Guoming DU

Prof. Dr. Guoming DU

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 March 2023 in Remote Sensing
Verified
Journal Article
Open access
PDF file
Kebiao MaoKebiao Mao
Kebiao Mao
HWHan Wang
Han Wang
JSJiancheng Shi
Jiancheng Shi
Essam HeggyEssam Heggy
Essam Heggy
Shengli WuShengli Wu
Shengli Wu
Guoming DUGuoming DU
Guoming DU
Soil moisture (SM) and land surface temperature (LST) are entangled, and the retrieval of one of them requires a priori specification of the other one. Due to insufficient observational information, retrieval of LST and SM from passive microwave remote sensing data is often ill-posed, and the retrieval accuracy needs to be improved. In this study, a novel fully-coupled paradigm is developed to robustly retrieve SM and LST from passive microwave data, which integrates deep learning, physical methods, and statistical methods. The key condition of the general paradigm proposed by us is that the output parameters of deep learning can be uniquely determined by the input parameters theoretically through a certain mathematical equation. Firstly, the physical method is deduced based on the energy radiation balance equation. The nine unknowns require the brightness temperatures of nine channels to construct nine equations, and the solutions of the physical method equations are obtained by model simulation. Based on the derivation of the physical method, the solution of the statistical method is constructed using multi-source data. Secondly, the solutions of physical and statistical methods constitute the training and test data of deep learning, which is used to obtain the solution curve of physical and statistical methods. The retrieval accuracy of LST and SM is greatly improved by smartly utilizing the mutual prior knowledge of SM and LST and cross iterative optimization calculations. Finally, validation indicates that the mean absolute error of the retrieved SM and LST data are 0.027 m3/m3 and 1.38 K, respectively, at an incidence angle of 0–65°. A model-data-knowledge-driven and deep learning method can overcome the shortcomings of traditional methods and provide a paradigm for retrieval of other geophysical variables. The proposed paradigm not only has physical meaning, but also makes deep learning physically interpretable, which is a milestone in the retrieval of geophysical remote sensing parameters based on artificial intelligence technology.

5 June 2022 in Land
Verified
Journal Article
Open access
PDF file
Bonoua FayeBonoua Faye
Bonoua Faye
Guoming DUGuoming DU
Guoming DU
RZRu Zhang
Ru Zhang
The main challenge for land use efficiency is population growth. Using a coupling coordination model and the Sustainable Development Goal 11.3.1 (Land Consumption Rate to Population Growth Rate, LCRPGR), the present study unravels the evolution of land use types and efficiency of land use in the Economic Community of West African States (ECOWAS) from 2003 to 2019. We interpreted the temporal land use types based on the long-time series Moderate Resolution Imaging Spectroradiometer (MCD12Q1-V6) land use images in 2003, 2007, 2011, 2015, and 2019. The World Bank data on population and socioeconomics were used to support the investigation on the coupled link. Our results showed that from 2003 to 2019, the total area of agricultural land decreased by −1.4%, while the built-up area expanded by 25.80%. Globally, the LCRPGR was 0.82, of which Cabo-Verdi and Niger had the highest (1.69) and lowest index (0.58), respectively. Correspondingly, the coupled coordination model indicated an overall value of 6.81, in which the highest and lowest value occurred in Liberia (29.78) and Niger (0.31), respectively. Furthermore, a rapidly growing population had become one of the most important factors causing inefficient land use in the study area. In summary, the ECOWAS was confronted with a lack of management to improve land use efficiency. Thus, future land use policies should pay more attention to balancing the coupled relationship between urban expansion and socio-economic interests.

22 September 2021 in Land
Verified
Journal Article
Open access
PDF file
Bonoua FayeBonoua Faye
Bonoua Faye
Guoming DUGuoming DU
Guoming DU
The study aims to reveal the transition features of agricultural land use in the Groundnut Basin of Senegal from 2009 to 2018, especially the impact of urbanization on agricultural land and the viewpoint of farmland spatiotemporal evolution. Integrated data of time series MCD12Q1 land-use images of 2009, 2012, 2015, and 2018 were used to provide a land transition in agricultural and urban areas through the synergistic methodology. Socio-economic data was also used to serve as a basis for the argument. The results highlight that: (1) Agricultural land increased by 14.53%, with a dynamic index of 1.45 from 2009–2018. (2) Over the same period, urbanization increased by 2.80%, with a dynamic index of 0.28. (3) In different regions, the transition of agricultural land in Kaffrine is most intense (expansion rate: 22.80%). The same situation of urbanization happened in Thiès Region with a value of 7.94%. Except for Thiès, agricultural land in other regions has not yet been subject to major pressure due to urbanization. Overall, the farming system in Groundnut Basin is an extensive model, the recommendations from the point of view of land-use planning and land law are necessary to ensure efficient agricultural land management in the area.

26 April 2023 in Land
Verified
Journal Article
Open access
PDF file
JYJiang Yun
Jiang Yun
Guoming DUGuoming DU
Guoming DU
HTHao Teng
Hao Teng
JWJun Wang
Jun Wang
HLHaolin Li
Haolin Li
Simulating the spatial response of ecosystem service value (ESV) caused by land use change in Heilongjiang Province under different scenarios in the future is of great significance for ensuring ecological security and sustainable development in the region. Heilongjiang Province, an important grain-producing region in China, is taken as the research area in this study. Five land use maps (2000, 2005, 2010, 2015, and 2020) were used to evaluate the change of ecosystem service value in Heilongjiang Province in the past 20 years. In addition, the area of each land use type under different future scenarios was predicted by Markov model and MOP model, the future land use pattern was simulated based on PLUS model, the ESV under different scenarios was calculated, and the spatial distribution and the degree of ESV clustering were further explored. The results show that: (1) During 2000–2020, the built-up land in Heilongjiang Province continued to grow, with a total increase of 5076 km2 during the 20-year period, the area of water continued to grow, the area of unused land gradually decreased, and the area of farmland and forest land changed less. (2) During the study period, the ESV in 2000, 2005, 2010, 2015, and 2020 were 1320.8 billion yuan, 1334.5 billion yuan, 1342.1 billion yuan, 1327.6 billion yuan, and 1328.4 billion yuan, respectively. Generally, it shows a fluctuating trend. (3) The ESV of natural development scenario (NDS), economic priority scenario (ERS) and ecological protection scenario (EPS) are 1334.3 billion yuan, 1254.8 billion yuan and 1455.6 billion yuan, respectively. The ESV of different scenarios is quite different. (4) The spatial distribution of ESV was higher in the northwest, central, and southeast, and lower in the east and west. The hot and cold spots of ESV are widely distributed and the degree of polymerization is high. The methods and conclusions of this study can provide scientific reference for the optimization of national spatial pattern and the formulation of sustainable development policy.

13 March 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Bonoua FayeBonoua Faye
Bonoua Faye
Guoming DUGuoming DU
Guoming DU
Quanfeng LiQuanfeng Li
Quanfeng Li
HFHélène Véronique Marie Thérèse Faye
Hélène Véronique Marie Thérèse Faye
JDJeanne Colette Diéne
Jeanne Colette Diéne
HSHenri Marcel Seck
Henri Marcel Seck
Understanding the factors influencing the vulnerability of forested areas is crucial for human well-being and effective governance of ecosystem supply and demand. Based on remote sensing data, this study also considered ten natural and human variables as indexes to explore the main influencing factors that may impact the vulnerability of the Thies region’s forested areas. The 2005, 2010, 2015, and 2020 satellite image data were processed using ArcGIS 10.6 and ENVI 5.1 software. The methodology includes using the transfer matrix approach and calculating the geographic landscape index to describe the dominant morphology of forested areas. Furthermore, a mixed linear regression model was built to establish the connection between forested areas and the potential contributing components. Our study revealed that the forested areas led to relative fragmentation, with an average of 88 patches for Aggregation Index (AI), 3.25 for Largest Patch Index (LPI), 2.50 for Patch Density (PD), and 112 for Landscape Shape Index (LSI) between 2005 and 2020. In addition, the transfer matrix indicated that the loss of forestry areas was about −78.8 km2 for agricultural land, −127.8 km2 for bare land, and −65.3 km2 for artificial surfaces. The most critical factors that influenced forested areas were agricultural and manufactural added value, rainfall (p < 0.05), slope, distance to the road, and agricultural sown area (p < 0.001). Overall, this investigation has revealed that the effective management of forested areas in the Thies region requires an understandable assessment. It was observed that both human anthropogenic and natural factors significantly contribute to the decline in forested areas.

24 August 2024 in Agriculture
Verified
Journal Article
Open access
PDF file
Guoming DUGuoming DU
Guoming DU
LHLe Han
Le Han
LYLongcheng Yao
Longcheng Yao
Bonoua FayeBonoua Faye
Bonoua Faye
Understanding the spatiotemporal patterns and driving mechanisms of cropping patterns’ evolution tailored to local conditions is crucial for the effective allocation of black soil in northeast China and the advancement of agricultural development. This study utilized the Google Earth Engine platform to extract the spatial distribution data of major grain crops in northeast China for the year 2022. Using crop classification data from 2000 to 2022, the spatial overlay analysis method identified cropping pattern types based on spatial and temporal changes. The primary cropping patterns identified were continuous maize cropping, maize–soybean rotation, mixed cropping, and continuous soybean cropping. Simultaneously, this research constructed three distinct crop periods: Period I (2000–2002), Period II (2010–2012), and Period III (2020–2022). Over three periods, these patterns covered 94.73%, 88.76%, and 86.39% of the area, respectively. The evolution of the dominant cropping pattern from Period I to Period II involved the transition from continuous soybean cropping to continuous maize cropping, while from Period II to Period III, the main shift was from continuous maize cropping to maize–soybean mixed cropping. From a spatial perspective, since Period I, maize has increasingly replaced soybean as the dominant crop, with continuous maize cropping expanding northward and continuous soybean cropping contracting. The maize–soybean rotation area also migrated northward, particularly in the core area of the Songnen Plain, evolving mostly into continuous maize cropping. Maize cropping areas exhibited significant regional characteristics, being densely distributed in the Sanjiang Plain and Liaohe Plain, and along major tributaries in northeast China. Consequently, the interplay of the natural environment, economic policies, and agricultural technologies drove these changes. The findings offer valuable insights for optimizing cropping patterns and developing rotation systems in northeast China.