Zhanwei Liu

Dr. Zhanwei Liu

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 September 2025 in Nature Sustainability
Verified
Journal Article
Open access
PDF file
BXBO XU
BO XU
ZLZhanwei Liu
Zhanwei Liu
SYShuyue Yan
Shuyue Yan
RSRafael J. P. Schmitt
Rafael J. P. Schmitt
XHXiaogang He
Xiaogang He
Hydropower is vital for climate mitigation by enabling low-carbon energy systems, but hydropower dams also trap sediment, a crucial resource for ecosystems and climate adaptation along downstream coastlines. Here we present a multisectoral integrated water–sediment–energy planning framework that fully internalizes the impacts of hydropower expansion, both on energy system costs and on foregone ecosystem services from reduced sediment supply for the Mekong River Basin. Our analysis indicates that full development of large hydropower could reduce sediment flows to the Mekong Delta by 41.2 ± 6.6 megatonnes per year (75 ± 9%). However, strategically replacing 19 high-sediment-trapping hydropower plants with solar, wind and energy storage alternatives could preserve up to 98% of sediment supply, with only a 4–6% (US$15.7–26.0 billion) increase in energy system costs over 2020–2050. Crucially, when sediment-related benefits—valued at US$12–28 million per megatonne per year—are considered, the additional costs of preserving sediment supply are nearly offset. The proposed framework offers a transferable approach to support sustainable low-carbon energy transitions while safeguarding sediment-dependent ecosystems worldwide.

1 October 2021 in Journal of Water Resources Planning and Management
Verified
Journal Article
Closed access
PDF file
SLShengli Liao
Shengli Liao
ZLZhanwei Liu
Zhanwei Liu
BLBenxi Liu
Benxi Liu
XWXinyu Wu
Xinyu Wu
CCChuntian Cheng
Chuntian Cheng
HSHuaying Su
Huaying Su
For low construction cost, the hydropower plant with multiple units sharing a common tunnel (HPMUST) has been widely used, which plays an essential role in the short-term scheduling of power systems worldwide. Thus, accurately solving short-term scheduling of HPMUST is important. Because water head loss of a tunnel is related to the discharge of multiple units, the existing methods noted in the literature tend to decompose this problem to find a near-optimal solution, instead of solving it integrally. With increasing requirements of short-term scheduling, this solution cannot meet practical operation requirements, and it becomes imperative to obtain high quality solutions by solving the original problem. For this purpose, this paper develops a mixed integer linear programming (MILP) model for short-term scheduling of HPMUST and proposes a gross water head-based iterative methodology to directly solve it. The MILP model takes account of not only head variations, common tunnel water head loss of multiple units, and water consumption during startup and shutdown of units, but also water consumption of crossing the vibration zone. Moreover, the paper proposes a novel generic linear programming formulation to model crossing the vibration zone, which can achieve a remarkable decline in number of times of crossing the vibration zone. The numerical results show that, compared with the average output model (AOM) and the real operation, the developed model and solution method can efficiently reduce water consumption of power generation within an acceptable time and meet practical operation requirements.

19 November 2025 in Communications Earth & Environment
Verified
Journal Article
Open access
PDF file
JXJingkai Xie
Jingkai Xie
ZLZhanwei Liu
Zhanwei Liu
SYShuyue Yan
Shuyue Yan
AZAlan D. Ziegler
Alan D. Ziegler
XSXiaogang Shi
Xiaogang Shi
XHXiaogang He
Xiaogang He
The Lower Mekong River Basin faces growing challenges in deploying variable renewable energy under intensifying hydroclimatic extremes. To evaluate these impacts, we develop an integrated modeling framework that links an energy expansion model with a synthetic streamflow generator. The results indicate that severe extended periods of low flows (10,000-year events) reduce hydropower output by 24.3 ± 8.7 terawatt-hour (−13.7% ± 4.9%), increasing wind capacity requirements by 10.6 ± 12.5 gigawatts (1.15% ± 1.36%) and raising system-wide costs by 8.7 ± 3.2 billion U.S. dollars. Conversely, extended periods of high flows of similar severity boost hydropower generation by 23.3 ± 9.2 terawatt-hours (+ 13.1% ± 5.2%), decreasing wind capacity needs by 9.3 ± 11.7 gigawatts (1.02% ± 1.27%) and lowering system costs by 7.8 ± 3.0 billion U.S. dollars. Prolonged drought conditions accelerate wind development to sustain decarbonization, while surplus flows from floods may delay investment—introducing long-term uncertainty for energy system resilience. These findings underscore the need for proactive, region-specific strategies to ensure resilient and sustainable energy transitions under intensifying hydroclimatic extremes.

10 November 2025 in Water Resources Research
Verified
Journal Article
Open access
XPXiao Peng
Xiao Peng
ZLZiwei Liu
Ziwei Liu
AZAlan D. Ziegler
Alan D. Ziegler
CZChao Zhang
Chao Zhang
ZLZhanwei Liu
Zhanwei Liu
XHXiaogang He
Xiaogang He
Conventional climate analyses rely on mean temperature trends to assess climate change impacts, yet this aggregation can obscure asymmetric shifts in the temperature distribution especially in threshold‐sensitive systems like snowpack. Here, we introduce a distributional diagnostic framework that decomposes winter temperature trends into median, cold‐tail (5th percentile), and warm‐tail (95th percentile) components across the Northern Hemisphere. Using 1981–2020 temperature records from the Berkeley Earth data set, we find that mean and median winter trends diverge substantially, with differences ranging from −0.19 to +0.41°C/decade (5th–95th percentile range) across snow‐affected grid cells. Mean trends systematically exceed median trends in 61% of locations, and these divergences are driven by spatially structured and climatology‐dependent tail behavior. We classify these patterns into four types based on directions of cold and warm tail trends relative to median trends and show that each aligns with distinct climatological regimes. Using multivariate regression, we show that median temperature trends consistently outperform means in explaining March snow water equivalent trends, while adding tail metrics further improves explanatory power. Tail contributions vary by climate: in extremely cold zones (−20°C) warm‐tail trends dominate, in moderately cold zones (−20 to −10°C) median trends and in near‐freezing regions (−10 to 0°C) cold‐tail trends. These results demonstrate that asymmetric distributional change is a key control on spring snowpack trends and highlight the need for percentile‐based diagnostics in climate impact assessments.

16 April 2026 in Water Resources Research
Verified
Journal Article
Open access
ZLZiwei Liu
Ziwei Liu
XPXiao Peng
Xiao Peng
ZLZhanwei Liu
Zhanwei Liu
AZAlan D. Ziegler
Alan D. Ziegler
Chao LiChao Li
Chao Li
XHXiaogang He
Xiaogang He
Cold region water resources derive from a delicate balance between rainfall‐derived runoff and snowmelt‐derived runoff. Global warming is expected to disrupt this balance by increasing the former meanwhile decreasing snowmelt‐derived contributions. However, significant uncertainties in climate model projections have made it difficult to quantify these changes with high confidence. In this study, we apply observational constraints based on global mean surface air temperature to refine projected runoff changes in snow‐affected regions in the Northern Hemisphere. Our findings suggest that the shift from snowmelt‐to rainfall‐derived runoff will occur gradually, with smaller changes than anticipated. Specifically, under the high‐emission scenario through the end of this century, projected decreases in snowmelt‐derived runoff are reduced by 13% (from unconstrained 23.8% to constrained 21.0%), while rainfall‐derived runoff increases are more moderate by 12% (from unconstrained 43.2% to constrained 37.8%), with associated uncertainties (>66% confidence) narrowing by 40%. This recalibration of runoff components reduces total runoff increases by 16% (from unconstrained 18.4% to constrained 15.4%). In regions with climatological winter temperatures below −15°C, substantial declines in snowmelt‐derived runoff are unlikely before 2100, as most precipitation continues to fall as snowfall despite rapid warming. While these refined projections provide sufficient time for adaptation, inevitable shifts in runoff timing and volume still pose flood and drought risks, emphasizing the need for informed water management strategies.

19 March 2026 in Energies
Verified
Journal Article
Open access
PDF file
JHJing Hu
Jing Hu
BXBO XU
BO XU
HZHuicheng Zhou
Huicheng Zhou
ZLZhanwei Liu
Zhanwei Liu
Quantifying the complex spatial–temporal correlations and generating representative high-dimensional coupled scenario sets are essential for the robust planning and risk assessment of large-scale hybrid energy systems (HESs). Although numerous models have been developed for this purpose, as the number of plants scales up to hundreds, existing approaches suffer from the curse of dimensionality, often resulting in high computational burden, posterior collapse, and distributional oversmoothing. To address this gap, this paper proposes a Spatial-Clustering Conditional Variational Auto-Encoder (SC-CVAE) framework, which employs spatial clustering to decompose the high-dimensional global problem into tractable subproblems and integrates adaptive deep networks to accurately capture high-dimensional spatiotemporal complementarity. Case studies on the Yalong River energy base, featuring massive wind and solar integration, demonstrate that SC-CVAE reduces global spatial correlation error by 56% compared to the Independent Baseline, while achieving a 2.4-fold computational speedup over the monolithic Global Baseline. Crucially, by mitigating posterior collapse to alleviate oversmoothing effects inherent in high-dimensional VAEs, the proposed framework improves the capture rate of high-impact extreme events by 3.4-fold and reduces the Energy Score error by 65%. This high-fidelity reconstruction of tail characteristics provides a more reliable basis for identifying supply-deficit risks in basin-wide HESs. The proposed framework enables scalable and high-fidelity generative modeling, establishing a robust methodology for stochastic optimization and long-term security assessments in the global transition toward decarbonized power systems.