Abstract
The intrinsic variability of solar and wind energy, compounded by their rapid expansion, has intensified power curtailment challenges1,2. Although spatiotemporal complementarity between these resources is widely recognized as a pathway to enhance renewable integration and reduce balancing requirements3,4,5,6,7,8,9,10,11,12,13,14,15,16, existing assessments are largely based on hypothetical deployments17,18,19,20,21,22,23,24. Consequently, how solar–wind complementarity manifests under real-world infrastructure and shapes system-level integration outcomes remains unclear. Here we develop a unified national inventory to enable a data-driven assessment of solar–wind complementarity. The inventory covers 319,972 solar photovoltaic facilities and 91,609 wind turbines in 2022, identified from sub-metre satellite imagery using a deep-learning-based framework. Using this dataset, we show that solar–wind complementarity substantially reduces generation variability, with effectiveness increasing as the geographic scope of pairing expands. At the system level, nationwide inter-provincial coordination raises effective renewable penetration by 99.88 TWh in an 80% dispatchable-flexibility system, corresponding to 9.1% of total solar and wind generation, or approximately 120 h of national average load. These findings demonstrate that energy complementarity is a scalable, system-wide mechanism for advancing solar and wind penetration, offering broadly applicable insights into the role of inter-regional coordination in enhancing renewable integration in large power systems.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to the full article PDF.
USD 39.95
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
High-resolution geospatial data describing individual solar PV facilities and wind turbines cannot be publicly released owing to sensitivity associated with critical energy infrastructure. Aggregated county-level hourly solar and wind generation time series used in the complementarity and penetration analyses, together with processed provincial load profiles, are publicly available at Zenodo (https://doi.org/10.5281/zenodo.19079543)59. Example geospatial vector data for wind turbines and solar PV installations in Qinghai province are also provided as an illustrative subset of the detection dataset. High-resolution satellite imagery used for facility identification was mainly accessed through Google Earth for non-commercial academic research. Further data supporting the findings of this study may be available from the corresponding author on reasonable request and subject to institutional approval. Source data are provided with this paper.
Code availability
Scripts used to reproduce the county-level complementarity analysis and provincial penetration analysis are publicly available at Zenodo (https://doi.org/10.5281/zenodo.19079543)59. Wind turbine detection and PV installation segmentation were implemented using the open-source frameworks mmdetection and mmsegmentation. Configuration files specific to the deep learning experiments conducted in this study are provided in the same repository.
References
Grams, C. M., Beerli, R., Pfenninger, S., Staffell, I. & Wernli, H. Balancing Europe’s wind-power output through spatial deployment informed by weather regimes. Nat. Clim. Change 7, 557–562 (2017).
Lu, X. et al. Challenges faced by China compared with the US in developing wind power. Nat. Energy 1, 16061 (2016).
Heide, D. et al. Seasonal optimal mix of wind and solar power in a future, highly renewable Europe. Renew. Energy 35, 2483–2489 (2010).
Heide, D., Greiner, M., von Bremen, L. & Hoffmann, C. Reduced storage and balancing needs in a fully renewable European power system with excess wind and solar power generation. Renew. Energy 36, 2515–2523 (2011).
Xiao, M., Wetzel, M., Pregger, T., Simon, S. & Scholz, Y. Modeling the supply of renewable electricity to metropolitan regions in China. Energies 13, 3042 (2020).
Pedruzzi, R. et al. Review of mapping analysis and complementarity between solar and wind energy sources. Energy 283, 129045 (2023).
Engeland, K. et al. Space-time variability of climate variables and intermittent renewable electricity production – a review. Renew. Sustain. Energy Rev. 79, 600–617 (2017).
Jurasz, J., Canales, F., Kies, A., Guezgouz, M. & Beluco, A. A review on the complementarity of renewable energy sources: concept, metrics, application and future research directions. Sol. Energy 195, 703–724 (2020).
Schindler, D., Behr, H. D. & Jung, C. On the spatiotemporal variability and potential of complementarity of wind and solar resources. Energy Convers. Manag. 218, 113016 (2020).
Kapica, J., Canales, F. A. & Jurasz, J. Global atlas of solar and wind resources temporal complementarity. Energy Convers. Manag. 246, 114692 (2021).
Hajou, A., El Mghouchi, Y. & Chaoui, M. A new solar-wind complementarity index: an application to the climate of Morocco. Renew. Energy 227, 120490 (2024).
Gao, Y. et al. The wind-solar hybrid energy could serve as a stable power source at multiple time scale in China mainland. Energy 305, 132294 (2024).
Prasad, A. A., Taylor, R. A. & Kay, M. Assessment of solar and wind resource synergy in Australia. Appl. Energy 190, 354–367 (2017).
Hu, J. et al. Reducing energy storage demand by spatial-temporal coordination of multienergy systems. Appl. Energy 329, 120277 (2023).
Li, P., Lian, J., Ma, C. & Zhang, J. Complementarity and development potential assessment of offshore wind and solar resources in China seas. Energy Convers. Manag. 296, 117705 (2023).
de Souza Nascimento, M. M. et al. Offshore wind and solar complementarity in Brazil: a theoretical and technical potential assessment. Energy Convers. Manag. 270, 116194 (2022).
Kruitwagen, L. et al. A global inventory of photovoltaic solar energy generating units. Nature 598, 604–610 (2021).
Zhang, X., Xu, M., Wang, S., Huang, Y. & Xie, Z. Mapping photovoltaic power plants in China using Landsat, random forest, and Google Earth Engine. Earth Syst. Sci. Data 14, 3743–3755 (2022).
Chen, Y., Zhou, J., Ge, Y. & Dong, J. Uncovering the rapid expansion of photovoltaic power plants in China from 2010 to 2022 using satellite data and deep learning. Remote Sens. Environ. 305, 114100 (2024).
Xia, Z. et al. Mapping the rapid development of photovoltaic power stations in northwestern China using remote sensing. Energy Rep. 8, 4117–4127 (2022).
Jiang, H. et al. Multi-resolution dataset for photovoltaic panel segmentation from satellite and aerial imagery. Earth Syst. Sci. Data Discuss. 2021, 1–17 (2021).
Yu, J., Wang, Z., Majumdar, A. & Rajagopal, R. Deepsolar: a machine learning framework to efficiently construct a solar deployment database in the United States. Joule 2, 2605–2617 (2018).
Ortiz, A. et al. An artificial intelligence dataset for solar energy locations in India. Sci. Data 9, 497 (2022).
Malof, J. M., Bradbury, K., Collins, L. M. & Newell, R. G. Automatic detection of solar photovoltaic arrays in high resolution aerial imagery. Appl. Energy 183, 229–240 (2016).
Zhang, Z. et al. Carbon mitigation potential afforded by rooftop photovoltaic in China. Nat. Commun. 14, 2347 (2023).
Mallapaty, S. et al. How China could be carbon neutral by mid-century. Nature 586, 482–483 (2020).
Wang, J. et al. Inherent spatiotemporal uncertainty of renewable power in China. Nat. Commun. 14, 5379 (2023).
National Energy Administration, China. National Energy Administration releases 2024 national power industry statistics. National Energy Administration https://www.nea.gov.cn/20250121/097bfd7c1cd3498897639857d86d5dac/c.html (2025).
Yin, J., Molini, A. & Porporato, A. Impacts of solar intermittency on future photovoltaic reliability. Nat. Commun. 11, 4781 (2020).
Howe, C. As China’s renewable capacity soars, utilisation lags, data show. Reuters https://www.reuters.com/sustainability/climate-energy/chinas-renewable-capacity-soars-utilisation-lags-data-show-2025-08-05/ (2025).
Ren, G., Wan, J., Liu, J. & Yu, D. Spatial and temporal assessments of complementarity for renewable energy resources in China. Energy 177, 262–275 (2019).
Jani, H. K., Kachhwaha, S. S., Nagababu, G. & Das, A. Temporal and spatial simultaneity assessment of wind-solar energy resources in India by statistical analysis and machine learning clustering approach. Energy 248, 123586 (2022).
Santos-Alamillos, F., Pozo-Vázquez, D., Ruiz-Arias, J. A., Von Bremen, L. & Tovar-Pescador, J. Combining wind farms with concentrating solar plants to provide stable renewable power. Renew. Energy 76, 539–550 (2015).
Xu, L., Wang, Z. & Liu, Y. The spatial and temporal variation features of wind-sun complementarity in China. Energy Convers. Manag. 154, 138–148 (2017).
Jerez, S. et al. An action-oriented approach to make the most of the wind and solar power complementarity. Earth’s Future 11, e2022EF003332 (2023).
Guo, Y. et al. Variation-based complementarity assessment between wind and solar resources in China. Energy Convers. Manag. 278, 116726 (2023).
Johlas, H., Witherby, S. & Doyle, J. R. Storage requirements for high grid penetration of wind and solar power for the MISO region of North America: a case study. Renew. Energy 146, 1315–1324 (2020).
Tong, D. et al. Geophysical constraints on the reliability of solar and wind power worldwide. Nat. Commun. 12, 6146 (2021).
Liu, L. et al. Optimizing wind/solar combinations at finer scales to mitigate renewable energy variability in China. Renew. Sustain. Energy Rev. 132, 110151 (2020).
Cao, K., Zhou, C., Church, R., Li, X. & Li, W. Revisiting spatial optimization in the era of geospatial big data and GeoAI. Int. J. Appl. Earth Obs. Geoinf. 129, 103832 (2024).
Li, C. & Song, L. Regional differences and spatial convergence of green development in China. Sustainability 14, 8511 (2022).
Fang, W. et al. Assessment of wind and solar power potential and their temporal complementarity in China’s northwestern provinces: insights from ERA5 reanalysis. Energies 16, 7109 (2023).
Hersbach, H. et al. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc. 146, 1999–2049 (2020).
Sun, K., Xiao, B., Liu, D. & Wang, J. Deep high-resolution representation learning for human pose estimation. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 5693–5703 (IEEE, 2019).
Wang, J. et al. Deep high-resolution representation learning for visual recognition. IEEE Trans. Pattern Anal. Mach. Intell. 43, 3349–3364 (2021).
Zhang, H., Wang, Y., Dayoub, F. & Sunderhauf, N. VarifocalNet: an IoU-aware dense object detector. In Proc. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 8514–8523 (IEEE, 2021).
Perez, R., Ineichen, P., Seals, R., Michalsky, J. & Stewart, R. Modeling daylight availability and irradiance components from direct and global irradiance. Sol. Energy 44, 271–289 (1990).
Reda, I. & Andreas, A. Solar position algorithm for solar radiation applications. Sol. Energy 76, 577–589 (2004).
Holmgren, W. F., Hansen, C. W. & Mikofski, M. A. pvlib python: a Python package for modeling solar energy systems. J. Open Source Softw. 3, 884 (2018).
Chen, S. et al. The potential of photovoltaics to power the belt and road initiative. Joule 3, 1895–1912 (2019).
Hu, W., Scholz, Y., Yeligeti, M., Bremen, L. & Deng, Y. Downscaling ERA5 wind speed data: a machine learning approach considering topographic influences. Environ. Res. Lett. 18, 094007 (2023).
Gruber, K., Regner, P., Wehrle, S., Zeyringer, M. & Schmidt, J. Towards global validation of wind power simulations: a multi-country assessment of wind power simulation from MERRA-2 and ERA-5 reanalyses bias-corrected with the Global Wind Atlas. Energy 238, 121520 (2022).
Goldwind. Goldwind sustainability report. Goldwind https://web.archive.org/web/20190819172242/http://goldwindglobal.com/images/about/duty/report/2017.pdf (2017).
Feng, J., Feng, L., Wang, J. & King, C. W. Evaluation of the onshore wind energy potential in mainland China—based on GIS modeling and EROI analysis. Resour. Conserv. Recycl. 152, 104484 (2020).
Rinne, E., Holttinen, H., Kiviluoma, J. & Rissanen, S. Effects of turbine technology and land use on wind power resource potential. Nat. Energy 3, 494–500 (2018).
National Energy Administration, China. Typical load profiles of provincial power grid. National Energy Administration https://www.ndrc.gov.cn/xwdt/tzgg/202012/P020201202546044875868.pdf (2020).
Jiang, H. et al. High-resolution analysis of rooftop photovoltaic potential based on hourly generation simulations and load profiles. Appl. Energy 348, 121553 (2023).
Edmonds, J. Paths, trees, and flowers. Can. J. Math. 17, 449–467 (1965).
Hu, Y. & Liu, Y. China Solar–wind complementarity analysis dataset and code. Zenodo https://doi.org/10.5281/zenodo.19079543 (2026).
Acknowledgements
We thank A. Python for helpful comments on the manuscript and Y. Wang for assistance with data annotation. This work was supported by the National Natural Science Foundation of China (grant nos. 42430106, 42401407, 42371468 and 72571005) and the High Performance Computing Platform of Peking University (no. HPC2306190166).
Author information
Authors and Affiliations
Contributions
Y.H., F.Z., C.Y., Z.J. and Y.L. conceived and designed the research. Y.H. performed the analysis and wrote the first version of the manuscript. J.Y. and C.Y. designed the deep learning pipeline and conducted the experiments. M.Z., Q.C., J.W. and H.Z. assisted with image data processing and cloud computing. Y.H., H.J., C.Z. and F.Z. contributed to discussions on the method. Y.H., H.J., Y.L., Z.J, F.Z., C.Z., L.Y., S.M., X.L., Q.D., C.W., G.Y. and W.Z. assisted with quality control and contributed to manuscript revision.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature thanks Wenxuan Hu, Hardik Jani and Yvonne Scholz for their contribution to the peer review of this work. Peer reviewer reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Deep learning framework diagram.
The workflow consists of four steps: (1) manual sample annotation; (2) deep learning model training; (3) nationwide image inference; and (4) manual verification. During the manual verification step, for regions in which certain installations were missed during the initial Google imagery inference, Jilin-1 satellite imagery was used as an alternative data source to re-infer these areas, thereby recovering undetected installations and improving overall recall. Base map derived from Natural Earth shaded relief data (public domain).
Extended Data Fig. 2 Wind generation patterns in five counties matched for inter-county solar–wind complementarity.
The 24-h wind power generation profiles show trends opposite to solar power, with higher outputs during night-time and lower outputs during daytime.
Extended Data Fig. 3 County-level multiyear complementarity stability (2000–2022).
a–c, Inter-annual variability of Kendall rank correlation τ. a, Same-county solar–wind complementarity (interquartile range (IQR) of τ across 2000–2022). b, Optimal partner selected under the solar-source strategy (IQR of τ). c, Optimal partner selected under the wind-source strategy (IQR of τ). Across all counties, inter-annual IQR values are small (typically <0.02–0.04), indicating that year-to-year meteorological fluctuations perturb complementarity only within a narrow band relative to the much larger spatial heterogeneity observed in the baseline year (2022). d–f, Stability of optimal partner relationships across meteorological years. Panels report the frequency with which the 2022 top partner remains within the top 10 candidate partners ranked by τ during 2000–2022. d, Solar–wind links. e, Wind–solar links. f, Wind–wind links. The top-1-in-top-10 frequency commonly exceeds 0.7, demonstrating that inter-annual variability mainly reshuffles a small set of near-equivalent candidates while preserving the broader spatial structure of complementarity. Counties with no solar or wind generation are shown in grey.
Extended Data Fig. 4 Province-level stability of optimal inter-provincial pairing across meteorological years (2000–2022).
a, Top-1 partner retention under strategies B and C (flexibility = 0.8). Maps show the fraction of years (2000– 2022) in which the 2022 optimal partner of each province remains the top-ranked partner. Strategy B (left) exhibits generally high retention, indicating that most provinces repeatedly select the same neighbouring partner or a very small set of adjacent partners. Strategy C (right) shows lower top-1 retention, not because complementarity is unstable but because several partner provinces often offer near-equivalent complementarity for a given province, so small year-to-year meteorological variations can shift the single best partner within this limited candidate set. b, Heatmaps of partner-selection frequency under strategies B and C (flexibility = 0.8). Each cell reports the share of years in which a partner province is the top-ranked match for the source province. Under strategy B (left), most provinces concentrate weight on one or two clearly dominant partners, consistent with the high retention in panel a. Under strategy C (right), many load provinces—particularly in central and eastern China—draw repeatedly from a small group of high-resource western and northwestern partners, leading to several partners sharing intermediate frequencies rather than a single persistent top-1 choice. Together, these results show that the structure of inter-provincial complementarity is highly robust across distinct meteorological years, even though the identity of the single best partner is not always unique.
Extended Data Fig. 5 Cost-aware provincial pairing confirms the advantage of nationwide complementarity.
a, Heatmap of the objective improvement when switching from neighbouring-only pairing (strategy B) to nationwide pairing (strategy C), computed as ∆objective = objB − objC across transmission-cost settings (rows) and system flexibility levels (columns). Positive values indicate that strategy C achieves a lower (better) value of the joint utilization–cost objective. b, Trade-off between renewable utilization and the joint objective when switching from strategy B to strategy C. Each point corresponds to one setting–flexibility combination, with the horizontal axis showing the change in renewable utilization (utilC − utilB) and the vertical axis showing the change in the joint objective (objC − objB); values below zero indicate that strategy C improves the objective. Across all evaluated settings and flexibility levels, strategy C consistently yields higher renewable utilization and a lower joint objective, demonstrating that the benefits of broader spatial coordination persist when transmission costs and existing-network feasibility constraints are explicitly incorporated.
Extended Data Fig. 6 Sensitivity of renewable curtailment to storage capacity and power constraints.
Renewable curtailment as a function of the storage scaling factor α for the four complementarity strategies (A–D) under different system flexibility levels (from 100% to 70%). The scaling factor α proportionally limits both storage energy capacity and charging/discharging power relative to the unconstrained reference. Across all flexibility levels, tighter storage constraints lead to increased curtailment, but the relative ordering of strategies remains unchanged, with broader spatial coordination consistently resulting in lower curtailment. Notably, nationwide provincial pairing (strategy C) and full national integration (strategy D) exhibit substantially reduced sensitivity to storage limitations compared with more localized strategies.
Supplementary information
Supplementary Information (download PDF )
This Supplementary Information file includes Supplementary Notes 1–12, Supplementary Figs. 1–19, Supplementary Tables 1–12 and further references.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Hu, Y., Jiang, H., Zhang, C. et al. Advancing solar and wind penetration in China through energy complementarity. Nature (2026). https://doi.org/10.1038/s41586-026-10570-z
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41586-026-10570-z
