For journals, later
Later: the missed-literature check on every submission.
Right now unkwn42 is a check authors run on their own paper. One day the same check could run on every submission a journal gets, the way plagiarism checks do. I'm not selling that yet. If you edit a climate or earth science journal, I can show you how it would work on a private sample of papers you've already published.
Things slip through at the borders
Editors know their own field. Give an ocean scientist a manuscript on ocean heat and they'll know the key papers before the end of the abstract.
Overlap gets missed at the edges, in an ecology paper that reported the same effect a decade ago under another name, or a hydrology method the authors have reinvented without knowing. Nobody on a board reads every neighbouring literature, so the overlap surfaces late, from a sharp referee in round two or a reader after publication.
Start with a private sample
We draw a random sample of your journal's recent published papers, with the seed recorded so anyone can repeat the draw. For each paper, the engine searches the corpus as it stood the day before the paper came out, leaves out everything the paper cites and its authors' own work, and puts papers found under another name or from another subfield first. Every flagged item is checked against both papers' abstracts, only the clearly relevant ones count, and you get the results privately, with the method and its limits.
The sample uses only published papers, so no submission data changes hands. If it's useful, we can talk about a pilot on live submissions, which would need a data processing agreement before it starts. The editor decides what, if anything, to do with each item.
Real results
We have no journal sample results to show yet. These are unedited results from the same engine, on inputs I wrote for this page. Every paper named is a public paper in our corpus, and each title links to it.
Reviewer 2 · the paper a referee would citerun 20 October 2026
Abstract in
We examine the largest winter floods in California's Russian River basin since 1997. Most of them coincide with narrow plumes of water vapour from the subtropical Pacific striking the coast and being lifted over the coastal mountains, which produces several days of heavy orographic rain. We argue that forecasting these plumes a few days ahead could improve flood warnings in the basin.
Prior-work defence brief
Ralph2006 already established the causal link between atmospheric rivers and the largest Russian River floods since 1997, which is the primary empirical claim of the abstract.
F. Martin Ralph et al. · 2006 · Geophysical Research Letters
Attack
This paper explicitly examines the largest floods on California's Russian River since October 1997 and establishes that all seven events were caused by landfalling atmospheric rivers producing heavy orographic rainfall. It directly validates the abstract's core observation that narrow plumes of water vapor striking the coast and lifting over mountains are the mechanism for these specific historical floods.
Defence
The authors can distinguish their work by clarifying that while Ralph et al. (2006) documented the retrospective occurrence and physical mechanism of these past floods, the current study focuses specifically on the operational utility of forecasting these plumes days in advance to improve warning systems, a predictive application not addressed in the observational case study.
F. Martin Ralph et al. · 2012 · Journal of Hydrometeorology
Attack
This study utilizes unique hourly observations near the Russian River to quantify how the duration of atmospheric river conditions modulates streamflow and runoff volume, noting that longer durations produce significantly higher peak flows. It already links the characteristics of these vapor plumes to flood magnitude in the exact same basin using similar observational data from the region.
Defence
The differentiation lies in the focus: Ralph et al. (2012) analyzed the relationship between storm duration and resulting hydrologic impacts to understand variance in runoff, whereas the current abstract proposes a specific forecasting framework to predict the arrival of these plumes days ahead to enhance early warning capabilities rather than just characterizing their impact after the fact.
The abstract never says "atmospheric river", the name the field uses for these plumes. Reviewer 2 still put the 2006 paper that made the same finding for the same river at the top. Nothing is left out of this brief.
Terminal · a claim already maderun 19 October 2026
Idea in
Marine heatwaves have become longer and more frequent across the global ocean over the past century.
Already claimedgrounding verified
read as: Marine heatwaves have become longer and more frequent across the global ocean over the past century.
Eric C. J. Oliver et al. · 2018 · Nature Communications
Using a range of ocean temperature data including global records... we identify significant increases in marine heatwaves over the past century. We find that from 1925 to 2016, global average marine heatwave frequency and duration increased by 34% and 17%, respectively...
Claimed, with the paper that already makes the claim and the passage where it does. The same result also listed seven adjacent studies of shorter records or single regions, left out here to keep it short.
Where it works, and where it doesn't
Coverage varies a lot by journal. Of the papers published in the last five years, we hold 86% of The Cryosphere and 68% of Journal of Climate, but 16% of Environmental Research Letters and 8% of PLOS Climate. Nothing published after June 2026 is in yet. Before running a sample, we measure how much of the work your journal's papers cite is in the corpus, and tell you if it's too little to be worth doing.
Journals we hold most of
- The Cryosphere86%
- International Journal of Climatology72%
- Ocean Science72%
- Weather and Climate Dynamics69%
- Journal of Climate68%
- Atmospheric Chemistry and Physics49%
Journals we hold less of
- Biogeosciences30%
- Environmental Research Letters16%
- Frontiers in Climate14%
- Climate of the Past10%
- PLOS Climate8%
The share of each journal's papers from the last five years that we hold, measured on 13 October 2026 against a random sample of 200 from OpenAlex. Each figure is a few points either way.
Questions
Our editors already know the field. What does this add?
Inside their field, very little. Its use is in work from neighbouring fields and older vocabulary, which is where any specialist has blind spots.
Does it work in Editorial Manager or ScholarOne?
No. A pilot would run alongside your system, with the list sent to you for each submission. Integration would come after a pilot shows it's worth having.
What about confidential submissions?
The private sample uses only published papers, so no data agreement is needed for it. Live submissions would need a data processing agreement before a pilot starts, and manuscripts are never used to train models.
How is this different from integrity tools?
Integrity tools look for misconduct, such as paper mills and image manipulation. This looks for honest overlap, meaning related work the authors missed. The two do different jobs.
Have you run this on a journal yet?
Not yet, so we quote no figures here. When we run a sample for you, the results go to you alone.
Ask for a private sample
Tell us the journal and the person to send the results to. I reply to every message myself. Or email hello@unkwn42.com.