The methodological heart of the study is the China Integrated Sustainable Power System Optimization model, known as CISPO, which integrates land-use constraints and renewable resource assessments directly into a capacity-expansion and hourly-dispatch framework. The researchers simulated five-year intervals from 2030 to 2060 across three contrasting land-supply scenarios, ranging from an open regime in which developers can access most technically suitable terrain to a conservative regime reflecting strict protections for cropland, forests, grasslands and water shorelines. These scenarios draw on China’s actual regulatory landscape, including national rules governing the use of forest land for photovoltaic stations and wind farms, the Grassland Law, the Land Administration Law, and guidelines restricting construction along rivers and lakes. By embedding these real policy boundaries into the optimization, the team captured how legal and environmental restrictions, not just physical geography, determine where wind and solar can actually be built.

One of the study’s most striking conceptual contributions is the treatment of land as a spatially heterogeneous production factor with a quantifiable marginal value. The researchers extracted the shadow prices of the variable renewable energy capacity constraints in each grid cell, effectively asking how much the whole power system would save if one more unit of renewable capacity could be squeezed into a particular location. The results reveal a dramatic geographic asymmetry. High-value land is concentrated in eastern and central China, close to the demand centers where electricity is consumed. Siting additional renewables there reduces the need for expensive long-distance transmission lines and for overbuilding remote generation in the resource-rich but demand-poor west. In effect, a square kilometer of suitable land near Shanghai or Wuhan is worth far more to the system than a comparable parcel in the sparsely populated northwest, even if the northwest enjoys better sunshine and stronger winds.

When stringent land-use restrictions are imposed, the optimization model responds in a predictable but consequential way: renewable deployment migrates. With the most accessible land near demand centers placed off limits, the model shifts wind and solar construction toward the resource-rich frontier regions. Northwestern solar deployment increases by approximately 48 percent relative to the base scenario by 2060. This relocation is not free. Moving generation farther from load centers raises inter-regional transmission requirements by roughly 20 percent in 2060, meaning China would need to build substantially more high-voltage corridors to carry electricity from the remote interior to the coastal megacities. Transmission lines themselves consume land, face their own permitting hurdles and add capital cost, creating a feedback loop in which land scarcity in one domain generates infrastructure pressure in another.

Despite these frictions, the study delivers a reassuring headline: carbon neutrality remains technically achievable even under the most restrictive land assumptions. The model finds feasible pathways to a zero-carbon power system in all scenarios, meaning land constraints do not derail decarbonization. What changes is the price tag and the configuration of the system. The 4.1 percent cost penalty, while modest relative to the enormous overall investment required for full decarbonization, represents tens of billions of dollars per year by 2060, resources that could otherwise fund storage, grid modernization or electrification of other sectors. The finding reframes land policy as an economic lever: decisions about which parcels are open to renewable development directly shape the national cost of climate mitigation.

The study also connects land scarcity to electricity prices, an issue of growing political salience as China deepens its power market reform. Using marginal-cost-based price formation, the researchers show that the cost differences across land-use scenarios could become visible in market-based electricity prices, depending on market design and cost-recovery mechanisms. Recent reforms, including the 2025 policy notice on deepening market-oriented reform of on-grid tariffs for new energy, are pushing renewables toward full market participation. If land restrictions systematically raise the marginal cost of supply, consumers and industries in land-constrained regions could face persistently higher prices, while the benefits of cheap remote renewables are eroded by transmission charges. This linkage between land policy and retail electricity costs is likely to become a central tension in China’s energy governance.

The research arrives amid a global reckoning with the land intensity of renewable energy. Solar and wind are far less land-efficient per unit of energy than fossil or nuclear plants, and studies from the western United States to Brazil have documented mounting conflicts between renewable buildout and agriculture, biodiversity and community interests. In China, a surprising share of existing photovoltaic capacity has been installed on cropland, prompting regulatory tightening. The new study quantifies, for the first time at national scale, what such tightening costs the power system, transforming an abstract land-use debate into concrete economic terms. It also highlights the value of land-saving alternatives that appear in the literature, including rooftop photovoltaics, floating solar on reservoirs, agrivoltaics that combine farming with electricity generation, and photovoltaic deployment on degraded or industrial land such as open-pit mines.

For policymakers, the implications cut in several directions. First, the shadow-price analysis offers a spatial prioritization tool: protecting the highest-value land near demand centers imposes the largest system costs, so land-use decisions in eastern and central China deserve the closest scrutiny. Second, the 20 percent increase in inter-regional transmission needs under restrictive scenarios argues for proactive grid planning, since transmission projects take years to permit and build. Third, the modest overall cost penalty suggests that China has genuine flexibility; even a conservative land regime does not make decarbonization prohibitively expensive, provided the system can re-optimize around the constraints. The authors’ framework, with data and code publicly available on GitHub, allows other researchers and planners to test alternative land policies, resource assumptions and technology costs.

The study’s scope is limited to the power sector, and the authors note that cost differences will surface in market prices only as reform deepens and depending on how costs are recovered. Extending the analysis to heating, hydrogen production and electrified transport, all of which will compete for the same renewable generation, is an obvious next step. Yet the core message is already clear and widely applicable: as the world pivots from whether to decarbonize to how, the humble question of where to put the panels and turbines has become a first-order economic problem. China’s experience suggests that every country plotting a net-zero grid should start counting not just its terawatt-hours but its square kilometers, because in the land-energy nexus, geography is destiny and scarcity has a price.

Subject of Research: Land availability constraints on China's transition to a carbon-neutral power system

Article Title: Navigating the land–energy nexus for carbon-neutral power systems in China

Article References: Zhu, Z., Yu, R., & Zhang, D. (2026). Navigating the land–energy nexus for carbon-neutral power systems in China. Nature Sustainability. https://doi.org/10.1038/s41893-026-01941-5

Image Credits: AI Generated

DOI: 10.1038/s41893-026-01941-5

Keywords: China, carbon neutrality, power system optimization, land use, renewable energy, solar power, wind power, transmission, shadow prices, electricity markets, Nature Sustainability, energy modelling

Researchers Aparajita Jaiswal and Gaurav Nanda of Purdue University studied 38 undergraduate students enrolled in a junior-level operations management course during a unit on linear programming in manufacturing contexts. The design of the assignment was deliberate. Students first solved a linear programming problem manually using the graphical corner-point method, which requires defining an objective function, identifying constraints, representing the feasible region, and finding the corner point that optimizes the objective. Only afterward did they solve the same problem using ChatGPT. The students then submitted written reflections describing how their manual reasoning compared with the AI’s approach, whether the AI helped them identify errors or confirm their logic, and what they learned about concepts such as optimal solutions, shadow prices, and sensitivity analysis. Where possible, the researchers also collected the students’ actual ChatGPT interaction records—chat transcripts, share links, screenshots, and uploaded images—to check whether what students said matched what they did.

To make sense of this variation, the team applied the ICAP framework, a well-established model from the learning sciences developed by Michelene Chi and Ruth Wylie. ICAP sorts cognitive engagement into four hierarchical modes: Passive, Active, Constructive, and Interactive. Passive engagement means simply receiving information; Active engagement means manipulating what is given, such as checking or confirming work; Constructive engagement means generating new ideas beyond the material, such as asking for explanations or drawing inferences; and Interactive engagement means dialogic co-construction, in which both parties make generative, contingent contributions that build on each other. The framework’s core claim is that deeper modes predict stronger learning—a claim the study put directly to the test in the context of human-AI interaction.

The results revealed three clearly documented patterns. Passive engagement appeared when students submitted the problem and read the AI’s output without using the interaction to compare, verify, reinterpret, or extend anything. One student wrote plainly, “With ChatGPT, I just copy pasted the question/problem and then it showed me the steps and got the correct solution.” Another described the process as “a simple copy and paste and it spit out the answer with work shown.” Interestingly, the passive cases were not uniform. Some students provided minimal direction; others uploaded course materials or submitted images of the problem so the AI would follow the class method. What united them was the absence of any intentional follow-through: noticing that the AI skipped the graph or omitted a feasible point, for instance, did not translate into probing questions or deliberate verification.

Active engagement looked different. Here, students positioned ChatGPT as a confirmation or reassurance tool, using it to check work they had already completed by hand. “AI confirmed my answer and the logic used seems to be the same,” one student reported. Another noted, “The AI approached the problem the same way I did and helped me confirm that I was following the steps correctly without error.” A third summed it up bluntly: “All it did to help was verify my answers and method.” In these cases, students were doing more than receiving—they were comparing and validating—but the interaction remained centered on correctness rather than on building new understanding. One illustrative case showed a student who asked ChatGPT to solve the problem with the corner-point method, received a mathematically accurate solution identifying the optimum at the point (0,3) with an objective value of 45, and then requested a sensitivity analysis. Although the second prompt extended the content, the exchange stayed confirmatory in structure, which is why the researchers classified it as Active rather than Constructive.

Constructive engagement emerged when students sought explanations and interpretation, particularly around the concepts they found hardest: sensitivity analysis, shadow prices, and binding versus non-binding constraints. “I didn’t really understand how sensitivity analysis and shadow prices worked until it explained them to me,” one student wrote. Another said the AI “pointed to details I would miss out on, such as identifying binding constraints and shadow prices,” and a third reflected that “sensitivity analysis made more sense once the AI broke it down in normal language.” In these cases, students used the chatbot not to obtain an answer but to understand what the answer meant—how a shadow price of 3.75 on a binding constraint translates into the value of relaxing that constraint, or how allowable ranges describe how far coefficients can shift before the optimal solution changes. The researchers were careful to note, however, that the study did not assess whether these AI-generated explanations actually improved conceptual understanding on later tests or transfer problems.

The most theoretically significant finding was what the researchers did not find. No student reflection and no available interaction record met the operational criteria for Interactive engagement, the highest ICAP mode requiring iterative, contingent co-construction in which each party’s contributions build on and challenge the other. Even students who used multi-turn exchanges tended to remain in a request-response pattern, soliciting more information without refining, challenging, or jointly developing ideas with the system. Longer conversations, the study emphasizes, were not automatically deeper ones: some multi-turn interactions stayed verification-oriented from start to finish. The authors are cautious here—the finding represents an absence of documented Interactive engagement, not proof that no unrecorded interactive exchange occurred, since many submitted records were incomplete. Still, the pattern is telling.

Why didn’t genuine dialogue with AI emerge? The instructional context offers clues. The task was convergent and well structured, with a known method and a single correct answer—conditions that naturally cue answer-seeking rather than exploration. Students received no explicit guidance on how to prompt the system dialogically, no models of iterative questioning, and no assessment incentives rewarding sensemaking; the assignment was graded for completion. Under those conditions, there was little reason for students to adopt an interactive stance. The authors argue that this absence should be read not as a failure of students or technology but as a design signal: interactive cognition with AI must be deliberately invited, scaffolded, and normed, and it will not appear simply because a conversational interface exists.

The broader implication is that “AI use” is not a meaningful analytic category in itself. Two students can produce identical, correct solutions—one by copying a prompt, the other by interrogating an explanation of shadow prices—and the cognitive consequences are worlds apart. The authors suggest that prompting should be understood as a cognitive and epistemic practice rather than a mere technical skill: a single copy-and-paste submission reflects a fundamentally different orientation toward learning than a sequence of targeted why, what if, and how would this change questions. For educators, the takeaway is concrete. If assessments reward only final answers, students will use AI passively or for verification. If tasks require explaining why a method works, comparing alternative approaches, or reasoning about how changed assumptions alter results, AI can be repositioned as a resource for reasoning rather than a shortcut around it. The educational value of generative AI, the study concludes, lies not in the tool itself but in the forms of thinking that instruction invites students to enact with it—and right now, in at least one typical quantitative course, those invitations mostly stop short of genuine dialogue.

Subject of Research: Cognitive engagement patterns of undergraduate students using generative AI for technical problem solving

Article Title: More than getting the answer: How students engage with generative AI in technical problem solving

Article References: Jaiswal, A., & Nanda, G. (2026). More than getting the answer: How students engage with generative AI in technical problem solving. International Journal of STEM Education, 13(1), Article 58. https://doi.org/10.1186/s40594-026-00646-7

Image Credits: AI Generated

DOI: 10.1186/s40594-026-00646-7

Keywords: generative AI, STEM education, cognitive engagement, ICAP framework, linear programming, ChatGPT, problem solving, sensitivity analysis, shadow prices, instructional design, higher education, than