Climate-risk intelligence, built for African finance.
ecosphere gives African banks, insurers, asset managers, development finance institutions, corporates and regulators the climate evidence they need to lend, underwrite, invest and supervise, with every figure traceable to its source.
One evidence base, from hazard to capital.
ecosphere is a climate-risk platform. It brings together observed hazard history from ERA5 reanalysis, CMIP6 climate projections and NGFS transition scenarios, aggregates them to every African country, admin-1 region and municipality, and models hazard at individual sites on bare-earth terrain.
It then carries those hazards through to the measures financial institutions already manage: carbon cost and revenue impact per counterparty, probability of default, expected credit loss and CET1 capital. Results are delivered in the web console, as Parquet data files and through an API, in each organisation’s reporting currency.
African exposures deserve African detail.
Climate risk is local. A loan book in Kenya, Nigeria or South Africa is exposed to the rains, heat and terrain of specific regions and sites, and its losses are counted in shillings, naira and rand.
Places you report on
Figures are aggregated to every country, admin-1 region and municipality on the continent, the units in which lending and supervision actually happen, rather than stopping at a national average.
Ground, not rooftops
Flood and landslide are modelled on bare-earth terrain, choosing LiDAR, our own DTM pipeline, FABDEM or Copernicus DSM in that order, because water follows the ground and not the tops of buildings and trees.
Numbers in your currency
Each organisation reports in its own currency, such as ZAR, KES or NGN, so stress results land in the same units as the balance sheet they are meant to inform.
Four rules every figure follows.
These are not slogans. They decide how the console draws a chart, how a data file is structured and what a report is allowed to say.
-
Provenance is a primitive.
Where a number came from is part of the number. Every figure carries its source, period and version, every delivery passes a published data contract and a second-person review, and synthetic or stand-in data is always labelled as such.
IllustrativeFlood depth 1-in-100, grain store1.1 msrc2D flood dtmFABDEM verv2026.10 reviewsecond person -
Uncertainty is shown, not hidden.
A single decimal place implies a precision climate science does not have. Central values come with ranges, projections with their ensemble spread, and incomplete years are flagged rather than quietly compared with full ones.
Central value, with its rangeIllustrative -
Two dimensions, never one blended score.
Physical risk and transition risk have different causes, horizons and remedies. Folding them into a single score hides which one you are facing, so ecosphere always reports them as two separate dimensions.
Same blended score, opposite risksIllustrative -
Comparison first.
No single scenario is the future. Scenarios, horizons and runs are laid side by side, so the question is always “how much worse under this path than that one?” rather than one number taken as a forecast.
CET1 impact, bps, 2030Illustrative
Start small, on your own book.
We would rather prove the figures on a sample of your exposures than ask you to take them on trust.
-
Pilot
We run a sample of your counterparties and sites through physical shocks and transition scenarios, and walk your team through every link from hazard to capital.
- You bring
- sample book
- You get
- stress result · report
-
Integrate
Connect the full book through uploads or the API, take Parquet files into your own models, and keep your organisation’s data isolated from everyone else’s.
- Delivery
- console · Parquet · API
- Isolation
- per organisation
-
Report
Run official stress tests alongside exploratory ones, export scenario reports for committees and supervisors, and ask the assistant questions of your own published data.
- Runs
- official · exploratory
- Outputs
- scenario report
Read how the figures are made on Data & methods, or see the questions we answer by sector.
Bring a sample portfolio. Leave with a stress result.
In a working session we run a sample of your counterparties through a physical shock and a transition scenario, and walk you through every step of the chain from hazard to capital.