{"id":"24e194ab-718c-46b9-a7c3-6af2f394f700","arxiv_id":"2505.07624","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"For a 100-hour, 42.5%-efficient storage technology, state-by-state model runs show viability costs from -$12 to $5,994/kW, with only Kansas, North Dakota, Nebraska, and Virginia passing the $1,100/kW DOE target in 2050.","lead":"Using a capacity expansion model, the authors compute the maximum price per kilowatt that 100-hour long-duration storage could cost and still pay off in each U.S. state in 2050. They find that at the U.S. Department of Energy's $1,100 per kilowatt target, only four states could replace gas and coal with storage without raising total system costs.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'only 4 states' result is not tested against the most consequential assumption: each state is isolated with no inter-state trade or new transmission (Sec. 4.6).","rationale":"The central claim is policy-relevant: it says the DOE's $1,100/kW target would justify full thermal replacement in only 4 states. This claim would be true only if the state-level viability costs are accurate. The single largest threat to accuracy is the island-state assumption in Sec. 4.6, because it changes the counterfactual replacement cost for many states. The reader's weakest assumption identifies exactly this, and I agree. The paper's internal consistency is not the issue; the issue is external validity. The model is coherent and the math seems plausible, but no sensitivity analysis or comparative test is provided. A multi-state run or import relaxation is the natural decisive check. If it moves the count, the paper's conclusion should be reframed as conditional on autarky. Given the absent test, CONDITIONAL remains the right verdict, so I recommend UNCHANGED.","tokens_in":81451,"tokens_out":5725,"duration_ms":60408,"concrete_test":"Re-run the opportunity value model for all 48 states with the Section 4.6 demand constraint relaxed so each state may import up to its projected 2050 net interchange (or run a joint multi-state model with the 134 BAs and existing transmission from Cambium/ReEDS). Then recount states with viability cost at or above $1,100/kW. If the set changes (e.g., NJ or DE becomes viable, or CA/CO crosses the threshold), the 4-state claim depends on the island assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 4.6 imposes that 'the combined generating capacity of all BAs within a state must meet the state’s demand,' with no inter-state electricity trade and no new transmission investment. This isolation assumption is structurally important: to retire gas and coal, a resource-poor state must build local intermittent capacity at high cost, lowering the avoided cost and hence the computed LDES viability cost. In reality, such states import clean power from neighboring regions, so their true cost of replacing thermal is lower and their viability cost higher. The five states with negative viability costs (AL, CT, DE, NJ, OH) are small or import-dependent systems, strongly suggesting the assumption is driving low values. The paper states the assumption but never tests it, so the headline 'only 4 states above $1,100/kW' is not robust to inter-state trade. The 4x/10x build caps and the single 100-h/42.5% storage configuration compound this but are secondary.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper estimates, for each of the 48 contiguous U.S. states in a 2050 scenario, the maximum capital cost (in $/kW) of a 100-hour, 42.5% round-trip-efficiency long-duration energy storage (LDES) system such that the state could retire all gas- and coal-fired generation and replace it with a mix of intermittent renewables, short-duration storage, and LDES without increasing total system cost. This 'viability cost' is computed as the ratio of the maximum opportunity value from a constrained optimization (Eqs. 2-3) to the installed LDES capacity. The headline result is that only Kansas, North Dakota, Nebraska, and Virginia have viability costs above the DOE's $1,100/kW target; five states have negative viability costs. The paper also reports aggregated LDES capacity requirements and correlates high viability costs with low thermal utilization, high wind shares, and higher thermal fixed O&M costs.","tokens_in":81572,"tokens_out":7379,"duration_ms":70300,"significance":"The paper provides a consistent, transparent method for state-level LDES valuation using public datasets, and its negative result about the DOE target, if robust, would be policy-relevant. The framework is reproducible in principle and the supplementary material is extensive. However, the headline ranking is conditioned on several strong modeling choices that are stated but not stress-tested, most notably the autarkic treatment of each state and the exclusion of all firm low-carbon alternatives other than gas/coal retirement. These choices could materially change the set of states above the target, so the central claim should be interpreted with caution until sensitivity analyses are provided.","major_comments":[{"comment":"The state isolation assumption is load-bearing for the headline claim. The model requires each state's balancing areas to meet the state's demand with no inter-state trade and no new transmission, which forces resource-poor states to build local intermittent capacity at high cost and lowers their computed avoided cost and hence their LDES viability cost. The five states with negative viability costs (AL, CT, DE, NJ, OH) are small or import-dependent systems, consistent with this mechanism. Since the paper's central result is that only four states clear the DOE target, the authors should either relax this assumption (e.g., group states into regions with trade) or provide a rigorous sensitivity analysis showing that the ranking is unchanged. Without that, the 'only 4 states' conclusion is not robust to a reasonable and realistic modification of the model.","section":"Section 4.6"},{"comment":"The opportunity-value scenario retires gas and coal and admits only intermittent generation, SDES, and LDES as replacements; nuclear, geothermal, and fossil with carbon capture are not candidate resources. The resulting viability cost is thus conditional on the exclusion of these alternatives, which in many states could provide firm capacity at lower cost than large amounts of local renewables plus storage. The paper should report at least a bounding analysis (e.g., adding a generic firm low-carbon resource at ATB cost) and discuss how the DOE-target comparison would change. As it stands, the policy conclusion that the target is sufficient in only four states is framed more strongly than the model supports.","section":"Sections 2.1 and 4.4"},{"comment":"The candidate renewable build limits of 4x (and 10x for CT, DE, PA) the existing installed capacity are ad hoc and could bind for states with limited land or already high renewable penetration. If these caps bind, the optimization cannot choose the least-cost replacement mix, and the implied viability costs are depressed. The authors do not report which states or technologies hit these caps, nor do they test the sensitivity of the state ranking to the multiplier. This should be documented and, ideally, relaxed in a sensitivity case.","section":"Section 4.7"}],"minor_comments":[{"comment":"The phrase 'baseline bodel' appears to be a typo for 'baseline model'.","section":"Section 4.2"},{"comment":"The variables C_over, q_over, c_VC, and x_power are introduced without definitions or units; the paper should include a notation table or refer explicitly to [7] with a short description so the metric is self-contained.","section":"Equations (2)-(3)"},{"comment":"The definition of 'minimum required LDES capacity' (646.09 GW) should be stated precisely; it is not clear whether this is the smallest capacity at which viability cost is non-negative or the capacity that maximizes viability cost.","section":"Section 2.2"},{"comment":"The supplementary figures appear to contain garbled text (sequences of 'uni0000' characters) in the provided version; the authors should verify the PDF rendering.","section":"Supplementary information"},{"comment":"The term 'IES' (intermittent energy sources) is used throughout but not defined at first use; please define all abbreviations on first appearance.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper's main claim is interesting and the method is sound, but I am concerned that the autarky assumption is a fundamental modeling choice that could invert the headline result. If the authors can add a regional-trade sensitivity or significantly moderate the conclusion, the paper could be suitable for publication. I would not reject outright, as the issue is testable within the paper's framework."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: this is a competent, clearly written extension of the authors' own California method to all 48 contiguous states, and the new output—per-state $/kW viability costs for 100h, 42.5% RTE LDES in 2050—is genuinely new and useful for policy conversation. The driver analysis (low thermal share, high wind share, higher thermal FO&M => higher viability cost) is plausible and supported by the table.\n\nThe modeling is internally coherent. The viability cost is the max avoided cost divided by LDES capacity, an optimization output compared to an external DOE target, so there's no circularity. The negative-cost states (AL, CT, DE, NJ, OH) are explained by weak renewables, not by a fitting artifact. The supplementary dashboard and per-state figures are a plus.\n\nThe soft spot that matters: Section 4.6 treats each state as an island with no inter-state trade or new transmission. The paper states this clearly, but never tests it. This is not a minor detail. States with poor local wind/solar must build expensive local intermittent capacity to retire gas and coal, which lowers the computed avoided cost and hence the viability cost. In reality those states would import clean power, which would raise their true viability cost. The five negative-cost states are small or import-dependent systems; that pattern strongly suggests the assumption is doing real work. So the headline 'only 4 states clear $1,100/kW' is conditional on an assumption that could plausibly change the ranking. The 4x/10x build caps and the single 100h/42.5% storage configuration are secondary but also need sensitivity analysis.\n\nThe reader's stress-test holds up on reading the paper; I don't see a way to dismiss it. That said, the central argument is not flawed—it's a conditional result that is honestly labeled. The paper would improve with a scenario allowing regional trade or at least a discussion of how imports would shift the threshold. No code or data files are provided, which limits reproducibility, though the public data sources are named.\n\nVerdict: this deserves peer review. I'd send it out. It's the kind of applied result a referee can push to a much stronger conditionally-accepted paper. For my own work I wouldn't cite the headline number without checking the sensitivity, but I'd bring it to reading group as a case study in how strong assumptions drive state-level storage results.","headline":"A solid, policy-relevant state-by-state LDES viability map whose headline 'only 4 states' claim hangs on an untested no-trade assumption and a single storage configuration.","tokens_in":82174,"tokens_out":1923,"would_cite":false,"duration_ms":20277,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Under the 2030 cost target of $1,100 per kilowatt, only four contiguous U.S. states would find 100-hour storage economical for replacing gas and coal in 2050.","keywords":["long-duration energy storage","viability cost","100-hour storage","capacity expansion","thermal generation retirement","state-wise analysis","energy storage economics","round-trip efficiency"],"falsifier":"Re-run the two models with cross-state transmission allowed at existing or planned transfer capacities; if the number of states clearing the $1,100/kW threshold exceeds four, the headline result is an artifact of the isolation assumption.","tokens_in":81152,"feed_emoji":"🔋","tokens_out":8096,"duration_ms":69434,"temperature":0.7,"pith_summary":"The paper tries to establish that the current U.S. cost target for multi-day storage is set too high to justify replacing gas and coal plants in most states. It computes a viability cost for each of the 48 contiguous states: the maximum price per kilowatt that a 100-hour, 42.5%-efficient storage system can carry and still keep total system cost no higher than a system that keeps its thermal plants. The results range from -$11.81/kW to $5,993.94/kW, and only Kansas, North Dakota, Nebraska, and Virginia clear the $1,100/kW target. The finding matters because it separates the cost question from the value question: even if storage technology reaches its target price, most states would not choose it for full thermal replacement under these assumptions.","feed_headline":"Long-duration storage pays off in just 4 U.S. states by 2050","feed_subtitle":"At the $1,100/kW target, Kansas, North Dakota, Nebraska and Virginia are the only states that clear it.","key_machinery":"The central object is the LDES viability cost, defined as the maximum avoided total system cost divided by the installed storage power capacity. It is computed by solving two optimization problems in sequence: a baseline model that minimizes total system cost with existing generators only, and an opportunity-value model that allows new intermittent capacity and batteries, retires gas and coal, fixes a candidate LDES capacity, and requires total cost to stay at or below the baseline. The largest per-kilowatt value found across the tested 0-150 GW capacity range is the state's maximum viability cost.","core_discovery":"For each state's 2050 system, the authors replace gas and coal generation with a mix of new wind, solar, 4-hour batteries, and a fixed amount of 100-hour storage, and then ask how much that storage could cost per kilowatt before the replacement stops being cost-neutral. They call this the viability cost. Across states it spans from -$11.81/kW to $5,993.94/kW, and the headline result is that only Kansas, North Dakota, Nebraska, and Virginia exceed the $1,100/kW multi-day storage target. Replacing thermal plants in the 43 states with positive viability costs would require at least 646.09 GW (64.61 TWh) of 100-hour storage, and 1,009.30 GW (100.93 TWh) at the capacities that maximize viability cost.","pith_inferences":["If states were allowed to trade electricity across their borders, the list of viable states would probably grow, because renewable-rich states could sell surplus power to resource-poor neighbours; the paper's isolation assumption is a plausible driver of the four-state result.","The gap between the paper's minimum 646 GW of 100-hour storage and the 225-460 GW national estimate suggests that wider demand projections assume storage plays roles other than full thermal replacement, or that full thermal replacement is too strict a benchmark for many states.","Negative-viability states could still find LDES worthwhile for services such as capacity adequacy or transmission deferral; the paper's value measure counts only avoided thermal-plant costs.","A direct test of robustness would vary round-trip efficiency and duration around the iron-air baseline (42.5% RTE, 100 hours) to see whether the set of four states changes when the storage technology's parameters move."],"forward_implications":["If the target falls to $500/kW, the viable list grows to nine states: California, Colorado, Idaho, Kansas, Montana, North Dakota, Nebraska, New Mexico, and Virginia.","At $300/kW, 17 states clear the bar, adding Maine, Minnesota, North Carolina, Nevada, Oklahoma, Texas, Washington, and West Virginia.","The minimum 100-hour storage needed for thermal replacement in the 43 states with positive viability costs is about 646.09 GW (64.61 TWh), and 1,009.30 GW (100.93 TWh) at the capacity that maximizes each state's viability cost.","States with high viability costs generally have low thermal participation, low thermal utilization, wind-dominated intermittent generation, and higher thermal fixed O&M costs; low-viability states generally rely on thermal, have solar-dominated intermittent mixes, and lower intermittent capacity factors.","Five states (Alabama, Connecticut, Delaware, New Jersey, and Ohio) have negative viability costs, meaning replacing gas and coal with intermittent generation plus storage raises total system cost at any storage price."],"supporting_citations":[{"why":"It supplies the baseline and opportunity-value optimization models used to compute viability costs.","marker":"[7]"},{"why":"It provides the 2050 state-level load and generator projections that seed both models.","marker":"[8]"},{"why":"It defines the $1,100/kW multi-day storage cost target and the 225-460 GW national LDES estimate used as comparison points.","marker":"[12]"},{"why":"It supplies investment and fixed O&M cost data for intermittent and thermal technologies.","marker":"[10]"},{"why":"It supplies the fuel price assumptions that determine how costly the replaced thermal generation is.","marker":"[11]"},{"why":"It provides the iron-air demonstration parameters behind the 42.5% round-trip efficiency assumption.","marker":"[13]"},{"why":"It gives the iron-air battery technology review that supports the same round-trip efficiency value.","marker":"[14]"},{"why":"It provides the ReEDS base scenario used for state-level fixed O&M and capacity-factor inputs.","marker":"[9]"},{"why":"It documents balancing area definitions and generator characteristics used to assemble each state's isolated system.","marker":"[17]"}],"fun_headline_variants":["Long-duration storage pays off in only 4 states","4 states pass $1,100/kW storage viability bar","Kansas, North Dakota, Nebraska, Virginia top storage viability","Wind-heavy states alone clear LDES cost target","Just 4 U.S. states find long-duration storage economic"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper models each state as an isolated grid with no electricity imports and no new transmission, so a state with poor local wind and solar cannot draw on clean power from outside; the four-state result depends on that isolation.","fun_headline_variants_meta":{"raw":{"variants":["Long-duration storage pays off in only 4 states","4 states pass $1,100/kW storage viability bar","Kansas, North Dakota, Nebraska, Virginia top storage viability","Wind-heavy states alone clear LDES cost target","Just 4 U.S. states find long-duration storage economic"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000201,"raw_usage":{"total_tokens":1350,"prompt_tokens":888,"completion_tokens":462,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":504,"completion_tokens_details":{"reasoning_tokens":379}},"tokens_in":504,"tokens_out":462,"duration_ms":4224,"temperature":1.0,"reasoning_tokens":379,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T22:13:19.240592+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the two models with cross-state transmission allowed at existing or planned transfer capacities; if the number of states clearing the $1,100/kW threshold exceeds four, the headline result is an artifact of the isolation assumption.","supporting_citations":[{"cited_title":"IEEE Transactions on Energy Markets, Policy and Regulation (2024) 15","cited_arxiv_id":null,"evidence_quote":"It supplies the baseline and opportunity-value optimization models used to compute viability costs."},{"cited_title":"Accessed on May 17, 2023","cited_arxiv_id":null,"evidence_quote":"It provides the 2050 state-level load and generator projections that seed both models."},{"cited_title":"Technical report, U.S","cited_arxiv_id":null,"evidence_quote":"It defines the $1,100/kW multi-day storage cost target and the 225-460 GW national LDES estimate used as comparison points."},{"cited_title":"Accessed on March 29, 2023","cited_arxiv_id":null,"evidence_quote":"It supplies investment and fixed O&M cost data for intermittent and thermal technologies."},{"cited_title":"Energy Information Administration: Annual Energy Outlook 2023","cited_arxiv_id":null,"evidence_quote":"It supplies the fuel price assumptions that determine how costly the replaced thermal generation is."},{"cited_title":"Accessed on March 7, 2025","cited_arxiv_id":null,"evidence_quote":"It provides the iron-air demonstration parameters behind the 42.5% round-trip efficiency assumption."},{"cited_title":"Walsh, F.: A review of the iron–air secondary battery for energy storage","cited_arxiv_id":null,"evidence_quote":"It gives the iron-air battery technology review that supports the same round-trip efficiency value."},{"cited_title":"Accessed on March 22, 2023","cited_arxiv_id":null,"evidence_quote":"It provides the ReEDS base scenario used for state-level fixed O&M and capacity-factor inputs."},{"cited_title":"Technical report, NREL (January 2023)","cited_arxiv_id":null,"evidence_quote":"It documents balancing area definitions and generator characteristics used to assemble each state's isolated system."}],"review_version":1}