Page:Quantifying a realistic, worldwide wind and solar electricity supply.pdf/12

 250

Y.Y. Deng et al. / Global Environmental Change 31 (2015) 239–252

5. Conclusion

Fig. 6. Potential results by distance. Annual achievable electricity potential on land is shown for 2070, differentiated by distance from infrastructure and per resource type. Top: High availability case, bottom: Low availability case.

Combining resources on land, sea and buildings yields renewable electricity potentials which outstrip the prospective long-term regional demand for almost all regions. The exception is South Asia, where the Medium availability case may not provide sufﬁcient electricity if the higher demand projections transpire. The picture looks even less uniform at individual country level: Brazil and Egypt have the prospect of being self-sufﬁcient even with high demand, based on solar and wind electricity alone. In contrast, India and Nigeria would not necessarily be able to satisfy demand from national sources alone at 24–40 GJ/cap/a, except in the High availability case: Nigeria’s long-term resource base is 16– 52 GJ/cap/a, although intra-regional trade within West Africa could satisfy regional demand in the Medium case. India’s longterm potential is 23–66 GJ/cap/a but it lies within a similarly constrained region. As such, India and its region will need to unlock the potentials in the High availability case or supplement supply with other energy options or long-distance transport of renewable electricity. These constraints are not imminent however: electricity demand today is only 2.1 GJ/cap/a in India and 0.5 GJ/cap/a in Nigeria. Based on these demand assumptions, between 60% and 90% of the world’s population in the 2080s will live in countries which could be self-sufﬁcient in the Medium case, depending on the demand per capita. The remaining 10–40% would need to mobilise potentials found in the High availability case, import electricity from neighbouring countries or use different sources of electricity. 4.5. How far do we have to transport this electricity? We addressed the question of usability of renewable potential in remote locations by differentiating potentials based on their distance from ‘‘infrastructure’’, which we use as a marker for the distance from an electricity grid. We deﬁned the existence of this marker infrastructure based on datasets of population density, railways, and urban land cover. The results are shown in Fig. 6 for the Low and High cases for 2070. On land, between one to two thirds of the potential are situated in the most accessible distance ranges. The much larger potential offshore is primarily concentrated in higher distance categories, consistent with the very restrictive availability factors we have used near-shore.

We have presented a bottom-up assessment of renewable electricity potentials for ﬁve technologies: PV on buildings and on land, CSP, wind on land and on sea. The study builds on high resolution publicly available geographic datasets and presents aggregated results at regional level. Global totals fall within the ranges found in the existing literature. The debate about electricity potentials is often dominated by the expected development of the conversion technology. Our study conﬁrms, however, that it is more important to carefully assess the overall availability of surface area, as this carries the largest uncertainty. Within the assessment of the land area we have taken care to try and estimate a feasible availability factor and show the inﬂuence that variations in this availability factor have on the ﬁnal results. In establishing the availability factor, we have tried to use countries with high technology penetration as a guide. For wind power, for example, current (end of 2011) installed capacities imply land penetration across all land types of around 1% of suitable area. Given that deployment still continues in these countries, but current societal debate suggests that it is reaching local thresholds in some regions, we have estimated the long-term availability factors for wind to be around 5–6% for our Medium case. To enable interpretation of our ﬁndings, we have also shown comparison of our long-term (2070) estimates with prospective future demand. We show that even our lowest estimates would provide (just) enough renewable electricity for a population of 10 billion, expected for the early 2080s, if the total achievable potential could be mobilised at global level. When focusing on individual regions and countries, differences emerge: some regions could easily satisfy current and future demand whereas others may run into supply shortages in the Low availability cases, especially where the distance to infrastructure may be problematic. Whether all of this potential could be deployed requires an energy system scenario calculation which also addresses issues around system integration of various sources and their interaction with demand patterns which have not been addressed explicitly in this study. However, focusing simply on the resource potential does deﬁne the range of possibilities for the future. While there is scope for substantial renewables growth from today’s levels, in the long run the large global potentials may contain within them constraints in speciﬁc countries or regions. These ﬁndings may guide society in formulating long-term visions for energy systems as a basis for energy policy in the decades to come. Acknowledgements Helen Saehr and Pim Rooijmans prepared the GIS maps. Thomas Winkel and Paul Noothout assisted with analysis for hydro- and geothermal electricity. Pieter van Breevoort found comparisons to previous studies. Kjell Bettgenha¨user provided input to the buildings reference country modelling. Kornelis Blok, Monique Hoogwijk, Jan Coelingh, Said Bijary, Dirk Schoenmakers, Peter Bange, and Faruk Dervis provided input into and review of the land- and sea approach and assumptions.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.gloenvcha.2015.01.005. References Adams, A.S., Keith, D.W., 2013. Are global wind power resource estimates overstated? Environ. Res. Lett. 8.