A suitable place
Bring land rights, soil, water and present use into one site decision. A map helps us ask better questions.
How land is assessedA proposal for Caribbean food resilience
More food.
Closer to home.
From the Caribbean’s agricultural past to a practical next chapter for local food.
CaribVista starts with a question: how can land, farming knowledge and local organisation help more people eat well? Explore the histories behind the landscapes, then follow the proposal from research and practical training to a first locally led farm pilot.
Across the Caribbean15 country reviews
Explore the 15 country reviews
Generalised orientation; markers are not project sites. Map data: Natural Earth.
PurposeFood people can access
Research coverage15 programme countries
Current stageProject preparation
01Why this idea exists
Land has a history.
Food needs a future.
In Barbados, the sugar plantation economy reshaped land, work and trade through the labour of enslaved Africans. In Jamaica, emancipation also brought a struggle for land and the growth of independent small farming. The region’s food story includes both export estates and people growing for their own communities.
That history raises the question behind CaribVista: what would it take to support a new chapter of locally grown food, with farmers and households at its centre? The answer has to begin in a particular place, with the people and agricultural knowledge already there.
Historical context: Barbados’s heritage submission and the National Library of Jamaica.
BarbadosJamaicaGuyana
From a sugar landscape to a broader food future
Sugar reshaped Barbados's land, labour and trade. Alongside that plantation history runs another story: people growing provisions, keeping livestock and building livelihoods from the land.
Land, independence and the people who grow Jamaica's food
Jamaica's farming story includes plantation estates and the free villages built around emancipation. Access to land gave families room to grow food, earn a living and shape their own futures.
A food-producing coast, a future to share
Guyana’s coastal rice and sugar fields grew through generations of labour and enterprise. Beyond them, forests and savannas sustain different communities and food traditions.
02The intention
A farm is part of
a much bigger picture.
Food resilience depends on more than what grows. It depends on people who can farm it, water when it is needed, reliable buyers and a way for households to afford or receive the harvest.
The proposal is to connect those pieces in one locally led programme—and make the work, costs and results open to scrutiny.
Local capability
Put farmers, practical training and qualified field support at the centre of daily production.
How the programme worksAccountable delivery
Connect harvests to buyers and households. Track the costs, the food delivered and the decisions behind it.
What success would meanThe landscape, in context
Now look at the land.
After the country story, the imagery adds another view: what the landscape looks like and how its vegetation signal changes. Pair those observations with farmers’ knowledge, field visits and land records to investigate a possible site.
From the satellite record
See the land.
Read the signal.
Explore the same landscape in natural colour and through its vegetation signal.
Sentinel-2 · January–September 2026
RGB · Natural colourBarbadosCloud-screened median compositesNDVI colour scale
-0.2 · Lower signal0.9 · Higher signal
Stronger vegetation signals can include forests, pasture and crops. The dark background shows masked water, areas outside the study boundary or missing data.
Source granules99May overlap on the same date
Mean sampled NDVI0.546Separate 1,000 m screening grid
Image display grid≈ 28 mCountry overview, not parcel detail
A monitoring example · Barbados
Greenness changes through the months.
Each point is the spatial mean of that month’s median NDVI over observed grid cells. Different months can have different clear views.
NDVI axis: 0 to 1. Seasonal conditions, cloud screening and changes in observed cells can influence this line. It is not a measure of harvests, crop health or available land.
Monthly values and observation coverage
| 2026 month | Mean NDVI | Observed cells | Grid-cell coverage | Source granules |
|---|---|---|---|---|
| Jan | 0.621 | 412 / 416 | 99.0% | 10 |
| Feb | 0.590 | 413 / 416 | 99.3% | 11 |
| Mar | 0.539 | 410 / 416 | 98.6% | 9 |
| Apr | 0.462 | 397 / 416 | 95.4% | 7 |
| May | 0.412 | 415 / 416 | 99.8% | 14 |
| Jun | 0.344 | 342 / 416 | 82.2% | 11 |
| Jul | 0.537 | 408 / 416 | 98.1% | 13 |
| Aug | 0.529 | 416 / 416 | 100.0% | 12 |
| Sept | 0.597 | 416 / 416 | 100.0% | 12 |
Coverage refers to cells on this fixed grid, not national area coverage or classification accuracy. Monthly medians cannot be averaged to reproduce the full-period median. The monthly grid is pinned separately from the archived screening calculation.
Download the monthly values and method ↓How to read these images and inspect the sources
Natural colour combines the red, green and blue bands (B4, B3, B2). NDVI compares near-infrared and red reflectance: (B8 − B4) / (B8 + B4). Both views use the same dated scene list and extent, with cloud and shadow screening. Each pixel summarises observations across the period; this is not a single photograph.
The colour scale is fixed across countries; values beyond its endpoints use the endpoint colours. The separate NDVI statistic is the spatial mean of the temporal-median index on the documented 1 km screening grid. Image pixels use the display grid stated above.
These views help guide a closer inspection. They do not identify available rental land, confirm ownership or measure agricultural suitability.
- A temporal median of observations in January–September 2026, not a single-date photograph.
- RGB and NDVI use identical bounds, display grid and valid-pixel masks. Higher NDVI indicates greater spectral greenness, not crop identity, yield, land vacancy or suitability.
- Transparent pixels represent the selected boundary/land mask or missing valid observations, not proof of absent vegetation.
- The overview grid is for display. The published numerical screening uses a separate 1 km grid; do not extract hectares or availability from these coloured pixels.
- Cloud and shadow masking can leave residual haze or remove valid surfaces. Native 10 m source bands do not make the overview a 10 m product.
Download the imagery reproduction package ↓
Contains modified Copernicus Sentinel data (2026), processed for CaribVista.
The next step is to connect an observed landscape to the soil, water, rights and people a workable farm would depend on.
From observations to a site decisionFrom a regional idea to a local test
Begin with one place.
Learn before growing.
A first pilot should be sized around real land, water, skills and demand. Barbados is an initial preparation candidate; its site and scale remain to be established.
Read the pilot approach- 01
Agree the place and the people
A willing local partner, clear land rights and a team responsible for daily delivery.
- 02
Make a workable farm plan
Site tests, crop choices, buyers and a budget built from current evidence.
- 03
Measure what reaches people
Harvests, paid work, cost per kilogram, food access and environmental outcomes.
- 04
Decide what deserves to grow
Use the results to expand, revise or stop—not to assume a regional rollout.
03Across the region
One shared method.
Fifteen local contexts.
Follow each country’s agricultural history, the pressures and possibilities facing its farmers, and the place CaribVista could have in that story. Then explore the land research and practical next steps.
Find a countryThe next conversation
Help define the first pilot.
Bring a local delivery team, a candidate site or a route to customers and households. Together, the next task is to agree the preparation work, its responsibilities and its budget.
Discuss a preparation partnershipAn invitation to prepare a project. Sites, funding and operating partnerships are still to be agreed.