Search
Keeping the global consumption within the planetary boundaries

November 13, 2024 | Nature | Source |

 

Introduction: Researchers from Shandong University (China) and the University of Maryland (USA) investigate the global distribution of environmental footprints across 168 countries, covering 98% of the global population. The study categorizes 201 global consumption groups by expenditure using the environmentally extended multi-regional input-output (EE-MRIO) model. This model estimates direct and indirect environmental impacts of consumption, quantifying six environmental indicators tied to planetary boundary (PB) transgressions, including CO2 emissions, land-system change, nitrogen fixation, phosphorus fertilizer use, blue-water consumption, and biosphere integrity. By addressing consumption inequality, the research aims to guide sustainable policymaking to restore ecological balance.

Key findings: The study reveals that the top 10% of global consumers are disproportionately responsible for 43% of CO2 emissions, 23% of land-system changes (HANPP), 26% of nitrogen fixation, and 25% of phosphorus use, alongside significant contributions to blue-water consumption and biosphere degradation. In contrast, the bottom 50% largely remain within planetary boundaries. Key drivers include food production, manufacturing, and services. Wealthier groups rely more on animal-based foods, while poorer groups depend on plant-based diets. Mitigation strategies aimed at reducing overconsumption among the wealthiest 20%, particularly in food and service sectors, could decrease environmental pressure by 25–53%, helping restore ecological balance and biodiversity.  

Targeted policies, such as progressive taxation, efficiency improvements, and equitable resource distribution, are essential to curb overconsumption and promote sustainability. However, data gaps and socioeconomic disparities pose challenges for universal implementation. Future research should broaden planetary boundary indicators, improve data tracking, and foster collaboration between governments, industries, and communities to enhance global sustainable practices.

 

Figure | The footprints of the six environmental indicators and the shares of each global expenditure decile in the total footprints in 2017. Bar and doughnut pie chart refers to the per capita footprints and the percentage share of each global decile in the total footprints, respectively. The expenditure level of each decile group increases as the colour deepens. The red circle represents the level of per capita boundaries.

Viewed Articles
Keeping the global consumption within the planetary boundaries
November 13, 2024 | Nature | Source |  Introduction: Researchers from Shandong University (China) and the University of Maryland (USA) investigate the global distribution of environmental footprints a
Read More
Drawing transformation pathways for making use of joint effects of food and energy production with biodiversity agriphotovoltaics and electrified agricultural machinery
June 2023 | Journal of Environmental Management | Volume335Introduction: Agricultural land does not have to choose between producing food, generating renewable energy, and protecting biodiversity. Sch
Challenges and prospects for agricultural greenhouse gas mitigation pathways consistent with the Paris Agreement
May 22, 2020 | Frontiers in Sustainable Food Systems |  Introduction: Global pathways consistent with the Paris Agreement rely on substantial reductions in agricultural methane (CH4) and N2O alongside
Potential of  enhanced-efficiency nitrogen fertilizers in reducing  nitrogen and carbon losses in a sandy soil integrated crop-livestock  system
Introduction: Conventional urea is widely used because of its high nitrogen content and low cost, but substantial nitrogen can be lost through ammonia (NH₃-N) volatilization, reducing fertilizer effi
Climate-smart pest management: building resilience of farms and landscapes to changing pest threats
February 25, 2019 | Journal of Pest Science |  Introduction: Climate change is intensifying crop pest threats globally, disrupting species distribution, outbreak timing, and pest–natural enemy dynamic
Hybrid Machine learning models for PV output prediction: Harnessing Random Forest and LSTM-RNN for sustainable energy management in aquaponic system
April 2025 | Energy Conversion and Management | Volume 330Introduction: Dewi and her research team demonstrated how hybrid machine learning could improve solar energy forecasting and support more reli
TOP