Is urban development ecologically sustainable? Ecological footprint analysis and prediction based on a modified artificial neural network model: A case study of Tianjin in China

Meiyu Wu, Yigang Wei, Patrick T.I. Lam, Fangzhu Liu, Yan Li

Research output: Journal article publicationJournal articleAcademic researchpeer-review

24 Citations (Scopus)


Cities face significant challenges in moving forward with sustainable development. Examples of such challenges are the conflict between economic growth and shortage of natural resources, the depletion of energy and the drastic reduction of environmental carrying capacity. This study evaluates the state of sustainable development and varying trends from 1994 to 2014 in the first-tier Chinese city of Tianjin. A host of sustainability indicators are investigated, including ecological footprint (EF), ecological capacity (EC), ecological deficit (ED)/surplus, optimum population size, EF of 104-yuan Gross National Product (GDP) and EF diversity (EFD). These indicators provide complete insights into the state and trend of urban sustainability. This study proposes a novel prediction model with improved precision based on artificial neural networks. Using the model, this study predicted the EF and EC for Tianjin between 2015 and 2030. Finding yielded the following observations within this period. The total EF increased significantly from 1.17 gha/cap (global hectare/capita) to 3.09 gha/cap, which is virtually a threefold increase. Pasture land, fishing grounds, built-up land and forest land accounted for a small proportion of the total EF, whereas those of fossil energy land and arable land were large. The total EC indicated a slight decrease from 0.27 gha/cap to 0.21 gha/cap. The ECs of pasture land, forest land and fishing grounds were relatively small, whereas those of arable land and built-up land were large. The total ED increased significantly from −0.2632 gha/cap to −3.0511 gha/cap, which indicates that the ecological resource endowments of Tianjin are insufficient to sustain human activities. The optimum population increased by 95.84%, which added from 7.22 × 106 to 14.14 × 106, while the actual population is consistently on overload. The EF of 104-yuan GDP and ecological footprint diversity had a downward trend, indicating the growing efficiency of resource utilisation. This paper proposes tenable suggestions for the progress of urban sustainability. Predictions of the autoregressive integrated moving average and back-propagation neural network models indicate the deterioration in the ecological balance of Tianjin will continue in the short- and mid-term unless effective measures are taken.

Original languageEnglish
Article number117795
JournalJournal of Cleaner Production
Publication statusPublished - 10 Nov 2019


  • Artificial neural network
  • China
  • Ecological footprint
  • Environmental carrying capacity
  • Urban sustainability

ASJC Scopus subject areas

  • Renewable Energy, Sustainability and the Environment
  • Environmental Science(all)
  • Strategy and Management
  • Industrial and Manufacturing Engineering

Cite this