About

Dr. Conghe Song is a Professor of Geography and Environment at the University of North Carolina at Chapel Hill and serves as the Department Chair. He is a member of the Remote Sensing and Ecological Modeling Group. His research interests include remote sensing, ecological modeling, and geographic analysis. Dr. Song has a comprehensive educational background with a Ph.D. in Geography from Boston University, an M.S. in Forest Ecology from Beijing Forestry University, and a B.S. in Forestry from Anhui Agricultural University. He also holds a Certificate in Education from Capital Normal University and a Certificate of Training from the Center for Creative Leadership. His professional experience includes serving as a Lecturer at Beijing Forestry University, an Assistant Professor, Associate Professor, and now Professor at UNC Chapel Hill. He has also been a Charles Bullard Fellow at Harvard University. Since 2023, Dr. Song has been the Chair of the Department of Geography and Environment at UNC Chapel Hill. Throughout his career, he has contributed to the fields of geography and ecological modeling through research, teaching, and leadership roles.

Research topics

• Geography
• Ecology
• Economics
• Computer Science
• Economic growth
• Political Science
• Natural resource economics
• Environmental science
• Business
• Geology
• Environmental resource management
• Psychology
• Atmospheric sciences
• Development economics
• Mathematics
• Agroforestry
• Demography
• Statistics
• Internal medicine
• Medicine
• Environmental health
• Climatology
• Biology
• Remote sensing

Selected publications

• The interactive effects of impervious surface, vegetation, and water on urban land surface temperature

  Sustainable Cities and Society · 2026-04-21

  article
  PublisherDOI

• Urbanization and climate change jointly shift land surface phenology in the northern mid-latitude large cities

  UNC Libraries · 2025-07-11

  articleOpen access
  PublisherDOI

• Syngas Production through Chemical-Looping-Enhanced Hydrogenation of Carbonates

  ACS Sustainable Chemistry & Engineering · 2025-12-09 · 1 citations

  article

  Direct hydrogenation of carbonates (DHC) technology represents a breakthrough alternative to conventional syngas production processes that rely on fossil fuels. Nevertheless, it is constrained by a low carbonate conversion efficiency and suboptimal hydrogen utilization. Here, a novel strategy of chemical-looping-enhanced hydrogenation of carbonates (HC-CL) is proposed for the coproduction of metal oxides and syngas, which divides the DHC process into two half-looping processes involving H2-reduction and decarbonation-oxidation. An excellent redox catalyst of spinel Ce5/MgFe2O4 is developed to drive the HC-CL, which can transform into the rock salt structure of Ce5/MgxFeyO–Fe0 to provide abundant Fe metal sites and oxygen vacancies (Ov) after H2-reduction, and regenerate itself through efficiently splitting CO2 from carbonates into CO during the decarbonation-oxidation. Reduction-regeneration mechanism of Fe species ensures the redox reversibility of Ce5/MgFe2O4, achieving a high CO2-splitting efficiency of 70% over 10 HC-CL cycles. The techno-economic analysis based on the process simulation further illustrates that the HC-CL process can save 2 GJ/tcarbonate of energy consumption, 0.5 tCO2/tcarbonate of CO2 emissions, and 55 $/tcarbonate of costs than the DHC process coupled with H2 separation and recycling utilization. Therefore, the HC-CL process turns out to be a promising novel technology for syngas production.

  PublisherDOI

• What motivates local leaders of community forests in Nepal? An examination of leaders’ expressed values and experiences

  Environmental Science & Policy · 2025-09-27

  articleOpen access

  Nepal’s community forestry (CF) program, a globally recognized model of participatory forest management, relies on voluntary local leaders to guide forest management and governance decisions. Sustaining voluntary leadership has become increasingly challenging because of outmigration, declining forest dependence, and growing urban influence on rural livelihoods. In this study, we explore the values and motivations of existing leaders of community forest user groups (CFUGs), which underpin the leadership characteristics in sustaining these local institutions. We surveyed 144 leaders of 49 CFUGs in Nepal’s mid-hills and used their responses as indicators of leadership values, derived from the “Motivation to Lead” and related theoretical frameworks. Using exploratory factor analysis and a multiple indicators multiple causes (MIMIC) model, we identify three motivation factors and examine their associations with leaders’ individual and CF characteristics. The results suggest that, out of the 16 indicators, eight explain core values and motives clustered into three latent motivation factors, indicating three axes of leadership motivation in Nepal’s CF program: environmental stewardship, altruism, and power and influence. Leaders were likely to be motivated by either environmental stewardship or altruism. However, leaders motivated by either altruism or environmental stewardship were also motivated by the power and influence. Furthermore, individual leadership characteristics such as leadership position and duration, and CF characteristics such as forest type, support from non-governmental organizations, fire incidences, and leadership experience in local governments, are associated with leadership motivation factors. These findings inform understanding of voluntary leadership drivers in CFUGs, for strengthening and sustaining community-based forest management in Nepal. • We developed a survey to capture CF leaders’ values through response statements. • Factor analysis and a MIMIC model identified three core leadership motivations. • Eight indicators revealed three motivation factors: Stewardship, altruism, power and influence. • Leaders valued either stewardship or altruism, often tied with power and influence. • Leadership role, experience, forest type, NGO aid, and fire events associated with motivations.

  PublisherDOI

• Understanding the continuous phenological development at daily time step with a Bayesian hierarchical space-time model: impacts of climate change and extreme weather events

  UNC Libraries · 2025-07-09

  articleOpen access
  PublisherDOI

• Interpretable Multi-Sensor Fusion of Optical and SAR Data for GEDI-Based Canopy Height Mapping in Southeastern North Carolina

  Remote Sensing · 2025-04-25 · 4 citations

  articleOpen access

  Accurately monitoring forest canopy height is crucial for sustainable forest management, particularly in southeastern North Carolina, USA, where dense forests and limited accessibility pose substantial challenges. This study presents an explainable machine learning framework that integrates sparse GEDI LiDAR samples with multi-sensor remote sensing data to improve both the accuracy and interpretability of forest canopy height estimation. This framework incorporates multitemporal optical observations from Sentinel-2; C-band backscatter and InSAR coherence from Sentinel-1; quad-polarization L-Band backscatter and polarimetric decompositions from the Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR); texture features from the National Agriculture Imagery Program (NAIP) aerial photography; and topographic data derived from an airborne LiDAR-based digital elevation model. We evaluated four machine learning algorithms, K-nearest neighbors (KNN), random forest (RF), support vector machine (SVM), and eXtreme gradient boosting (XGB), and found consistent accuracy across all models. Our evaluation highlights our method’s robustness, evidenced by closely matched R2 and RMSE values across models: KNN (R2 of 0.496, RMSE of 5.13 m), RF (R2 of 0.510, RMSE of 5.06 m), SVM (R2 of 0.544, RMSE of 4.88 m), and XGB (R2 of 0.548, RMSE of 4.85 m). The integration of comprehensive feature sets, as opposed to subsets, yielded better results, underscoring the value of using multisource remotely sensed data. Crucially, SHapley Additive exPlanations (SHAP) revealed the multi-seasonal red-edge spectral bands of Sentinel-2 as dominant predictors across models, while volume scattering from UAVSAR emerged as a key driver in tree-based algorithms. This study underscores the complementary nature of multi-sensor data and highlights the interpretability of our models. By offering spatially continuous, high-quality canopy height estimates, this cost-effective, data-driven approach advances large-scale forest management and environmental monitoring, paving the way for improved decision-making and conservation strategies.

  PublisherOA PDFDOI

• Impacts of COVID-19 pandemic on the livelihoods of rural households in the community forestry landscape in the Middle Hills of Nepal

  UNC Libraries · 2025-05-03

  articleOpen accessSenior author
  PublisherDOI

• Tracking Major Sources of Water Contamination Using Machine Learning

  UNC Libraries · 2025-06-24

  articleOpen accessSenior author

  Current microbial source tracking techniques that rely on grab samples analyzed by individual endpoint assays are inadequate to explain microbial sources across space and time. Modeling and predicting host sources of microbial contamination could add a useful tool for watershed management. In this study, we tested and evaluated machine learning models to predict the major sources of microbial contamination in a watershed. We examined the relationship between microbial sources, land cover, weather, and hydrologic variables in a watershed in Northern California, United States. Six models, including K-nearest neighbors (KNN), Naïve Bayes, Support vector machine (SVM), simple neural network (NN), Random Forest, and XGBoost, were built to predict major microbial sources using land cover, weather and hydrologic variables. The results showed that these models successfully predicted microbial sources classified into two categories (human and non-human), with the average accuracy ranging from 69% (Naïve Bayes) to 88% (XGBoost). The area under curve (AUC) of the receiver operating characteristic (ROC) illustrated XGBoost had the best performance (average AUC = 0.88), followed by Random Forest (average AUC = 0.84), and KNN (average AUC = 0.74). The importance index obtained from Random Forest indicated that precipitation and temperature were the two most important factors to predict the dominant microbial source. These results suggest that machine learning models, particularly XGBoost, can predict the dominant sources of microbial contamination based on the relationship of microbial contaminants with daily weather and land cover, providing a powerful tool to understand microbial sources in water.

  PublisherDOI

• Urban Planning Scenarios Significantly Shape Future Urban Growth and Landscape Patterns in China’s Megacity, Shanghai

  Earth Systems and Environment · 2025-08-18 · 2 citations

  article
  PublisherDOI

• Climate Variability Masked Greening Effects on Water Yield in the Yangtze River Basin During 2001–2018

  UNC Libraries · 2025-04-25

  articleOpen access

  Rapid global vegetation greening has been observed for the past two decades, but its implications to the hydrological cycle are not well understood in many regions, including the Yangtze River Basin (YRB). This study used a remote sensing‐driven ecosystem model, the Coupled Carbon and Water model, to fully examine the individual and combined hydrological effects of vegetation and climate changes through a series of modeling experiments. During the study period (2001–2018), the vegetation showed a significant greening trend with the mean annual normalized difference vegetation index increasing at a rate of 0.4% per year ( p  < 0.001). In contrast, climate exhibited a marginal wetting trend with annual precipitation increasing at a rate of 6.7 mm/yr ( p  = 0.08). Annual evapotranspiration (ET) in the YRB significantly increased (3.1 mm/yr, p  = 0.01) primarily due to enhanced ecosystem productivity associated with vegetation greening, rather than climatic factors. However, the enhancement in ET did not lead to a significant decline in total water yield at the YRB scale. The large inter‐annual variability of precipitation masked the effects of vegetation greening on water yield. Overall, our study indicated that the recent land “greening up” has accelerated the regional hydrological cycle through increasing ET and resulted in enhanced risks of water resource shortage. Our findings highlighted the close connection between land cover dynamics and hydrological cycle under climate variability in one of the world’s largest river systems. Effective basin water resource management must consider hydrological response to vegetation greening and climate change. Key Points The Yangtze River basin experienced significant vegetation greening during 2001–2018 The recent vegetation greening led to a significant increase in evapotranspiration (ET) Precipitation variability masked the effects of ET increase on water yield

  PublisherDOI

Recent grants

• Scaling Up Forest Ecosystem Carbon Budget from Stand to Landscape: Impacts of Forest Structures

  NSF · $160k · 2004–2008

• CNH: The Effects of China's Grain-for-Green Program on the Dynamics of Coupled Natural-Human System in Rural China

  NSF · $1.2M · 2013–2018

• DISES: Influence of Community Forestry on the Dynamics of the Integrated Socio-Environmental Systems

  NSF · $1.6M · 2021–2026

Frequent coauthors

• Yulong Zhang

  China North Industries Group Corporation (China)

  34 shared

• Fangfang Yao

  Ningbo No. 2 Hospital

  30 shared

• Shuai Zhang

  29 shared

• Quanfa Zhang

  Wuhan Botanical Garden

  28 shared

• Xiao Yang

  Southern Methodist University

  27 shared

• Lawrence E. Band

  University of Virginia

  25 shared

• Junxiang Li

  Xi'an Jiaotong University

  25 shared

• Qi Zhang

  24 shared

Education

• PhD, Geography

  Boston University

  2001

• MS, Forestry

  Beijing Forestry University

  1991

• BS, Forestry

  Anhui Agricultural University

  1988

Awards & honors

• Charles Bullard Fellow, Harvard University (2005-2006)