Dynamic Changes, Spatiotemporal Differences, and Ecological Effects of Impervious Surfaces in the Yellow River Basin, 1986–2020
Abstract
:1. Introduction
2. Materials and Methods
2.1. Study Area
2.2. Data
2.2.1. Landsat Data
2.2.2. NTL Data
2.2.3. Impervious Surfaces Products
2.2.4. Other Auxiliary Data
2.3. Methodology
2.3.1. Binary Classification
2.3.2. Accuracy Assessment
2.3.3. Spatiotemporal Dynamics
2.3.4. Construct RSEI
2.3.5. IS and RSEI
3. Results
3.1. Accuracy Assessment of the Classification Results
3.1.1. Based on Validation Samples
3.1.2. Based on Impervious Surfaces Products
3.2. Analysis of Impervious Surfaces Dynamics
3.2.1. Annual Impervious Surfaces Dynamics
3.2.2. Comparison of Impervious Surfaces Fraction
3.2.3. Evaluation of Impervious Surfaces Expansion from 1986 to 2020
3.3. Analysis of the Effects of Impervious Surfaces Expansion by RSEI
3.3.1. Ecological Status in the YRB
3.3.2. The Effects of IS change on RSEI
4. Discussion
4.1. Comparison of the Results between IS-YRB with GAIA and NUACI/GAUD
4.2. Interpretation of the Variation of IS-YRB
4.3. The Effect of EIS on Ecological Conditions in the YRB
5. Conclusions
Supplementary Materials
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Weng, Q.H. Remote sensing of impervious surfaces in the urban areas: Requirements, methods, and trends. Remote Sens. Environ. 2012, 117, 34–49. [Google Scholar] [CrossRef]
- Gong, P.; Li, X.C.; Wang, J.; Bai, Y.; Chen, B.; Hu, T.Y.; Liu, X.P.; Xu, B.; Yang, J.; Zhang, W.; et al. Annual maps of global artificial impervious area (GAIA) between 1985 and 2018. Remote Sens. Environ. 2020, 236, 111510. [Google Scholar] [CrossRef]
- Yang, Q.Q.; Huang, X.; Yang, J.; Liu, Y. The relationship between land surface temperature and artificial impervious surface fraction in 682 global cities: Spatiotemporal variations and drivers. Environ. Res. Lett. 2021, 16, 024032. [Google Scholar] [CrossRef]
- Lu, L.; Peng, G. Urban and air pollution: A multi-city study of long-term effects of urban landscape patterns on air quality trends. Sci. Rep. 2020, 10, 18618. [Google Scholar] [CrossRef]
- Zhou, D.C.; Zhao, S.Q.; Zhang, L.X.; Liu, S.G. Remotely sensed assessment of urbanization effects on vegetation phenology in China’s 32 major cities. Remote Sens. Environ. 2016, 176, 272–281. [Google Scholar] [CrossRef] [Green Version]
- Zheng, Z.H.; Wu, Z.F.; Yang, Z.W.; Francesco, M. Analyzing the ecological enviroment and urbanization characteristics of the Yangtze River Delta Urban Agglomeration based on Google Earth Engine. Acta Ecol. Sin. 2021, 41, 717–729. [Google Scholar] [CrossRef]
- Du, J.Q.; Fu, Q.; Fang, S.F.; Wu, J.H.; He, P.; Quan, Z.J. Effects of rapid urbanization on vegetation cover in the metropolises of China over the last four decades. Ecol. Indic. 2019, 107, 105458. [Google Scholar] [CrossRef]
- Song, Y.S.; Li, F.; Wang, X.K.; Xu, C.Q.; Zhang, J.Y.; Liu, X.S.; Zhang, H.X. The effects of urban impervious surfaces on eco-physiological characteristics of Ginkgo biloba: A case study from Beijing, China. Urban For. Urban Green 2015, 14, 1102–1109. [Google Scholar] [CrossRef]
- Strohbach, M.W.; Doring, A.O.; Mock, M.; Sedrez, M.; Mumm, O.; Schneider, A.K.; Webers, S.; Schroder, B. The “Hidden Urbanization”: Trends of Impervious Surface in Low-Density Housing Developments and Resulting Impacts on the Water Balance, Front. Environ. Sci. Eng. 2019, 7, 29. [Google Scholar] [CrossRef] [Green Version]
- Zhang, L.; Weng, Q.H.; Shao, Z.F. An evaluation of monthly impervious surface dynamics by fusing Landsat and MODIS time series in the Pearl River Delta, China, from 2000 to 2015. Remote Sens. Environ. 2017, 201, 99–114. [Google Scholar] [CrossRef]
- Shahtahmassebi, A.R.; Song, J.; Zhang, Q.; Blackburn, G.A.; Wang, K.; Huang, L.Y.; Pan, Y.; Moore, N.; Shahtahmassebi, G.; Haghighi, R.S.; et al. Remote sensing of impervious surface growth: A framework for quantifying urban expansion and re-densification mechanisms. Int. J. Appl. Earth Obs. Geoinf. 2016, 46, 94–112. [Google Scholar] [CrossRef] [Green Version]
- Elvidge, C.D.; Tuttlle, B.T.; Sutton, P.C.; Baugh, K.E.; Howard, A.T.; Milesi, C.; Bhaduri, B.L.; Nemani, R. Global Distribution and Density of Constructed Impervious Surfaces. Sensors 2007, 7, 1962–1979. [Google Scholar] [CrossRef] [PubMed]
- Huang, X.M.; Schneider, A.; Friedl, M.A. Mapping sub-pixel urban expansion in China using MODIS and DMSP/OLS nighttime light. Remote Sens. Environ. 2016, 175, 92–108. [Google Scholar] [CrossRef]
- Guo, W.; Li, G.Y.; Ni, W.J.; Zhang, Y.H.; Lu, D.S. Exploring improvement of impervious surface estimation at national scale through integration of nighttime light and Proba-V data, GISci. Remote Sens. 2018, 55, 699–717. [Google Scholar] [CrossRef]
- Ma, Q.; He, C.Y.; Wu, J.G.; Liu, Z.F.; Zhang, Q.F.; Sun, Z.X. Quantifying spatiotemporal patterns of urban impervious surfaces in China: An improved assessment using nighttime light data. Landsc. Urban Plan. 2014, 13, 36–49. [Google Scholar] [CrossRef]
- Huang, X.; Yang, J.J.; Li, J.Y.; Wen, D.W. Urban functional zone mapping by integrating high spatial resolution nighttime light and daytime multi-view imagery. ISPRS J. Photogramm. 2021, 175, 403–415. [Google Scholar] [CrossRef]
- Zhang, Q.L.; Seto, K.C. Mapping urbanization dynamics at regional and global scales using multi-temporal DMSP/OLS nighttime light data. Remote Sens. Environ. 2021, 115, 2320–2329. [Google Scholar] [CrossRef]
- Zhao, M.; Zhou, Y.Y.; Li, X.; Cao, W.T.; He, C.Y.; Yu, B.L.; Li, X.; Elvidge, C.; Cheng, W.M.; Zhou, C.H. Applications of Satellite Remote Sensing of Nighttime Light Observations: Advances, Challenges, and Perspectives. Remote Sens. 2019, 11, 1971. [Google Scholar] [CrossRef] [Green Version]
- Gao, F.; De Colstoun, E.B.; Ma, R.H.; Weng, Q.H.; Masek, J.G.; Chen, J.; Pan, Y.Z.; Song, C.H. Mapping impervious surface expansion using medium-resolution satellite image time series: A case study in the Yangtze River Delta, China. Int. J. Remote Sens. 2012, 33, 7609–7628. [Google Scholar] [CrossRef]
- Zhang, X.; Liu, L.Y.; Wu, C.S.; Chen, X.D.; Gao, Y.; Xie, S.; Zhang, B. Development of a global 30m impervious surface map using multisource and multitemporal remote sensing datasets with the Google Earth Engine platform. Earth Syst. Sci. Data 2020, 12, 1625–1648. [Google Scholar] [CrossRef]
- Cao, X.M.; Gao, X.H.; Shen, Z.Y.; Li, R.X. Expansion of Urban Impervious Surfaces in Xining City Based on GEE and Landsat Time Series Data. IEEE Access 2020, 8, 147097–147111. [Google Scholar] [CrossRef]
- Liu, X.P.; Hu, G.H.; Chen, Y.M.; Li, X.; Xu, X.C.; Li, S.Y.; Pei, F.S.; Wang, S.J. High-resolution multi-temporal mapping of global urban land using Landsat images based on the Google Earth Engine Platform. Remote Sens. Environ. 2018, 209, 227–239. [Google Scholar] [CrossRef]
- Xu, H.Z.Y.; Wei, Y.C.; Liu, C.; Li, X.; Fang, H. A Scheme for the Long-Term Monitoring of Impervious−Relevant Land Disturbances Using High Frequency Landsat Archives and the Google Earth Engine. Remote Sens. 2019, 11, 1891. [Google Scholar] [CrossRef] [Green Version]
- Gong, P.; Li, X.C.; Zhang, W. 40-Year (1978–2017) human settlement changes in China reflected by impervious surfaces from satellite remote sensing. Sci. Bull. 2019, 64, 756–763. [Google Scholar] [CrossRef] [Green Version]
- Li, P.L.; Liu, X.P.; Huang, Y.H.; Zhang, H.H. Mapping impervious surface dynamics of Guangzhou downtown based on Google Earth Engine. J. Geo-Inf. Sci. 2020, 22, 638–648. [Google Scholar] [CrossRef]
- Song, X.P.; Sexton, J.O.; Huang, C.Q.; Channan, S.; Townshend, J.R. Characterizing the magnitude, timing and duration of urban growth from time series of Landsat-based estimates of impervious cover. Remote Sens. Environ. 2016, 175, 1–13. [Google Scholar] [CrossRef]
- Huang, X.; Li, J.Y.; Yang, J.; Zhang, Z.; Li, D.R.; Liu, X.P. 30 m global impervious surface area dynamics and urban expansion pattern observed by Landsat satellites: From 1972 to 2019. Sci. China Earth Sci. 2021, 64, 1922–1933. [Google Scholar] [CrossRef]
- Liu, C.; Zhang, Q.; Luo, H.; Qi, S.H.; Tao, S.Q.; Xu, H.Z.Y.; Yao, Y. An efficient approach to capture continuous impervious surface dynamics using spatial-temporal rules and dense Landsat time series stacks. Remote Sens. Environ. 2019, 229, 114–132. [Google Scholar] [CrossRef]
- Kennedy, R.E.; Yang, Z.Q.; Gorelick, N.; Braaten, J.; Cavalcante, L.; Cohen, W.B.; Healey, S. Implementation of the LandTrendr Algorithm on Google Earth Engine. Remote Sens. 2018, 10, 691. [Google Scholar] [CrossRef] [Green Version]
- Wang, Y.L.; Li, M.S. Urban impervious surface detection from remote sensing images a review of the methods and challenges. IEEE Geosci. Remote Sens. Mag. 2019, 7, 64–93. [Google Scholar] [CrossRef]
- Su, S.S.; Tian, J.; Dong, X.Y.; Tian, Q.J.; Wang, N.; Xi, Y.B. An impervious surface spectral index on multispectral imagery using visible and Near-Infrared bands. Remote Sens. 2022, 14, 3391. [Google Scholar] [CrossRef]
- Zhang, L.; Zhang, M.; Yao, Y.B. Mapping seasonal impervious surface dynamics in Wuhan urban agglomeration, China from 2000 to 2016. Int. J. Appl. Earth Obs. Geoinf. 2018, 70, 51–61. [Google Scholar] [CrossRef]
- Zhang, H.S.; Zhang, Y.Z.; Lin, H. Seasonal effects of impervious surface estimation in subtropical monsoon regions. Int. J. Digit. Earth 2014, 7, 746–760. [Google Scholar] [CrossRef]
- Shrestha, B.; Stephen, H.; Ahmad, S. Impervious Surfaces Mapping at City Scale by Fusion of Radar and Optical Data through a Random Forest Classifier. Remote Sens. 2021, 13, 3040. [Google Scholar] [CrossRef]
- Xu, H.Q. A remote sensing urban ecological index and its application. Acta Ecol. Sin. 2013, 33, 7853–7862. [Google Scholar] [CrossRef]
- He, Y.L.; You, N.S.; Cui, Y.P.; Xiao, T.; Hao, Y.Y.; Dong, J.W. Spatio-temporal changes in remote sensing-based ecological index in China since 2000. J. Nat. Resour. 2021, 36, 1176–1185. [Google Scholar] [CrossRef]
- Xu, H.Q.; Wang, M.Y.; Shi, T.T.; Guan, H.D.; Fang, C.Y.; Lin, Z.L. Prediction of ecological effects of potential population and impervious surface increases using a remote sensing based ecological index (RSEI). Ecol. Indic. 2018. 93, 730–740. [CrossRef]
- Chen, W.; Huang, H.P.; Tian, Y.C.; Du, Y.Y. Monitoring and assessment of the eco-environment quality in the Sanjiangyuan region based on Google Earth Engine. J. Geo-Inf. Sci. 2019, 21, 1382–1391. [Google Scholar] [CrossRef]
- Zhu, D.Y.; Chen, T.; Wang, Z.W.; Niu, R.Q. Detecting ecological spatial-temporal changes by Remote Sensing Ecological Index with local adaptability. J. Environ. Manag. 2021, 299, 113655. [Google Scholar] [CrossRef]
- Shan, Y.F.; Dai, X.; Li, W.L.; Yang, Z.C.; Wang, Y.L.; Qu, G.; Liu, W.N.; Ren, J.S.; Li, C.; Liang, S.E.; et al. Detecting Spatial-Temporal Changes of Urban Environment Quality by Remote Sensing-Based Ecological Indices: A Case Study in Panzhihua City, Sichuan Province, China. Remote Sens 2022, 14, 4137. [Google Scholar] [CrossRef]
- Li, J.S.; Sun, W.; Li, M.Y.; Meng, L.L. Coupling coordination degree of production, living and ecological spaces and its influencing factors in the Yellow River Basin. J. Clean Prod. 2021, 298, 126803. [Google Scholar] [CrossRef]
- Zhang, K.L.; Liu, T.; Feng, R.R.; Zhang, Z.C.; Liu, K. Coupling Coordination Relationship and Driving Mechanism between Urbanization and Ecosystem Service Value in Large Regions: A Case Study of Urban Agglomeration in Yellow River Basin, China. Int. J. Environ. Res. Public Health 2021, 18, 7836. [Google Scholar] [CrossRef] [PubMed]
- The Yellow River Water Resources Commission of the Ministry of Water Resources of the People’s Rep. Huanghe Nian Jian. Yellow River Almanac Society 2020, 67–76.
- General Office of the State Council of the People’s Republic of China. Framework of plan for ecological protection and high- quality development of the Yellow River Basin. Gaz. State Counc. People’s Repub. China 2021, 30, 15–35. [Google Scholar]
- Mills, S.; Weiss, S.; Liang, C. VIIRS day/night band (DNB) stray light characterization and correction. Earth Obs. Syst. XVIII 2013, 8866, 549–566. [Google Scholar] [CrossRef]
- Liu, X.P.; Huang, Y.H.; Xu, X.C.; Li, X.C.; Li, X.; Ciais, P.; Lin, P.R.; Gong, K.; Ziegler, A.D.; Chen, A.N.; et al. High-spatiotemporal-resolution mapping of global urban change from 1985 to 2015. Nat. Sustain. 2020, 3, 564–570. [Google Scholar] [CrossRef]
- Li, X.C.; Gong, P.; Zhou, Y.Y.; Wang, J.; Bai, Y.Q.; Chen, B.; Hu, T.Y.; Xiao, Y.X.; Xu, B.; Yang, J.; et al. Mapping global urban boundaries from the global artificial impervious area (GAIA) data. Environ. Res. Lett. 2021, 15, 094044. [Google Scholar] [CrossRef]
- Zou, Z.H.; Xiao, X.M.; Dong, J.W.; Qin, Y.W.; Doughty, R.B.; Menarguez, M.A.; Zhang, G.L.; Wang, J. Divergent trends of open-surface water body area in the contiguous United States from 1984 to 2016. Proc. Natl. Acad. Sci. USA 2018, 115, 3810–3815. [Google Scholar] [CrossRef] [Green Version]
- Guo, X.J.; Zhang, C.C.; Luo, W.R.; Yang, J.; Yang, M. Urban Impervious Surface Extraction Based on Multi-features and Random Forest. IEEE Access 2020, 8, 226609–226623. [Google Scholar] [CrossRef]
- Dong, X.G.; Meng, Z.G.; Wang, Y.Z.; Zhang, Y.Z.; Sun, H.T.; Wang, Q.S. Monitoring Spatiotemporal Changes of Impervious Surfaces in Beijing City Using Random Forest Algorithm and Textural Features. Remote Sens 2021, 13, 153. [Google Scholar] [CrossRef]
- Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef] [Green Version]
- Speiser, J.L.; Miller, M.E.; Tooze, J.; Ip, E. A comparison of random forest variable selection methods for classification prediction modeling. Expert Syst. Appl. 2019, 134, 93–101. [Google Scholar] [CrossRef] [PubMed]
- Li, X.C.; Gong, P.; Lu, L. A 30-year (1984–2013) record of annual urban dynamics of Beijing City derived from Landsat data. Remote Sens. Environ. 2015, 116, 78–90. [Google Scholar] [CrossRef]
- Sun, Y.; Zhang, X.C.; Zhao, Y.; Xin, Q.C. Monitoring annual urbanization activities in Guangzhou using Landsat images (1987–2015). Int. J. Remote Sens. 2017, 38, 1258–11276. [Google Scholar] [CrossRef]
- Xu, H.Q.; Wang, Y.F.; Guan, H.D.; Shi, T.T.; Hu, X.S. Detecting Ecological Changes with a Remote Sensing Based Ecological Index (RSEI) Produced Time Series and Change Vector Analysis. Remote Sens. 2019, 11, 2345. [Google Scholar] [CrossRef] [Green Version]
- Zhou, W.Q.; Yu, W.J.; Qian, Y.G.; Han, L.J.; Pickett, S.T.A.; Wang, J.; Li, W.F.; Ouyang, Z.Y. Beyond city expansion: Multi-scale environmental impacts of urban megaregion formation in China. Nat. Sci. Rev. 2022, 9, nwab107. [Google Scholar] [CrossRef]
- Sun, Z.C.; Xu, R.; Du, W.J.; Wang, L.; Lu, D.S. High-Resolution Urban Land Mapping in China from Sentinel 1A/2 Imagery Based on Google Earth Engine. Remote Sens. 2019, 11, 752. [Google Scholar] [CrossRef] [Green Version]
- Yang, L.M.; Xian, G.; Klaver, J.M.; Deal, B. Urban Land-Cover Change Detection through Sub-Pixel Imperviousness Mapping Using Remotely Sensed Data. Photogramm. Eng. Remote Sens. 2003, 69, 1003–1010. [Google Scholar] [CrossRef]
- Wu, W.B.; Ma, J.; Meadows, M.E.; Banzhaf, E.; Huang, T.Y.; Liu, Y.F.; Zhao, B. Spatio-temporal changes in urban green space in 107 Chinese cities (1990–2019): The role of economic drivers and policy. Int. J. Appl. Earth Obs. Geoinf. 2021, 103, 102525. [Google Scholar] [CrossRef]
- Lin, Z.H.; Mo, X.G.; Li, H.X.; Li, H.B. Comparison of Three Spatial Interpolation Methods for Climate Variables in China. Acta Geogr. Sin. 2002, 57, 47–56. [Google Scholar] [CrossRef]
- Chai, T.; Draxler, R.R. Root mean square error (RMSE) or mean absolute error (MAE)?-Arguments against avoiding RMSE in the literature. Geosci. Model Dev. 2014, 7, 1247–1250. [Google Scholar] [CrossRef] [Green Version]
- Yin, D.; Qian, J.X.; Zhu, H. Living in the “Ghost City”: Media Discourses and the Negotiation of Home in Ordos, Inner Mongolia, China. Sustainability 2017, 9, 2029. [Google Scholar] [CrossRef]
- Li, G.D.; Sun, S.; Fang, C.L. The varying driving forces of urban expansion in China: Insights from a spatial-temporal analysis. Landsc. Urban Plan. 2018, 174, 63–77. [Google Scholar] [CrossRef]
- Wang, G.Z. 70 Years of China: The Changes of Population Age Structure and the Trend of Population Aging, Chinese. J. Popul. Sci. 2019, 3, 2–15. [Google Scholar]
- Yang, Z.K.; Tian, J.; Li, W.Y.; Su, W.R.; Guo, R.Y.; Liu, W.J. Spatio-temporal pattern and evolution trend of ecological environment quality in the Yellow River Basin. Acta Ecol. Sin. 2021, 41, 514. [Google Scholar] [CrossRef]
- Alberti, M. The effects of urban patterns on ecosystem function. Int. Reg. Sci. Rev. 2005, 28, 168–192. [Google Scholar] [CrossRef]
- Yao, R.; Cao, J.; Wang, L.C.; Zhang, W.W.; Wu, X.J. Urbanization effects on vegetation cover in major African cities during 2001-2017. Int. J. Appl. Earth Obs. Geoinf. 2019, 75, 44–53. [Google Scholar] [CrossRef]
- Zhang, Y.; She, J.Y.; Long, X.R.; Zhang, M. Spatio-temporal evolution and driving factors of eco-environmental quality based on RSEI in Chang-Zhu-Tan metropolitan circle, central China. Ecol. Indic. 2022, 114, 109436. [Google Scholar] [CrossRef]
- Ariken, M.; Zhang, F.; Liu, K.; Fang, C.L.; Kung, H.T. Coupling coordination analysis of urbanization and eco-environment in Yanqi Basin based on multi-source remote sensing data. Ecol. Indic. 2020, 114, 106331. [Google Scholar] [CrossRef]
Points | Polygons | Total | ||
---|---|---|---|---|
Samples of pervious surface | 7500 | 1000 | 8500 | |
Samples of impervious surface | 1986–2000 | 1350 | 150 | 1500 |
2001–2005 | 1350 | 350 | 1700 | |
2006–2010 | 1350 | 550 | 1900 | |
2011–2020 | 1350 | 750 | 2100 |
Year | PC1 | Contribution | |||
---|---|---|---|---|---|
NDVI | WET | LST | NDBSI | ||
1986 | 0.33280 | 0.69872 | −0.60621 | −0.18315 | 65.2721 |
1990 | 0.31870 | 0.72534 | −0.57887 | −0.19293 | 73.4465 |
2000 | 0.42276 | 0.58320 | −0.65819 | −0.21894 | 73.2763 |
2010 | 0.40444 | 0.71649 | −0.51518 | −0.24013 | 76.8011 |
2020 | 0.47879 | 0.67823 | −0.48274 | −0.27880 | 73.0507 |
Circle | 1986 | 1990 | 2000 | 2010 | 2020 | |||||
---|---|---|---|---|---|---|---|---|---|---|
ISCR | RSEImean | ISCR | RSEImean | ISCR | RSEImean | ISCR | RSEImean | ISCR | RSEImean | |
1 | 2.2281 | 0.4516 | 2.8231 | 0.4725 | 4.2871 | 0.4958 | 7.3249 | 0.4207 | 16.0866 | 0.4119 |
2 | 2.0486 | 0.4782 * | 2.5960 | 0.5111 * | 3.9433 | 0.5363 * | 6.7495 | 0.4404 | 14.8637 | 0.4269 |
3 | 2.0428 | 0.4706 | 2.5884 | 0.5063 | 3.9317 | 0.5313 | 6.7297 | 0.4619 | 14.8246 | 0.4571 |
4 | 2.0369 | 0.4647 | 2.5810 | 0.5003 | 3.9208 | 0.5250 | 6.7107 | 0.4655 | 14.7866 | 0.4645 |
5 | 2.0312 | 0.4606 | 2.5736 | 0.4957 | 3.9095 | 0.5202 | 6.6915 | 0.4667 | 14.7487 | 0.4685 |
6 | 2.0254 | 0.4579 | 2.5663 | 0.4924 | 3.8988 | 0.5167 | 6.6731 | 0.4673 * | 14.7137 | 0.4712 |
7 | 2.0195 | 0.4556 | 2.5589 | 0.4895 | 3.8876 | 0.5137 | 6.6538 | 0.4672 | 14.6750 | 0.4729 |
8 | 2.0140 | 0.4538 | 2.5520 | 0.4871 | 3.8769 | 0.5112 | 6.6352 | 0.4667 | 14.6350 | 0.4742 |
9 | 2.0083 | 0.4527 | 2.5447 | 0.4853 | 3.8659 | 0.5092 | 6.6164 | 0.4664 | 14.5949 | 0.4754 |
10 | 2.0026 | 0.4517 | 2.5374 | 0.4836 | 3.8547 | 0.5075 | 6.5972 | 0.4659 | 14.5531 | 0.4763 * |
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. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zhang, J.; Du, J.; Fang, S.; Sheng, Z.; Zhang, Y.; Sun, B.; Mao, J.; Li, L. Dynamic Changes, Spatiotemporal Differences, and Ecological Effects of Impervious Surfaces in the Yellow River Basin, 1986–2020. Remote Sens. 2023, 15, 268. https://doi.org/10.3390/rs15010268
Zhang J, Du J, Fang S, Sheng Z, Zhang Y, Sun B, Mao J, Li L. Dynamic Changes, Spatiotemporal Differences, and Ecological Effects of Impervious Surfaces in the Yellow River Basin, 1986–2020. Remote Sensing. 2023; 15(1):268. https://doi.org/10.3390/rs15010268
Chicago/Turabian StyleZhang, Jing, Jiaqiang Du, Shifeng Fang, Zhilu Sheng, Yangchengsi Zhang, Bingqing Sun, Jialin Mao, and Lijuan Li. 2023. "Dynamic Changes, Spatiotemporal Differences, and Ecological Effects of Impervious Surfaces in the Yellow River Basin, 1986–2020" Remote Sensing 15, no. 1: 268. https://doi.org/10.3390/rs15010268
APA StyleZhang, J., Du, J., Fang, S., Sheng, Z., Zhang, Y., Sun, B., Mao, J., & Li, L. (2023). Dynamic Changes, Spatiotemporal Differences, and Ecological Effects of Impervious Surfaces in the Yellow River Basin, 1986–2020. Remote Sensing, 15(1), 268. https://doi.org/10.3390/rs15010268