Shared costs

Gold in Sulawesi Utara

6 sites share the same problems (acid mine drainage, cassava land, mercury, heavy metals). If they carry the solutions together, the clean-up costs Rp 276 bn over 10 years instead of Rp 306 bn: 9.9% less.

TogetherRp 276 bnover 10 years

Alone, summedRp 306 bnover 10 years

SavedRp 30.3 bn9.9%

Solutions shared66 sites

No site pays more than aloneIndicative: most unit prices are working assumptionsModel cluster.v1 · Computed 23 Oct 2026

The sites and what connects them

Circle size is the hectares to restore. Colour shows whether the land lies inside a mining permit.

Inside a permit
No permit
Farmers who supply a factory
Catchment
River
Main river of the catchment
POCO route from Pelabuhan Bitung
Depot

Tap a site on the map or in the list for its details.

What they share

Cost over 10 years, Rp million. Each group is made of sites that can really share the solution.

SolutionSitesAlone, summedTogetherSaving

POCO plus tailored microbes on every mined cell of every site, driven from site to site on a planned route.Working assumption6253,752239,136−6%

One microbe mix designed per goal (for example gold plus mercury) and tested in one lab, instead of one design and one courier run per site.Working assumption610,2435,811−43%

Drone flights, satellite checks and plots, with one team driving past all sites.68,3114,196−50%

One project document, validation and verification for several sites.36,7232,707−60%

A factory that turns cassava roots into flour indoors, within 48 hours of harvest.224,93222,607−9%

One nursery for the seedlings of all sites in half a day's drive, instead of buying or building one each.32,0821,289−38%

POCO and microbes go on every mined cell of every site: there is no separate POCO zone.

ESMI unit rent, capacity and price tiers are working assumptions until the ESMI price list is published.

Data grade: A measured or official, B derived from a public source, C estimate with a source, D working assumption.

From separate bills to one shared bill

Each bar is what one shared solution saves, Rp million over 10 years.

The axis starts at Rp 252 bn, so the savings are readable.

Amounts per site and per party are for members

The cluster, its problems, the solutions and the total saving are public. The bill per site and the split over the parties need a verified account, because they name what each site carries.

The parties in the split

Donors and fundsClean-up of land with no responsible party, paid from money ESMI raises externally.
CooperativeLabour in kind, and income-earning assets such as the MOCAF factory.
VillageVillage fund and cleaner water.
LandownersLand in kind, and a share of carbon income.
Permit holder (legal duty)The part of the site inside its permit. The legal reclamation duty stays with the holder, who has its own bill.
Government (Pemda)The government leads and is the single point of contact, but does not pay. The share that used to sit with the government is now carried by donors; ESMI raises that money externally.
ESMIESMI's discount on POCO and the lab (grows with the group) and its share as carbon developer.

The polluter comes first: a permit holder carries the share of the site inside its permit, in full and on its own bill. It does not move to others.

How it gets funded

The full clean-up budget of this cluster is Rp 276 bn over 10 years (Rp 306 bn if every site did it alone), for at least 1,454 ha mapped; hectares found in the field are added.

ESMI raises the money externally: donors and international funds for land with no responsible party, and impact finance repaid from what the cleaned land yields for permit land. The reclamation guarantee of the holder is the backstop.

The legal duty stays where the law puts it: with the permit holder for land inside its permit, and with the government for land without a responsible party. The government leads and is the single door; it is not the payer.

Carbon: alone or together

Credits are a scenario per site, not a promise. Only land that is additional counts: the part inside a permit is a legal duty and earns no credits.

Amounts per site and per party are for members

Is the split fair?

The share of each site in one solution, under six ways of splitting. The platform proposes Shapley and checks that nobody and no sub-group pays more than alone.

6 of 6 groups are stable under the proposed split.

Amounts per site are shown to members.

The group council chooses the method. Shapley is the default because it is the easiest to explain and the only one that treats equal sites equally.

Joining and leaving

Every site decides on its own whether to join, through the vote in Shared decisions. A group council with one delegate per site decides about the group.

Nobody already in may pay more because of a newcomer. If that would happen, the newcomer pays the difference on top of its share, or the council refuses.

Joining and leaving happen at fixed moments: the next verification for carbon, the next route planning for the POCO unit (every quarter), every quarter for the rest.

Legal form (proposal, to be confirmed with legal advice): villages through a BKAD, cooperatives through a secondary cooperative, permit holders and the government through a cooperation agreement.

Data and what we do not know

Sources: footprints Cheng et al. 2026; permits ESDM Minerba register; rivers and catchments HydroSHEDS (Lehner and Grill 2013); NO₂ Sentinel-5P; plants Global Energy Monitor (CC BY 4.0); prices: working assumptions in model cluster.v1.

Distances are straight lines times 1.4 (a road factor, grade D). Real routing over the road network comes later.
Catchments come from a 90 m model. For small coastal rivers the border can be off by a few hundred metres.
Mercury is measured on few sites. Splitting by pollution load is therefore not reliable where it is missing.
Most unit prices are grade D and have no quotes. The structure and the proportions are more reliable than the absolute amounts.

← All clusters