Global wood fuel production estimates and implications

Abstract

Global wood fuel production can indicate opportunities and also challenges in sustainable development, forest management, and energy access. Estimates of wood fuel removals and charcoal production are essential for tracking global goals yet reliable measurements are rare. We synthesize existing understanding through a mechanistic, conceptual model and build on it to develop statistical models from official statistics and over 2000 newly identified data points. For 2019, we estimate 2525.7 million m³ of wood fuel removals globally, approximately 30% higher than previously understood. Our estimates are 50% higher in Africa and 40% higher in Asia, 10% lower in the Americas and 20% lower in Europe. Global production of wood charcoal is estimated at 70.5 million tonnes, approximately 50% higher than previous values; our estimates are 20% higher in Africa and 200% higher in Asia. These estimates describe global shifts in wood fuel removals and charcoal production and improve our understanding of the forest sector; they will likely underlie global models used to forecast future trends.

Fig. 1. The conceptual model for wood fuel production, describing the mechanistic drivers of wood fuel removals at the national level.

These boxes do not represent the predictor variables in the model but rather the theoretical foundation for selection of predictor variables from globally available data. Unshaded headings describe a first level assessment of drivers or types of indicator variables, e.g., key drivers, household needs, or impacts. Shaded boxes describe specific indicators of these drivers for which availability of global data was assessed. Final selection of predictor variables for clustering and for modelling is described in the Methods. Predictor variables used are referenced in Supplementary Table 1. GDP Gross domestic product; LPG liquid petroleum gas.

Fig. 2. Country and territory clusters used in modelling (Supplementary Table 2).

Note that cluster names are for simplicity only; all countries included in each cluster are listed in Supplementary Table 2. Countries or territories displayed in white were not included in the cluster analysis (see Methods). Country and territory boundaries are based on 2020 United Nations Geospatial mapping data. The boundaries and names shown and the designations used on these maps do not imply the expression of any opinion whatsoever on the part of the Food and Agriculture Organization of the United Nations (FAO) concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers and boundaries. Dashed lines on maps represent approximate border lines for which there may not yet be full agreement; colors chosen for each cluster are for visual clarity only and have no meaning.

Fig. 3. Estimated 2019 wood fuel removals per capita (m³).

We did not produce modelled estimates for Christmas Island, Cook Islands, Curaçao, Faroe Islands, Gibraltar, Greenland, Iceland, Malta, Marshall Islands, Monaco, Nauru, San Marino, Tokelau, Tuvalu, and Western Sahara. Instead, we used official data, where available, or assumed zero wood fuel removals. For the Russian Federation, we used current values from FAOSTAT (accessed 1 November 2024). Country and territory boundaries are based on 2020 United Nations Geospatial mapping data. The boundaries and names shown and the designations used on these maps do not imply the expression of any opinion whatsoever on the part of FAO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers and boundaries. Dashed lines on maps represent approximate border lines for which there may not yet be full agreement.

Fig. 4. Trends in wood fuel removals and charcoal production.

Estimated wood fuel removals (a) and charcoal production (b) by continent. Countries included in each continental summary are those in FAOSTAT. Points (shapes) show our estimates and dashed lines show current FAOSTAT values. Solid lines are smoothed long-term trends in the estimates (locally estimated scatterplot smoothing).

Fig. 5. Comparison of current FAOSTAT values and our estimates.

Current FAOSTAT values and our estimates are of a wood fuel removals and b charcoal production per total capita (triangles) and per rural capita (grey circles). Each point represents one year and country or territory for which estimates were produced. Triangle colors indicate whether values retained from FAOSTAT or found data were available, or whether neither retained nor found data were available on wood fuel demand for each point (year and country or territory).

Fig. 6. Variable permutation importance scores.

Variable permutation importance scores for the wood fuel demand per capita models (a) and fraction of demand met with charcoal models (b). Within each cluster of countries, scores are relative to the highest importance score, calculated by dividing each importance value by the highest importance score. Numbers to the right of the bars are the importance score rankings of variables for each cluster.

Fig. 7. Estimated charcoal production per capita (kg).

We did not produce modelled estimates for Christmas Island, Cook Islands, Curaçao, Faroe Islands, Gibraltar, Greenland, Iceland, Malta, Marshall Islands, Monaco, Nauru, San Marino, Tokelau, Tuvalu, and Western Sahara. Instead, we used official data, where available, or assumed zero wood fuel removals. For the Russian Federation, we used current values from FAOSTAT (accessed 1 November 2024). Country and territory boundaries are based on 2020 United Nations Geospatial mapping data. The boundaries and names shown and the designations used on these maps do not imply the expression of any opinion whatsoever on the part of FAO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers and boundaries. Dashed lines on maps represent approximate border lines for which there may not yet be full agreement.