How much carbon can regeneration sequester? (+ the AI hacks you asked for)

Last week's AI tools for COP30 were a hit. So we're keeping the momentum going.

This week, we're supporting your climate work in two ways:

For your next project: We built a Regeneration Atlas that calculates carbon sequestration potential for any location—because showing stakeholders real numbers is easier with the right tool.

For your daily workflow: After the response to our AI tools last week, here are 5 prompting hacks that AI superusers (us, the ClimateAligned team) swear by, to get better results from ChatGPT, Claude, or whatever you're using.

How much carbon does land regeneration actually sequester?

Try our Regeneration Atlas

In honour of COP30, we remixed our energy project impact calculator to focus on regeneration across different biomes.

Using IPCC, FAO, and WRI data, we built a custom dataset that estimates carbon sequestration potential for every country's major ecological zones over 20 years.

How it works: Enter any location and land area— our Regeneration Atlas, calculates how much carbon you'd sequester through active regeneration.

Example: Regenerating an area the size of the City of London (290 hectares):

• Amazon rainforest (Brazil): 60,944 CO2e sequestered (equivalent to 13,249 cars off the road)
• Temperate oceanic forest (UK): 27,781 CO2e sequestered (equivalent to 6,039 cars off the road)

Geography matters—the same land area in the Amazon sequesters roughly twice as much carbon as in London.

Try it here →

AI Wizard: 5 AI Prompt Hacks

We're a small team using AI to speed up our work and increase output. AI is far from perfect—but there are ways to make your results significantly better.

Here are 5 techniques we use daily:

1. Ask it to check its work

When solving problems with right/wrong answers, get your answer then immediately ask: "Can you check that for errors?" LLMs catch their own mistakes surprisingly well.

2. Start a new chat when you're stuck

If you're going in circles, don't keep arguing. LLMs are prediction machines that get confused by messy context. Fresh chat = fresh start.

3. Kill the "ChatGPT voice"

Make a list of your AI pet peeves (mine: emojis, excessive bold, "delve," "unlock"). Explicitly ban them in your prompt. Even better: save it to Custom Instructions or memory.

4. Ask for 3 options instead of 1

Not getting what you want? Stop trying to perfect a single answer. Ask: "Give me 3 different approaches to this" and pick the best direction.

5. Shortcut spreadsheet tagging

Need to label/categorise spreadsheet rows? Copy 30 rows at a time, tell the AI exactly what labels you want, and ask for "just a plain text list, no explanations." Paste the results straight back into your sheet.

(This is how we tagged biome names for ~500 biomes across 195+ countries in the Regeneration Atlas—what would've taken days took ~30 minutes.)

Before we go - special thanks to Fintech Global for including us in 2025's ESG FinTech 100. We're honoured to be included alongside the most innovative companies transforming the ESG industry.

See you next Wednesday!

Natasha and the ClimateAligned team

Explore our products and free tools:

app.climatealigned.co - Access our labelled bond data service with insights on 30,000+ bonds. Check out a free sample!

brief.green - Run AI-powered ESG assessments on-demand. Sign up for free!

AI footprint calculator - Estimate the environmental impact of your AI usage. Try it now!

Project impact web app - Estimate the environmental impact of sustainability projects. Try it now!

Report publication patterns - Analyse corporate reporting seasons by region from our extensive dataset. Try it now!

Regeneration Atlas - Calculate the climate impact of land regeneration worldwide. Try it now!

Want to learn more about what we do? Visit climatealigned.co

Want to automate a workflow you find annoying and slow? We'd love to hear more! Fill out our Typeform.