Article
Environmental Sciences
Weibo Zeng, Qiuyan Deng, Xingyue Zhao, Dehua Li, Xinran Min
Summary: This paper proposes a remote sensing image stitching method that considers the impact of topography and geomorphology. By optimizing feature matching, reducing matching redundancy using Delaunay triangle mesh, and applying weighted fusion algorithm, the proposed method achieves good results in terms of accuracy, efficiency, and visual effect.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Katsunoshin Nishi, Masaaki Kawai, Bowo Eko Cahyono, Mirza Muhammad Waqar, Kaori Nishi, Josaphat Tetuko Sri Sumantyo
Summary: Concerns about land subsidence recurrence have risen in Japan due to urban development and groundwater use. While land subsidence observation wells have been used for monitoring, the aging facilities have sparked interest in remote sensing technology. This paper evaluates the substitutability of Consecutive DInSAR with the land subsidence observation well, using SARPROZ software to analyze Sentinel 1 images from August 2017 to March 2022 in Kanagawa prefecture. The study also introduces a new calculation model to estimate land subsidence based on the law of material conservation. The results show that Consecutive DInSAR aligns with the observation well data, indicating potential future substitution.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Kaili Zhang, Yonggang Chen, Wentao Wang, Yudi Wu, Bo Wang, Yanting Yan
Summary: In the study, a spectral-spatial feature called Pixel Neighbourhood Similarity (PNS) index was constructed, and it showed distinct boundaries between different land types. The PNS index exhibited relatively higher performance compared to other spectral-spatial features. Additionally, a feature combination called PNS-CFS, which included PNS index and 19 other features, significantly improved the classification accuracy compared to other feature combinations.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Sinan Bulut, Alkan Gunlu, Gunay Cakir
Summary: This study aims to estimate stand volume, basal area, number of trees, mean diameter, and top height using Landsat 8 and Sentinel-2 satellite images. Results showed that the support vector machine technique provided the best performance for estimating number of trees and basal area, and texture values were found to be more suitable than reflectance and vegetation indices for estimating stand parameters.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Tao Hai, Biju Theruvil Sayed, Ali Majdi, Jincheng Zhou, Rafid Sagban, Shahab S. Band, Amir Mosavi
Summary: A hybrid machine learning method is proposed for wildfire susceptibility mapping, utilizing a GIS database with 11 influencing factors and 262 fire locations. A multivariate adaptive regression splines (MARS) model is developed using the database and tuned with a cat swarm optimization (CSO) algorithm to generate accurate susceptibility maps. Results show strong correlations between land use, temperature, slope angle, and fire severity. The MARS-CSO model outperforms other models in terms of prediction capability, and the resulting wildfire risk map categorizes 20% of the study areas as very low risk and 40% as very high risk.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Shahfahad, Swapan Talukdar, Mohd Waseem Naikoo, Atiqur S. Rahman, Alexandre S. Gagnon, Abu Reza Md Towfiqul Islam, Amirhosein Mosavi
Summary: This study compares the accuracy of Landsat satellites' OLI and OLI-2 sensors in land use land cover (LULC) mapping. Image fusion techniques were applied to improve the spatial resolution of OLI and OLI-2 multispectral images, followed by LULC mapping using a support vector machine (SVM) classifier. The results demonstrate that OLI-2 provides more accurate LULC classification than OLI. Validation of the classified LULC maps reveals better performance of OLI-2 in distinguishing dense and sparse vegetation, as well as darker and lighter objects. The relationship between LULC maps and surface biophysical parameters using Local Moran's I also demonstrates the superiority of OLI-2 in LULC mapping compared to OLI.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Nam-Thang Ha, Tien-Dat Pham, Huu-Ty Pham, Dang-An Tran, Ian Hawes
Summary: In this study, a fusion of SAR Sentinel-1 (S-1), multi-spectral Sentinel-2 (S-2), and advanced machine learning models was used to improve the estimation of TOC stock in Zostera muelleri meadows in New Zealand. The best prediction of seagrass TOC was achieved by fusing S1 and S2 images, using the CatBoost ML model and the grey wolf optimization algorithm. The results provide new ideas for low-cost, scalable, and reliable estimates of seagrass TOC globally.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Arnob Bhattacharjee, S. M. Quamrul Hassan, Papri Hazra, Tapos Kormoker, Shahana Islam, Edris Alam, Md Kamrul Islam, Abu Reza Md. Towfiqul Islam
Summary: This research investigates the future changes in summer monsoon rainfall and temperature in Bangladesh. The study identifies the best models for projecting temperature and rainfall. The results show that the projected rainfall in most parts of Bangladesh will increase, while the northwest and west-central areas may experience the most significant rise in temperature.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Bohao Cui, Yili Zhang, Linshan Liu, Changjun Gu, Zhaofeng Wang, Bo Wei, Dianqing Gong
Summary: This study investigated the spatial and temporal evolution characteristics of forest disturbance and recovery in the central Himalayas from 1995 to 2018 using random forests model, LandTrendr temporal segmentation algorithm, and geographical detector. The results showed that forest disturbance was dominant and greater than recovery. Both disturbance and recovery exhibited a significant decreasing trend, with disturbance mainly occurring at low elevations and the Gandaki basin, while recovery mainly occurred in eastern Nepal. The main influencing factors on forest dynamics were elevation, temperature, and population, with their interaction synergistically enhanced.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Alysha van Duynhoven, Suzana Dragicevic
Summary: This study proposes a geospatial sample weighting approach using class-level landscape metrics to assign importance to different categories, aiming to address the imbalance of land cover data for deep learning models.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Luan Thanh Pham, Saulo Pomponet Oliveira, Hao Van Duong, Kamal Abdelrahman, Mohammed S. Fnais, David Gomez-Ortiz, Dat Viet Nguyen, Quynh Thanh Vo, Thong Kieu Duy, Ahmed M. Eldosouky
Summary: In this study, the subsurface structures of the northern Kontum massif were determined using aeromagnetic data for the first time. The results showed that the area is dominated by E-W and ENE-WSW trending lineaments, with depths ranging from 50 to 550 m. The presence of probable granitic intrusions was also observed. This magnetic interpretation provides valuable insights into the structural features of the northern Kontum massif.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Pride Mafuratidze, Tendai Polite Chibarabada, Munyaradzi Davis Shekede, Mhosisi Masocha
Summary: Hailstorms have become more frequent and intense over the past decade, leading to significant agricultural losses. This study aimed to develop and test a new normalized vegetation index approach to assess the severity of hailstorm damage on sugarcane plants in a large estate in southeastern Zimbabwe. The results showed that the & UDelta;NDTI index computed using multi-spectral datasets within two weeks after a hailstorm can effectively detect and characterize the damage on sugarcane, providing reliable information for customized crop insurance packages.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Wendiwesen Megersa, Kiros Tsegay Deribew, Girmay Abreha, Tebarek Liqa, Mitiku Badasa Moisa, Samuel Hailu, Kenate Worku
Summary: This study analyzed the negative impacts of urban expansion on the environment in Mettu area, Ethiopia, using stochastic modeling with the CA-Markov chain and multi-scenario prospects. Landsat images from 1986, 2000, and 2021 were used. The results show a significant increase in built-up areas, mainly at the expense of cropland and forest. The model predicts a twofold expansion of urban areas before the 2040s, posing challenges for ecological and economic livelihoods, unless green economy principles are implemented effectively.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Joshua Sos, Kim Penglase, Tom Lewis, Prashant K. Srivastava, Harikesh Singh, Sanjeev K. Srivastava
Summary: The use of drones and remote sensing combined with geospatial analysis is a cost-efficient method for monitoring energy distribution networks in fire-prone areas. This study evaluated vegetation height and volume in power line corridors in Southeast Queensland, Australia using image and photogrammetric analysis with segmentation algorithms. Drone-generated models were used to assess the effectiveness of various fuel reduction techniques. Strong correlation was observed between field observations and drone-derived models, indicating the efficacy of this approach in assessing fuel heights.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Zhiye Wang, Chuanming Ma, Yan Zhang, Bo Hu, Shuming Xu, Zhenfen Dai
Summary: Expanding urbanization has increased the risk of urban flooding, posing a hazard to humans. The Analytic Hierarchy Process-Press-State-Response (AHP-PSR) model was used to assess urban flooding vulnerability in Jining City, China. The results provide important references for government in preventing and mitigating urban flooding.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Cheng Wang, Jingyuan Chen, Dan Li, Yunbin Zhang, Meng Zhu, Fang Rong, Zhiqiang Gan
Summary: This study examines the level of urban polycentricity in 41 cities in China's Yangtze River Delta region and investigates the driving factors behind its spatial differentiation. The results show that the distribution of hot and cold spots of the urban polycentricity index varies between different periods, and population size and topographic relief play key roles in driving urban polycentricity.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Aimin Li, Xiangyu Yan, Xuan Kang
Summary: Due to the complexity of spectral characteristics for inland lake water, high-precision atmospheric correction methods are essential in remote sensing. This research conducted atmospheric correction on the Tiande lake zone in Zhengzhou, China using four methods (QUAC, Dark Subtract, FLAASH, and 6S) and compared the water reflectance with ground-based spectrometer data. The analysis shows that FLAASH method is best suited for Zhuhai-1 and Sentinel-2 images, while 6S method is best suited for Planet and GF-2 images. The correction results of the four image types demonstrate the applicability of FLAASH and 6S in lake water color remote sensing inversion.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Junkui Xu, Hao Zhang, Chun Liu, Jianzhong Guo
Summary: This study proposes a method for building shape recognition using a deep graph filter neural network. By treating shape recognition as a combination of subjective and objective graph signal filtering process, and constructing a shape features extraction framework, the tasks of shape classification and embedding of buildings are achieved.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Muhammad Ahmad Raza, Mohammed M. A. Almazah, Ijaz Hussain, Fuad S. S. Al-Duais, A. Y. Al-Rezami, Talha Omer
Summary: This study examines the spatial patterns of seasonal drought frequency and inter-seasonal drought persistence in the northeastern region of Pakistan. The results indicate that some areas in the study region are more prone to drought and vulnerable to inter-seasonal drought persistence. The Bayesian logistic regression model provides more accurate and precise regional seasonal drought forecasts.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Yenny Cuellar, Liliana Perez
Summary: This study examines the effect of different neighborhood sizes and spatial resolutions on the performance of the Future Land Use Simulation (FLUS) model using Cellular Automata. The FLUS model employs an Artificial Neuronal Network to calculate the relationship between land uses and drivers and estimate the probability of each land use. Sensitivity analysis is conducted by varying the neighborhood sizes and spatial resolutions, and various metrics are used to assess the model's performance. The results show that a 3 x 3 neighborhood size and 5 meters spatial resolution yield the best accuracy.
GEOCARTO INTERNATIONAL
(2023)
