§01 · The brief, in one sentence
My father planted the first coffee on this ground in 1971. The trees are still there. We are not replanting on instinct. We ran a soil sample, two weather years, and a long argument with an agronomist before we took out the first stump. This post is the case for the ten acres under Morceau, written before the first cherry came off.
§02 · The coordinates, and why elevation isn't a cliché here
0.1542°S, 35.1997°E. Ten acres in Muhoroni, Kisumu County, on the eastern shoulder of the Lake Victoria basin. Equator. 1,200 meters above sea level. Bimodal rainfall: long rains March through May, short rains October through November.
The math on elevation is simple enough to be repeated badly. Higher means slower bean development, denser cellular structure, more sugar concentration in the seed. The cliché stops at 1,200m on this part of the equator because the day length doesn't change with the seasons. There is no winter to slow ripening for you. Slow comes from elevation alone. 1,200m is the ledge where it kicks in here. Below 1,000m we'd grow a perfectly fine commodity arabica. We would not have an 88+ cup.
The other thing the equator gives you is consistency. Daylight in Muhoroni varies by about 18 minutes across the year. A pollinated flower in March follows the same ripening clock as one pollinated in October. The protocol stops being a moving target.
We did not pick this ground. We inherited it, ran the numbers, and decided not to leave.From the mill log, day 01
§03 · What the soil panel actually said
We pulled three independent core samples from across the ten acres in March 2024 and sent them to Egerton University for analysis. The summary, unedited from the report:
- pH: 6.1 to 6.4 (target for arabica is 5.5 to 6.5; we sit at the upper boundary)
- Organic matter: 4.8 percent (good; coffee likes 3 to 6)
- CEC, cation exchange capacity: 18.4 meq per 100g (high, good buffering)
- Texture: clay-loam, well-drained on the slope, slightly heavier in the bottom three rows
- No glyphosate or chlorothalonil residue (we'd never expected any; previous use was subsistence maize)
The pH at the upper boundary is the reason we can grow a clean cup without aggressive amendment. Most Kenyan farms in the lower coffee belt pull pH down with sulfur. We don't. The CEC is high enough that the soil holds onto nutrients you put in instead of letting them leach, which means the compost program we run pays back instead of running off into the watershed.
§04 · The organic thesis, in plain language
No synthetic nitrogen, no synthetic herbicide, no synthetic fungicide. Native shade canopy at thirty-eight percent cover. Cover crops between rows on a two-year rotation. Aerobic compost from coffee pulp, green waste, and bovine manure, all of it sourced within four kilometers. Hand-turned drying beds. No till.
The trade is real and worth naming. Yields per acre are below the chemical-input average for the region by something like fifteen percent in year three, narrowing year over year as the soil builds. We accept that. The cup is what we sell, and the cup off this protocol is a different beverage. Roasters who've tasted both don't need a pitch.
Three decisions we made up front, written down so we can be held to them:
Most farms in the lower coffee belt amend with sulfur to bring pH down. Our soil sat at 6.1 to 6.4, the upper boundary of the arabica window. Amending would have squeezed out a bit more cherry weight; not amending leaves the microbiome intact, and that compounds across years. The cost of not amending is one less line item on the spend sheet and a gentler soil. We took the gentler soil.
The bottom of the slope tested heavier: more clay, slower drainage. Affordable, available, contiguous. We passed. Heavier soils favor a different cup profile (more body, less acidity) and we are not building two lots. The ten acres we kept share a single clay-loam profile end to end, which means the cup is consistent and the picking schedule is coherent.
First panel was Egerton University. Second was a private lab in Nairobi we paid double for. The two reports agreed within rounding. We knew before planting that the ground was not lying to us. For ten acres of capital that's cheap insurance.
§05 · Why this acreage, and no other
Family land, three generations. Water rights at the upper boundary so the entire estate can be irrigated by gravity in a dry shoulder. Forty minutes by truck to a wet mill cooperative we use for the back half of processing. A picker labor pool of roughly two hundred households within a four-kilometer walk, which is the difference between hiring on the day and waiting on the second harvest.
None of these are accidents and none are romantic. We did not pick this acreage; we inherited it and made the math work. If the soil panel had come back at 5.2 pH or the CEC at single digits we'd have leased it out and looked elsewhere. It came back the way it did, and we stayed.