstrates that data-driven models integrating thermal infrared remote sensing technology are a promising tool for the mapping of potential urban geothermal resources for further exploration. Moreover, after correction, the reclassified LST results of urban areas are more authentic and suitable for the mapping of potential geothermal resources. In the future, the method applied in this study may be considered in the exploration of more southeastern coastal cities in China.

1. Introduction

In the quest for sustainable and eco-friendly energy solutions, geothermal resources stand out as viable alternatives that are capable of meeting our escalating energy demands [1,2]. The development and utilization of energy stored in the form of heat, derived from the Earth’s internal heat, has mostly relied on direct observation via field work and drilling, which are costly and restricted by accessibility [3,4]. Therefore, many researchers have tried to develop methodologies that use remote sensing information. As one of the most significant surface environmental parameters, the land surface temperature (LST) can be derived from satellite information and used to monitor the spatial distribution characteristics of geothermal anomalies. Many studies have found this to be a cost-effective method for the geothermal exploration of large areas, and it is becoming more commonplace [5,6,7]. With the progress in space-related information technology, geographic information system (GIS) technology has been used in the research of thermal resources [8,9]. Coolbaugh et al. (2004, 2007) combined the LST, Bouguer gravity, a digital elevation model (DEM), faults, seismic activity, and other factors to predict and evaluate areas with geothermal potential in Nevada and the Western Basin; thus far, a total of 170 potential areas have been found [10,11]. Mutua et al. (2011) demonstrated that multi-spectral and multi-spatial satellite images can be used in the mapping of potential geothermal regions [12]. Xiong et al. (2017) applied remote sensing techniques to identify the factors controlling geothermal fields in Tengchong City and found a close relationship between geothermal anomalies and the geological conditions [13]. Bian et al. (2021) again emphasized that thermal infrared remote sensing technology can efficiently identify areas of geothermal anomalies and will guide subsequent work exploring geothermal resources [14].
However, relatively few studies have been conducted on its use in the field of urban geothermal resources. Urban areas are increasingly facing the challenge of sustainable energy management, with a growing need to harness renewable energy sources efficiently. The detection of urban areas represents a critical frontier in the sustainable management and utilization of urban subsurface resources [15,16,17]. There are certain restrictions related to thermal infrared remote sensing (TIRS) technology, as the detection of geothermal anomalies can be affected by factors other than the geothermal energy flux, particularly human factors. Urban settings present a complex array of factors that must be considered when surveying and analyzing the geothermal potential; human factors can significantly influence the detection and interpretation of geothermal anomalies. Nonetheless, adopting analysis models integrating thermal infrared remote (TIR) sensing technology for urban geothermal resource exploration brings numerous advantages [18,19,20]. Zhang et al. (2012) used the LST, distance to urban areas, surface slope, and other factors to build a weighted model capable of evaluating areas of geothermal potential in Tengchong, Yunan [21]. Moghaddam (2013) used factors such as volcanos, fault strikes, hot springs, hydrothermal alteration zones, geothermal gradients, and ground heat flows to construct models and conducted a detailed analysis on the advantages and disadvantages of these models [22]. Sadeghi (2015) applied an exponential superposition model and fuzzy logic model to map areas with geothermal potential [23]. Among them, information models have been used to predict the distribution of geological mineral resources and produce estimations.
In this study, we chose to assess Zhuhai City, which is situated along the southeastern coast of China. Based on thermal infrared remote sensing technology, we employed machine learning algorithms to identify factors related to the detection and interpretation of geothermal anomalies. The aim of this work was to test and compare the application of thermal infrared remote sensing technology coupled with data-driven analysis models to detect potential areas of geothermal resources in urban settings. Furthermore, considering the factors interfering in the detection of geothermal resources in urban areas, we also identified the influence of artificial areas in the mapping of potential urban areas of geothermal resources that should be corrected. Furthermore, we anticipate that our combined method will represent a sophisticated approach to accurately identifying geothermal anomalies in urban areas, in which traditional methods may struggle to differentiate between anthropogenic heat and genuine geothermal anomalies. In addition, the results of this study will enhance the detection and evaluation of urban geothermal resources.

2. Materials and Methods

2.1. Study Area

Zhuhai is located at the mouth of the Pearl River Delta, Guangdong Province, with a lengthy coastline, numerous islands, and urban conglomerates. With the rapid urban expansion in recent years, Zhuhai has become a typical urban city renowned for its strategic location and economic vitality [24,25,26]. The geothermal field in the south of the Pearl River Delta is one part of the geotropic along the southeastern coast of China. The geothermal systems found in Guangdong are mostly hydrothermal geothermal systems, which are mainly characterized by the exposure of geothermal fluids [27,28,29], making Zhuhai an ideal case study for the exploration of the viability of geothermal energy in coastal urban areas. By focusing on Zhuhai, our study aims to address the critical need for sustainable energy solutions in rapidly developing coastal cities.
The geothermal resources of Zhuhai are believed to be associated with its geological structures. It is characterized by significant deep fracture zones, which are influenced by the subduction of the Indian Ocean Plate, the Pacific Plate, and the Philippine Sea Plate [24,30]. Based on the research, 16 geothermal hot springs have been identified at present (Figure 1). In the meantime, accompanied by crustal upraise and multi-stage strong magmatic activity, favorable conditions for the occurrence of geothermal resources are created. The mainland of Zhuhai is cut into a landform unit of block uplift and subsidence by the NE and NW faults, forming a block uplift mountain and subsidence plain, and the outer islands are also controlled by the NE tectonic lines [31,32]. In addition, there are still active faults, which have existed since the Quaternary, mainly inheriting the old fractures extending to the NNW, NNE, and NEE, forming the present tectonic pattern. These geological features are conducive to the formation of geothermal systems, as they can trap geothermal fluids and facilitate their circulation [33,34]. Combined with the regional geological data, the strata mainly consist of Cambrian, Devonian, Cretaceous, and Quaternary systems, and the lithology consists of mainly siltstone, sandstone, and mud shale. Moreover, Quaternary sediment has developed widely in the study area [35]. The unique geological, environmental, and policy factors that make Zhuhai an ideal case study provide valuable insights that can be extrapolated to other coastal urban areas.

2.2. Data Source and Methodology

With the literature research and previous expert opinions, the Landsat 8 satellite remote sensing images from February and December 2023 were used in this study, coupled with the topographic and geological map of 1:100,000, a 30 m resolution DEM, and other geological survey reports, as the major data.
We calculated the LST of the study area from the Landsat 8 TIR data and synthesized multiple types of remote sensing information and geological maps to obtain more environmental variables, such as the geological structure, elevation, lithology, and so on. A support vector machine (SVM) and multilayer perceptron (MLP) were employed to produce the geothermal resource potential maps. Finally, the receiver operating characteristic (ROC) curve was utilized to assess the accuracy of the model predictions. Then, urban geothermal anomaly susceptibility maps were produced (Figure 2).

2.3. Geological Indicators

The geothermal system comprises a setup in which magma from the interior of the Earth is transferred to the surface from the fractured and weak zones, as well as accumulating in the permeable reservoir rock after heating in the depths [36,37,38]. This heat transfer is primarily facilitated by a combination of geological conditions, such as the geological structures, lithology, DEM, and derivatives. These geological indicators serve as critical parameters in the identification and analysis of geothermal resources and their anomalies [39,40].

2.3.1. Geological Structure

In the geothermal system, the geological structure, including faults and fractures, is one of the most important parameters, enabling the passage of geothermal fluids [41]. Combing the DEM data and multi-source remote sensing data, we interpret the faults of Zhuhai City and split them into known geological faults, buried geological faults, and supposed geological faults, which all can provide a good channel for fluid transport.

2.3.2. Analysis of Digital Elevation Models

DEMs and their derivatives are crucial in understanding the surface characteristics and inferring the subsurface geological conditions that can influence geothermal activity [42,43]. The mainland of Zhuhai is divided into three parts, namely the eastern hilly valley, the central hilly plain belt, and the western low hilly plain belt. We analyzed the elevation and calculated the slope and aspect for subsequent classification.

2.3.3. Lithology

The lithology, which refers to the physical and chemical composition of rocks, significantly influences the thermal properties of the Earth’s crust and is essential in identifying promising geothermal reservoirs [44,45]. According to the geological map, coupled with existing reports, we interpreted the lithology of the study area based on satellite images [46,47].

2.4. Land Surface Temperature

The temperature at the top of the Earth’s surface form the TIR bands is essential to define geothermal fields. In the detection of urban geothermal anomalies, the first step is to calculate the LST values [48,49,50]. We employed the single-window (SW) algorithm to extract the surface temperature in this study as follows:
T δ = a ( 1 − C − D ) + ( b ( 1 − C − D ) + C + D ) T a C
where a = −67.355351; b = 0.458606; C = ε δ τ δ ; D = ( 1 − τ δ )[1 + τ δ (1 − ε δ )]; ε δ is the emissivity of the land surface; τ δ is the atmospheric transmissivity; and Ta is the average action temperature. The SW algorithm formula applicable to the Landsat 8 thermal infrared band is as follows [51,52,53]:
T s = γ [ φ 1 L sen + φ 2 ε + φ 3 ] + δ
γ   ≈   T sen 2 / b γ L sen
δ = T sen − T sen 2 / b γ
where ε is the surface-specific emissivity; γ is a parameter determined by Equation (3); δ is a parameter determined by Equation (4); for TIRS Band 10 and Band 11, bγ is 1324 K and 1199 K, respectively; L sen is the spectral radiation value corresponding to the image element (W·m−2·sr−1·μm−1); T sen represents the brightness temperature value after conversion from the thermal infrared band; and φ 1 , φ 2 , φ 3 are the atmospheric function parameters, respectively.

2.5. Urban Region Correction

According to the previous studies, LST anomalies can be distinguished into two types: non-man-made geothermal anomaly areas and man-made geothermal anomaly areas, such as airports, shopping malls, and so on [54,55]. To eliminate the effects of urban heat islands, it is necessary to correct the original LST by identifying the urban thermal anomaly areas through urban thermal anomaly leapfrog fusion extraction (UTALFE) [56]. By setting the sliding window area of different orders, the extraction results of multiple thermal anomalies were fused, taking into account the extraction effects of large and small windows, so as to obtain a more complete range of urban thermal anomalies [57,58].
D = S 10 k · d − 2
where D is the sliding window size; S is the total area of the study area (km2); d is the area of a single unit (km2); and k is a non-negative integer, indicating the number of steps extracted by fusion.
UHI = UHI 0 ∪   UHI 1 ∪   UHI 2 … … ∪   UHI k
where UHI is the total range of urban thermal anomalies; UHI 0 ,   UHI 1 , … … , UHI k is the thermal anomaly range obtained by leapfrog calculation.

2.6. Analysis Models

Data-driven models, characterized by their ability to process vast amounts of data and uncover complex patterns, have emerged as effective tools in geothermal prospecting. The application of information models mostly adopts the form of factor superposition, and the reliability of the predictive models is contingent upon the integrity of the underlying data [59,60,61,62]. In this study, we employed SVM and MLP to calculate the susceptibility.

2.6.1. Support Vector Machine (SVM)

The support vector machine (SVM) is widely used to carry out classification and regression tasks. As a supervised machine learning algorithm, it can identify the optimal classification hyperplane within the sample space [63,64,65]. Thus, in the field of prediction mapping, SVMs are useful in identifying patterns and classifying areas based on the potential for geothermal springs by constructing hyperplanes in a multidimensional space that separates different classes, and they have been extensively utilized in the analysis of susceptibility [66,67,68,69].

2.6.2. Multilayer Perceptron (MLP)

The MLP is one of the most popular and widely used artificial neural networks (ANN) and consists of an input layer, an output layer, and one or more intermediate layers. These intermediate layers are hidden layers between the output and input layers, increasing the network’s ability to model complex functions [70,71,72]. In the MLP, there is no definite algorithm to determine the number of hidden layers and the number of neurons and this is often done by trial and error. The training process of the MLP is initiated by assigning arbitrary initial connection weights, which are constantly updated to reach acceptable and stable training accuracy [73,74,75].

3. Results

3.1. Geological Characteristics

Combing the DEM data and multi-source remote sensing data, we interpret the geological characteristics of the study area. The results show that the directions of the faults in Zhuhai City are mainly NW, NE, and NNE. They can be split into known geological faults, buried geological faults, and supposed geological faults (Figure 3a). In addition, the buffer distance to a geological structure is a crucial parameter in geothermal anomaly detection. A smaller buffer distance may result in a higher density of detected anomalies near the geological structure, while a larger buffer distance may encompass a broader area but potentially miss subtle anomalies closer to the structure. In this study, we convert all structure interpretations into a grid format, and the buffer zone is divided into six classes at 500 m intervals from the central lines within the study area (Figure 3b).
We calculated various DEMs and derivatives and divided them into different classes with suitable deviations. The elevation ranged from −7.0 to 561 m, and the slope ranged from 0° to 50.13° and was grouped into eight classes according to 1/2 standard deviation (Figure 4a,b). Meanwhile, the aspect of the slope, which generally refers to the steepness, incline, or gradient of a surface or line, was categorized into nine distinct classes (Figure 4c).
According to the geological map, coupled with existing reports in the study area, the interpretation of the lithology indicates that Zhuhai’s lithological diversity includes sedimentary, metamorphic, and igneous rocks (Figure 5). In the mainland of Zhuhai, the sediments of the Quaternary are widely developed, and the intrusions of acidic magma in the Middle Yanshan period were intense.

3.2. LST and Threshold Segmentation

The satellite data used in this study were Landsat 8 data with a resolution of 60 m, from February and December 2023. We employed the single-window (SW) algorithm to extract the surface temperature, and the results were 12.3375–42.1688 °C and 12.0584–35.2906 °C, respectively (Figure 6).
Furthermore, the correlation coefficient can indicate the predictability between the temperature changes in the two months and provides evidence of the temperature consistency. In this study, we selected 500 random points to compare the LST values of these two months, and there was a good linear relationship between the two in general, reaching 0.7336, showing high consistency (Figure 7). This high consistency suggests that the LST values across these two months are closely related, providing important insights into the significance of the linear correlation.
Taking the high consistency into consideration, there is a similar trend among the LST values’ distribution. During winter, many regions experience a significant reduction in vegetation cover and often exhibit more stable atmospheric conditions with less turbulence and fewer convective weather events compared to summer. By leveraging the unique conditions of winter, the images obtained in winter are more suitable for computation. Specifically, the results indicate that December experiences lower temperatures, making anomalies in the surface temperature potentially easier to detect and identify. Therefore, in our subsequent research efforts, we select the LST results from December to proceed with the next phase of analysis.
According to previous studies, LST anomalies can be influenced by artificial areas. The most common artificial areas in Zhuhai City can be divided into farmland, industrial areas, commercial areas, and residential areas (Table 1, Figure 8a), and we consider the remaining areas as natural features.
We applied Equation (5) to eliminate the thermal anomalies in the main urban areas, especially the interference of the urban heat island effect, and reclassified the LST results for the next stage (Figure 8b).

3.3. Geothermal Potential Anomaly Mapping

Based on the identified geothermal hot springs in the study area, SVM and MLP data-driven models are employed to predict the geothermal probability. We not only the compare the results between different models but also the results based on the reclassified and original LST values. The geothermal potential mapping outputs are categorized into five susceptibility levels according to natural breaks classification: very low, low, moderate, high, and very high (Figure 9).
For the results calculated based on the reclassified and original LST values in the study area, the proportion of the moderate level is always the largest (Table 2). Although the results for the five susceptibility levels in the above models are different, the area with the highest level is always located in the middle of the city and expands towards the northeast, while the susceptibility level of the southeast and northwest of Zhuhai is lower. Meanwhile, the identified geothermal hot springs are mostly in the areas with the highest and high levels, showing the greatest possibility for geothermal energy, and this is also an indicator that can be used to detect urban geothermal anomalies.
Whether the AUC value of the SVM model or MLP model is applied above, the result calculated based on the reclassified LST has higher precision than that using the original LST and is more than 0.65 (Figure 10), showing that it is necessary to take urban thermal anomaly leapfrog fusion into consideration. Simultaneously, the AUC value of SVM is the highest, reaching 0.815.

4. Discussion

At present, it is evident that LST anomalies correlate with geothermal anomalies, meaning that we can monitor the spatial distribution. However, in reality, the transfer of heat from geothermal systems to the surface is primarily facilitated by a combination of geological conditions. With the development of GIS technology, synthesized multi-remote sensing information technology is a cost-effective means of exploring large areas of geothermal potential [76,77]. The detection of geothermal resources in urban areas will play a significant role in the future sustainable management and utilization of energy [78,79]. However, relatively little research has been conducted on the simultaneous application of various methods of detecting geothermal resources. In this work, we present the application of Landsat-8-derived LST values from Zhuhai City, adopting SVM and MLP models to evaluate areas of geothermal potential. Considering the influence of various factors on the LST and geothermal anomaly mapping, particularly the urban heat island effect, we corrected the LST through UTALFE, so as to obtain more accurate results [80,81].
We produced Zhuhai’s geothermal susceptibility maps through the above technology, and our results demonstrate that analysis models integrating TIR technology can be used to identify the spatial distribution of urban geothermal anomalies; they are an effective technical means of carrying out geothermal pre-surveys [82,83,84]. Furthermore, we observed more objective differentiation among the various models, which allowed for the derivation of superior model combinations. Additionally, in this study area, the AUC value of the SVM reached a peak of 0.815, showing optimal applicability. Based on the results of various analysis models, this method can avoid the one-sidedness of geothermal surveys conducted using a single method, and it greatly improves the efficiency and reduces the cost of investing areas with geothermal resources. Geothermal energy offers a reliable and continuous energy source, which is crucial for China’s energy security, especially in coastal urban areas prone to energy demand fluctuations. Coastal urban areas face unique challenges, such as high population densities and limited space. Our findings provide a refined methodology for the identification of optimal locations for geothermal energy extraction, and we hope to provide a template for other researchers to follow in different geographical and climatic contexts.
However, there are also some limitations to the methodology employed in this study. Previous studies have reported that many algorithms can be used to calculate the Landsat 8 LST, and these may result in different values. Within the same context, some areas might produce LST errors and influence the LST thresholds. Other sources of uncertainty are related to the emissivity, which is relatively constant in natural landscapes and urban regions. In this study, we classified the urban regions into four classes, and further research on different land cover types is needed to validate the LST thresholds. We also lacked comprehensive time series data for the study area; therefore, different locations were estimated using different time series. In future research, we will analyze and interpret the root causes of the geothermal anomalies and storage conditions, focus on the full potential of multi-source data, and delve into the intrinsic relationships between various factors influencing data-driven models. Additionally, more information can be obtained via field investigations, and we expect the above combination to offer a sophisticated approach to accurately identifying geothermal anomalies in urban areas and enhance the detection and evaluation of urban geothermal resources.

5. Conclusions

In this study, data-driven models integrating TIR technology were employed to produce maps of potential urban geothermal resources in Zhuhai City, and the following conclusions were drawn.
(1)
Applying TIR technology to identify the spatial distribution of the LST is both feasible and effective, and, through UTALFE, we can reclassify the LST results to eliminate the effects of urban heat islands for the future detection of geothermal resources in urban areas.
(2)
SVM and MLP data-driven models have been used to predict the geothermal probability. Through model interpretation, the moderate-susceptibility class is found to dominate, at 26.90% (SVM) and 46.27% (MLP), and the results shift to 24.16% (SVM) and 28.67% (MLP) after the correction of the original LST by identifying urban areas of thermal anomalies via the UTALFE method.
(3)
Although the results regarding the five susceptibility levels for the above models are different, the highest results are consistently located in the middle of the city, expanding towards the northeast. In our subsequent study, we will focus on these areas in our fieldwork.
(4)
We found that the results calculated based on the reclassified LST had higher precision than the original LST values, and the AUC value of SVM was the highest, at 0.815, showing optimal applicability for the mapping of potential geothermal resources. In the future, the method applied in this study may be considered to carry out prospective work in the exploration of more southeastern coastal cities in China.
The application of thermal infrared remote sensing technology coupled with analysis models, as described in this study, not only sets a precedent for future urban geothermal anomaly mapping in similar cities, but also contributes to the broader field of sustainable and eco-friendly energy. In the future, we will continue to integrate additional machine learning models and corroborate the causes and mechanisms of the identified geothermal anomalies for the further exploration of interior geothermal resources.

Author Contributions

Y.B.: conceptualization, methodology, formal analysis, data curation, writing—original draft, writing—review and editing, visualization. Y.Y.: data curation, conceptualization, supervision, and funding acquisition. L.C.: validation, project administration, writing—review and editing. Y.N.: validation, project administration, resources. Y.G.: investigation, formal analysis, project administration. J.C.: writing—review and editing, funding acquisition. J.W.: writing—review and editing, supervision. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Project of Zhuhai City Geological Survey (including information) (grant number: 440401-2022-02369 [MZCD-2201-008]) and the Major Projects of High Resolution Earth Observation (grant number: 30-H30C01-9004-19/21).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are contained within the article.

Acknowledgments

We thank Yong Ni for his assistance with data collection and study conceptualization and particularly for valuable discussions.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Gao, Y.F.; Li, B.Z.; Wang, Z.H.; Yang, Y.L. Thermal environment analysis of mountainous towns based on GIS. J. Cent. S. Univ. 2009, S1, 230–234. [Google Scholar]
  2. Jurišić, M.; Frangeš, S.; Plaščak, I.; Šiljeg, A. Methodology of Development of Purpose Maps in GIS Environment–Resource Management. Geod. List 2013, 1, 1–12. [Google Scholar]
  3. Gemitzi, A.; Dalampakis, P.; Falalakis, G. Detecting geothermal anomalies using Landsat 8 thermal infrared remotely sensed data. Int. J. Appl. Earth Obs. Geoinf. 2021, 96, 102283. [Google Scholar] [CrossRef]
  4. Liu, J.Y.; Chao, J.Q.; Zhao, Z.F.; Xu, S.; Lai, Z.; Yang, H.; Chen, Q.; Tu, Y. Identification of geothermal potential based on land surface temperature derived from remotely sensed data. Environ. Sci. Pollut. Res. 2023, 47, 104726–104741. [Google Scholar] [CrossRef] [PubMed]
  5. Hamamoto, H.; Yamno, M.; Goto, S.; Taniguchi, M. Estimation of the past ground surface temperature history from subsurface temperature distribution: Application to the Bangkok area. Geophys. Explor. 2009, 62, 575–584. [Google Scholar]
  6. Liu, W.Y. Development of Geothermal Resources in Hongtong Realize Energy Saving and Emission Reduction. Shanxi Coking Coal Sci. Technol. 2011, 35, 54–56. [Google Scholar]
  7. Rae, A.J.; Rosenberg, M.D.; Bignall, G.; Kilgour, G.N.; Milicich, S.D. Geological results of production well drilling in the western steamfield, Ohaaki geothermal system: 2005–2007. In Proceedings of the 29th New Zealand Geothermal Workshop; University of Auckland: Auckland, New Zealand, 2007. [Google Scholar]
  8. Xiao, D.; Wei, H.; Shi, S.M.; Zhu, H.; Gan, S.; Wu, X.; Feng, W. Research of Geothermal Resources on Hot Dry Rock in North Songliao Basin of China. Int. J. Energy 2023, 2, 1–4. [Google Scholar] [CrossRef]
  9. Vaccaro, M. Multidisciplinary Approach for the Sustainable Utilization of Medium-Low Temperature Geothermal Resources. Ph.D. Thesis, University of Pisa, Pisa, Italy, 2013. [Google Scholar]
  10. Coolbaugh, M.F.; Raines, G.L.; Zehner, R.E. Assessment of exploration bias in data-driven predictive models and the estimation of undiscovered resources. Nat. Resour. Res. 2007, 16, 199–207. [Google Scholar] [CrossRef]
  11. Coolbaugh, M.F.; Shevenell, L.A. A method for estimating undiscovered geothermal resources in nevada and the great basin. Geotherm. Resour. Counc. Trans. 2004, 28, 13–18. [Google Scholar]
  12. Mutua, J.; Mibei, G. Remote sensing application in geothermal exploration: Case study of Barrier volcanic complex Kenya. Trans.-Geotherm. Resour. Counc. 2011, 35, 943–947. [Google Scholar]
  13. Xiong, Y.Z.; Chen, F.; Huang, S.P. Application of remote sensing technique to the identification of geothermal anomaly in Tengchong area, southwest China. J. Chengdu Univ. Technol. (Sci. Technol. Ed.) 2016, 43, 109–118. [Google Scholar]
  14. Bian, Y.; Yang, Y.P.; Li, M.; He, X.; Tang, H.; Sun, A.; Ju, X. Application of thermal infrared remote sensing techniques in geothermal resources surveying. China Min. Mag. 2021, 5, 153–157. [Google Scholar]
  15. Papale, R.; Parri, R.; Xodo, L.; Pesenti, D.; Tori, G.; Brogi, R.; Giannecchini, L. A Remote-Diagnostics and Predictive Maintenance Solution for Geothermal Power Plants. In Proceedings of the 8th African Rift Geothermal Conference, Nairobi, Kenya, 2–8 November 2020. [Google Scholar]
  16. Gao, W.; Zhao, J. Prediction of geothermal temperature field by multi-attribute neural network. Geotherm. Energy 2024, 12, 22. [Google Scholar] [CrossRef]
  17. Tang, Q.Y.; Zhang, Y. Assessment algorithm of the regional geothermal based on neural network. Renew. Energy Resour. 2013, 31, 90–93. [Google Scholar]
  18. Tura, N.; Ojanen, V.; Paloviita, T.; Piiparinen, S. Multiple Helix Approach in Advancing Sustainable Urban Energy Ecosystems. In Proceedings of the 2018 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM), Bangkok, Thailand, 16–19 December 2018; IEEE: Piscataway, NJ, USA, 2018. [Google Scholar]
  19. Theodoridou, I.; Karteris, M.; Mallinis, G.; Papadopoulos, A.M.; Hegger, M. Assessment of retrofitting measures and solar systems’ potential in urban areas using Geographical Information Systems: Application to a Mediterranean city. Renew. Sustain. Energy Rev. 2012, 16, 6239–6261. [Google Scholar] [CrossRef]
  20. Cumo, F.; Garcia, D.A.; Calcagnini, L.; Rosa, F.; Sferra, A.S. Urban policies and sustainable energy management. Sustain. Cities Soc. 2012, 4, 29–34. [Google Scholar] [CrossRef]
  21. Zhang, N.; Qin, Q.; He, L.; Jiang, H. Remote sensing and GIS based geothermal exploration in southwest Tengchong, China. In Proceedings of the IEEE International Geoscience and Remote Sensing Symposium, Munich, Germany, 22–27 July 2012; IEEE: Piscataway, NJ, USA, 2012; pp. 5364–5367. [Google Scholar]
  22. Moghaddam, M.; Pourbaige, M.; Tabar, H.K.; Farhadi, N.; Hosseini, S.M.A. Composition and Antifungal Activity of Peppermint (Mentha piperita) Essential Oil from Iran. J. Essent. Oil Bear. Plants 2013, 16, 506–512. [Google Scholar] [CrossRef]
  23. Sadeghi, B.; Gholamhoseinpoor, F. A study on the stability and green synthesis of silver nanoparticles using Ziziphora tenuior (Zt) extract at room temperature. Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 2015, 134, 310. [Google Scholar] [CrossRef]
  24. Zheng, W.B.; Deng, H.W. High-resolution sequence stratigraphic division and distribution of tidal deposits in the Zhuhai Formation, Huizhou Sag, Pearl River Mouth Basin, South China Sea. Geol. Mag. 2012, 149, 722–728. [Google Scholar] [CrossRef]
  25. Luo, C.W.; Huang, L.C.; Heng, J.W.; Huang, H.W.; Cai, S. The influence of temperature on geothermal power plant performance based on geology in Huangshadong geothermal field, China. Therm. Sci. 2023, 27, 4361–4373. [Google Scholar] [CrossRef]
  26. Lu, G.P.; Liu, R.F. Aqueous Chemistry of Typical Geothermal Springs with Deep Faults in Xinyi and Fengshun in Guangdong Province, China. J. Earth Sci. 2015, 26, 60–72. [Google Scholar] [CrossRef]
  27. Xiong, X.B.; Qin, Y.Z.; Sheng, X.J.; Liu, Y.P.; Xu, Q. Research on Engineering Geology Teaching—Take Hong Kong-Zhuhai-Macau Bridge and Tunnel Engineering as a Case. Educ. Res. Front. 2014, 4, 137–142. [Google Scholar]
  28. Xie, L.H.; Lin, C.S.; Dong, W.; Qin, C.G.; Zhang, Z.T.; Liang, X. High-resolution sequence stratigraphy of Zhujiang Formation and Zhuhai Formation in Panyu low uplift of the Zhujiang River mouth basin. Geol. China 2009, 36, 366–377. [Google Scholar]
  29. Chen, C.; Zheng, Z.; Zeng, L.F.; Xiao, F.; Tian, L.P.; Huang, K.Y. A combined geophysical and lithological study on eruptive history and Quaternary lacustrine stratigraphy of a maar in Leizhou Peninsula, China. J. Palaeogeogr. 2021, 10, 1–14. [Google Scholar] [CrossRef]
  30. Xie, H.; Yang, Z.; Deng, J. Assessment of Geothermal Resource Potential in the Guangdong-Hong Kong-Macao Greater Bay Area. Adv. Eng. Sci. 2019, 51, 5–12. [Google Scholar]
  31. Yan, X.; Lin, W.; Gan, H.; Yue, G.; Wang, G. Hydrogeochemical Characteristics of Huangshadong Geothermal Filed in Guangdong. IOP Conf. Ser. Earth Environ. Sci. 2019, 237, 032128. [Google Scholar] [CrossRef]
  32. Luo, W.; Fang, Y.; Gong, C.; Li, M.H.; Hu, X.Y. Study on geological characteristics of Sankeng geothermal field in Qingyuan, Guangdong Province and investigation practice. Yangtze River 2015, 46, 22–25. [Google Scholar]
  33. Chen, W.Y.; Jim, C.Y. Cost–benefit analysis of the leisure value of urban greening in the new Chinese city of Zhuhai. Cities 2008, 25, 298–309. [Google Scholar] [CrossRef]
  34. Chen, Z.; Davey, G. Subjective Quality of Life in Zhuhai City, South China: A Public Survey Using the International Wellbeing Index. Soc. Indic. Res. 2009, 91, 243–258. [Google Scholar] [CrossRef]
  35. Liu, B.J. Characteristics of continental delta deposits in Zhuhai Formation of Baiyun Depression in Pearl River Mouth Basin. Acta Pet. Sin. 2007, 28, 49–51. [Google Scholar]
  36. Bignall, G.; Milicich, S.D.; Ramirez, L.E.; Rosenberg, M.; Kilgour, G.; Rae, A. Geology of the Wairakei-Tauhara Geothermal System, New Zealand. In Proceedings of the Proceedings World Geothermal Congress 2010, Bali, Indonesia, 25–29 April 2010. [Google Scholar]
  37. Waibel, A.F. An overview of the geology and secondary mineralogy of the high temperature geothermal system in Dixie Valley, Nevada. Res. Gate 1987, 11, 479–486. [Google Scholar]
  38. Li, X.Y.; Cai, H.T.; Ren, Q.L. A Tentative Discussion on the Geothermo–Geological Characteristics and Development of Geothermal Resources in Cangzhou City. Acta Geosci. Sin. 2000, 21, 171–176. [Google Scholar]
  39. Bai, Y.C.; Li, Y.L.; Qi, F.L.; Zhang, F.L. Analysis of metallogenic conditions of geothermal resources in heiyu lake of daqing. Adv. Mater. Res. 2012, 550–553, 2472–2477. [Google Scholar] [CrossRef]
  40. Wood, C.P. Geology of the rotorua geothermal system. Geothermics 1992, 21, 25–41. [Google Scholar] [CrossRef]
  41. Bertini, G.; Casini, M.; Gianelli, G.; Pandeli, E. Geological structure of a long_iving geothermal system, Larderello, Italy. Terra Nova 2006, 18, 163–169. [Google Scholar] [CrossRef]
  42. Coolbaugh, M.F.; Kratt, C.; Fallacaro, A.; Calvin, W.M.; Taranik, J.V. Detection of geothermal anomalies using advanced spaceborne thermal emission and reflection radiometer (aster) thermal infrared images at Bradys hot springs, Nevada, USA. Remote Sens. Environ. 2007, 106, 350–359. [Google Scholar] [CrossRef]
  43. Kuria, Z.N.; Woldai, T.; Meer, F.D.V.D.; Barongo, J.O. Active fault segments as potential earthquake sources: Inferences from integrated geophysical mapping of the Magadi fault system, southern Kenya rift. J. Afr. Earth Sci. 2010, 57, 345–359. [Google Scholar] [CrossRef]
  44. Manzella, A. Geophysical Methods in Geothermal Exploration; Italian National Research Council International Institute for Geothermal Research: Pisa, Italy, 2005. [Google Scholar]
  45. D’Antonio, M. Lithology of the basement underlying the Campi Flegrei caldera: Volcanological and petrological constraints. J. Volcanol. Geotherm. Res. 2011, 200, 91–98. [Google Scholar] [CrossRef]
  46. Chorowicz, J.; Lavreau, J.; Tamain, G.; Theunissen, K.; Bardinet, C.; Girault, F.; Klerkx, J.; Le Page, A.; Nanyaro, J.; Nimpagaritse, G.; et al. The evaluation of satellite imagery and field Spectro-radiometer data for the study of the lithology and the ancient and reactivation tectonics in the vicinity of the east African rift. ISPRS J. Photogramm. Remote Sens. 1990, 45, 355–381. [Google Scholar] [CrossRef]
  47. Mawardi, S.; Sukiyah, E.; Haryanto, I. Morphotectonic characteristics of Cisadane Watersshed based on satellite images analysis. J. Geol. Dan Sumberd. Miner. 2019, 20, 175. [Google Scholar] [CrossRef]
  48. Chakrabarti, A.; Cormode, G.; Mcgregor, A. A near-optimal algorithm for estimating the entropy of a stream. ACM Trans. Algorithms 2010, 6, 145–148. [Google Scholar] [CrossRef]
  49. Jimenez-Munoz, J.C.; Sobrino, J.A.; Skokovic, D.; Mattar, C.; Cristobal, J. Land surface temperature retrieval methods from landsat-8 thermal infrared sensor data. IEEE Geosci. Remote Sens. Lett. 2014, 11, 1840–1843. [Google Scholar] [CrossRef]
  50. Wang, S.; He, L.; Hu, W. A temperature and emissivity separation algorithm for landsat-8 thermal infrared sensor data. Remote Sens. 2015, 12, 9904–9927. [Google Scholar] [CrossRef]
  51. Nishar, A.; Richards, S.; Breen, D.; Robertson, J.; Breen, B. Thermal infrared imaging of geothermal environments and by an unmanned aerial vehicle (UAV): A case study of the Wairakei–Tauhara geothermal field, Taupo, New Zealand. Renew. Energy 2016, 86, 1256–1264. [Google Scholar] [CrossRef]
  52. Sobrino, J.A.; Jiménez-Muñoz, J.C. Minimum configuration of thermal infrared bands for land surface temperature and emissivity estimation in the context of potential future missions. Remote Sens. Environ. Interdiscip. J. 2014, 148, 158–167. [Google Scholar] [CrossRef]
  53. Sobrino, J.A.; Jiménez-Muñoz, J.C.; Paolini, L. Land surface temperature retrieval from LANDSAT TM 5. Remote Sens. Environ. Interdiscip. J. 2014, 90, 434–440. [Google Scholar] [CrossRef]
  54. Chen, L.; Bai, Z.P.; Su, D.; You, Y.; Li, H.; Liu, Q. Application of land use regression to simulate ambient air PM10 and NO2 concentration in Tianjin City. China Environ. Ence 2009, 29, 685–691. [Google Scholar]
  55. Jiménez-Muñoz, J.C.; Sobrino, J.A.; Gillespie, A.R. Surface emissivity retrieval from airborne hyperspectral scanner data: Insights on atmospheric correction and noise removal. IEEE Geosci. Remote Sens. Lett. 2012, 9, 180–184. [Google Scholar] [CrossRef]
  56. Arola, T.; Korkka-Niemi, K. The effect of urban heat islands on geothermal potential: Examples from quaternary aquifers in finland. Hydrogeol. J. 2014, 22, 1953–1967. [Google Scholar] [CrossRef]
  57. Changnon, S.A.J.; Huff, F.A. The urban-related nocturnal rainfall anomaly at St. Louis. J. Clim. Appl. Meteorol. 2010, 25, 1985–1995. [Google Scholar] [CrossRef]
  58. Allen, A.; Milenic, D.; Sikora, P. Shallow gravel aquifers and the urban ‘heat island’ effect: A source of low enthalpy geothermal energy. Geothermics 2003, 32, 569–578. [Google Scholar] [CrossRef]
  59. Achour, Y.; Pourghasemi, H.R. How do machine learning techniques help in increasing accuracy of landslide susceptibility maps? Geosci. Front. 2019, 11, 871–883. [Google Scholar] [CrossRef]
  60. Mahdadi, F.; Boumezbeur, A.; Hadji, R.; Kanungo, D.P.; Zahri, F. GIS-based landslide susceptibility assessment using statistical models: A case study from Souk Ahras province, NE Algeria. Arab. J. Geosci. 2018, 11, 476. [Google Scholar] [CrossRef]
  61. Daya, S.B.; Cheng, Q.; Agterberg, F. Handbook of Mathematical Geosciences; Springer International Publishing: Cham, Switzerland, 2018. [Google Scholar]
  62. Tien Bui, D.; Tuan, T.A.; Klempe, H.; Pradhan, B.; Revhaug, I. Spatial prediction models for shallow landslide hazards: A comparative assessment of the efficacy of support vector machines, artificial neural networks, kernel logistic regression, and logistic model tree. Landslides 2016, 13, 361–378. [Google Scholar] [CrossRef]
  63. Hong, H.; Liu, J.; Zhu, A.-X.; Shahabi, H.; Pham, B.T.; Chen, W.; Pradhan, B.; Bui, D.T. A novel hybrid integration model using support vector machines and random subspace for weather–triggered landslide susceptibility assessment in the Wuning area (China). Environ. Earth Sci. 2017, 76, 652. [Google Scholar] [CrossRef]
  64. Shevade, S.K.; Keerthi, S.S.; Bhattacharyya, C.; Murthy, K.R.K. Improvements to the SMO algorithm for SVM regression. IEEE Trans. Neural Netw. 2000, 11, 1188–1193. [Google Scholar] [CrossRef] [PubMed]
  65. Tarabalka, Y.; Fauvel, M.; Chanussot, J.; Benediktsson, J.A. SVM and MRF–based method for accurate classification of hyperspectral images. IEEE Geosci. Remote Sens. Lett. 2010, 7, 736–740. [Google Scholar] [CrossRef]
  66. Zhang, X.; Lu, X.; Shi, Q.; Xu, X.Q.; Leung, H.C.E.; Harris, L.N.; Iglehart, J.D.; Miron, A.; Liu, J.S.; Wong, W.H. Recursive svm feature selection and sample classification for mass–spectrometry and microarray data. BMC Bioinform. 2006, 7, 197. [Google Scholar] [CrossRef]
  67. Lee, Y.J.; Mangasarian, O.L. SSVM: A smooth support vector machine for classification. Comput. Optim. Appl. 2001, 20, 5–22. [Google Scholar] [CrossRef]
  68. Keerthi, S.S.; Gilbert, E.G. Convergence of a generalized SMO algorithm for SVM classifier design. Mach. Learn. 2002, 46, 351–360. [Google Scholar] [CrossRef]
  69. Fine, S.; Scheinberg, K. Efficient SVM training using low-rank kernel representations. J. Mach. Learn. Res. 2001, 2, 243–264. [Google Scholar]
  70. Wang, J.; Yang, Y.; Mao, J.H.; Huang, Z.; Huang, C.; Xu, W. CNN-RNN: A Unified Framework for Multi-label Image Classification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 2285–2294. [Google Scholar]
  71. Amid, S.; Gundoshmian, T.M. Prediction of output energies for broiler production using linear regression, ANN (MLP, RBF), and ANFIS models. Environ. Prog. Sustain. Energy 2017, 36, 577–585. [Google Scholar] [CrossRef]
  72. Ouma, Y.O.; Moalafhi, D.B.; Anderson, G.; Nkwae, B.; Odirile, P.; Parida, B.P.; Qi, J. Dam water level prediction using vector AutoRegression, random forest regression and MLP–ANN models based on land–use and climate factors. Sustainability 2022, 14, 14934. [Google Scholar] [CrossRef]
  73. Yilmaz, I.; Kaynar, O. Multiple regression, ANN (RBF, MLP) and ANFIS models for prediction of swell potential of clayey soils. Expert Syst. Appl. 2011, 38, 5958–5966. [Google Scholar] [CrossRef]
  74. Jeong, M.S.; Jang, H.W.; Kim, J.S.; Lee, J.H. Roc evaluation for MLP ANN drought forecasting model. J. Korea Water Resour. Assoc. 2016, 49, 877–885. [Google Scholar]
  75. Qayyum, F.; Jamil, H.; Iqbal, N.; Kim, D.H.; Afzal, M.T. Toward potential hybrid features evaluation using MLP–ANN binary classification model to tackle meaningful citations. Scientometrics 2022, 127, 6471–6499. [Google Scholar] [CrossRef]
  76. Li, X.; Huang, C.; Chen, W.; Li, Y.; Han, J.; Wang, X.; Bai, X.; Yin, Z.; Li, X.; Hou, P.; et al. Gis model for geothermal advantageous target selection. Sci. Rep. 2023, 13, 6024. [Google Scholar] [CrossRef]
  77. Noorollahi, Y.; Itoi, R.; Fujii, H.; Tanaka, T. Gis model for geothermal resource exploration in Akita and Iwate prefectures, northern japan. Comput. Geosci. 2007, 33, 1008–1021. [Google Scholar] [CrossRef]
  78. Zhao, X.G.; Wan, G. Current situation and prospect of Chinas geothermal resources. Renew. Sustain. Energy Rev. 2014, 32, 651–661. [Google Scholar] [CrossRef]
  79. Tureyen, O.I.; Sarak, H.; Altun, G.; Satman, A. A modeling analysis of unitized production: Understanding sustainable management of single-phase geothermal resources with multiple lease owners. Geothermics 2015, 55, 159–170. [Google Scholar] [CrossRef]
  80. Haak, V.; Ritter, O.; Ritter, P. Mapping the geothermal anomaly on the island of milos by magnetotellurics. Geothermics 1989, 18, 533–546. [Google Scholar] [CrossRef]
  81. Chunhua, L.; Lizhi, Y.; Zhongye, L. Assessment and utilization of geothermal resources in Linqing City, Shandong province. Urban Geol. 2009, 212, 194–208. [Google Scholar]
  82. Romaguera, M.; Vaughan, R.; Ettema, J.; Izquierdo-Verdiguier, E.; Hecker, C.; van der Meer, F. Detecting geothermal anomalies and evaluating LST geothermal component by combining thermal remote sensing time series and land surface model data. Remote Sens. Environ. Interdiscip. J. 2018, 204, 534–552. [Google Scholar] [CrossRef]
  83. Peng, F.; Xiong, Y.Z.; Cheng, Y.X.; Fan, Q.C.; Huang, S.P. Towards application of remote sensing technology in geothermal prospecting in Xilingol in eastern inner Mongolia, NE China. Adv. Mater. Res. 2013, 610–613, 3628–3631. [Google Scholar] [CrossRef]
  84. Zouzias, D.; Miliaresis, G.C.; Seymour, K.S. Probable regional geothermal field reconnaissance in the Aegean region from modern multi-temporal night LST imagery. Environ. Earth Sci. 2011, 62, 717–723. [Google Scholar] [CrossRef]
Sustainability 16 07501 g001
Figure 1. Study area.
Figure 1. Study area.
Sustainability 16 07501 g001
Sustainability 16 07501 g002
Figure 2. Overall flow chart.
Figure 2. Overall flow chart.
Sustainability 16 07501 g002
Sustainability 16 07501 g003
Figure 3. (a) Geological structure; (b) buffer distance to faults.
Figure 3. (a) Geological structure; (b) buffer distance to faults.
Sustainability 16 07501 g003
Sustainability 16 07501 g004
Figure 4. Calculation according to DEM: (a) elevation; (b) slope; (c) aspect.
Figure 4. Calculation according to DEM: (a) elevation; (b) slope; (c) aspect.
Sustainability 16 07501 g004
Sustainability 16 07501 g005
Figure 5. Lithology.
Figure 5. Lithology.
Sustainability 16 07501 g005
Sustainability 16 07501 g006
Figure 6. (a) LST of February; (b) LST of December.
Figure 6. (a) LST of February; (b) LST of December.
Sustainability 16 07501 g006
Sustainability 16 07501 g007
Figure 7. Linear correlation.
Figure 7. Linear correlation.
Sustainability 16 07501 g007
Sustainability 16 07501 g008
Figure 8. (a) Urban regions of the study area; (b) reclassified LST results (based on December).
Figure 8. (a) Urban regions of the study area; (b) reclassified LST results (based on December).
Sustainability 16 07501 g008
Sustainability 16 07501 g009
Figure 9. Susceptibility mapping by the models: (a) SVM; (b) MLP; (c) SVM (reclassified LST results); (d) MLP (reclassified LST results).
Figure 9. Susceptibility mapping by the models: (a) SVM; (b) MLP; (c) SVM (reclassified LST results); (d) MLP (reclassified LST results).
Sustainability 16 07501 g009
Sustainability 16 07501 g010
Figure 10. AUC curves of each model: (a) original LST results; (b) reclassified LST results.
Figure 10. AUC curves of each model: (a) original LST results; (b) reclassified LST results.
Sustainability 16 07501 g010
Table 1. Artificial areas in Zhuhai City.
Table 1. Artificial areas in Zhuhai City.
ClassArea (km2)Image Feature
Farmland121.73Sustainability 16 07501 i001
Industrial66.84Sustainability 16 07501 i002
Commercial3.62Sustainability 16 07501 i003
Residential50.32Sustainability 16 07501 i004
Table 2. Distribution of each susceptibility group.
Table 2. Distribution of each susceptibility group.
ClassBased on Original LSTBased on Reclassified LST
SVMMLPSVMMLP
Very low16.99%11.63%16.71%16.56%
Low23.80%19.01%20.11%22.65%
Moderate26.90%46.27%20.34%28.67%
High22.77%16.32%24.16%18.70%
Very High9.54%6.76%18.68%13.42%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Bian, Y.; Ni, Y.; Guo, Y.; Wen, J.; Chen, J.; Chen, L.; Yang, Y. Urban Geothermal Resource Potential Mapping Using Data-Driven Models—A Case Study of Zhuhai City. Sustainability 2024, 16, 7501. https://doi.org/10.3390/su16177501

AMA Style

Bian Y, Ni Y, Guo Y, Wen J, Chen J, Chen L, Yang Y. Urban Geothermal Resource Potential Mapping Using Data-Driven Models—A Case Study of Zhuhai City. Sustainability. 2024; 16(17):7501. https://doi.org/10.3390/su16177501

Chicago/Turabian Style

Bian, Yu, Yong Ni, Ya Guo, Jing Wen, Jie Chen, Ling Chen, and Yongpeng Yang. 2024. "Urban Geothermal Resource Potential Mapping Using Data-Driven Models—A Case Study of Zhuhai City" Sustainability 16, no. 17: 7501. https://doi.org/10.3390/su16177501

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop