Solar Energy for Agribusiness in Colombia

A grid-connected solar system produces during the day. Its return depends mainly on how much of that energy is consumed at the same instant, because that is what avoids paying the full tariff; the surplus is fed to the grid and compensated under article 25 of Resolución CREG 174 de 2021, amended by Resolución CREG 101 072 de 2025, at a value that need not equal that tariff.

Most agribusiness operations have exactly the consumption profile solar rewards: irrigation and pumping run by day, refrigeration and the cold chain operate steadily, and drying and post-harvest concentrate in sunlight hours. It is the opposite of a home, which empties out just when generation peaks.

To that timing factor add two physical advantages the countryside has and the city does not: large roofs on warehouses, sheds and processing plants, and land available for ground-mounted arrays when the roof falls short. Less space constraint means more freedom to size the system by consumption, which is how it should be done, and not by the square metres that happen to fit.

A system's output depends on the site's irradiation, and Colombia is not uniform. According to the IDEAM Solar Radiation Atlas, regions such as the Caribbean, La Guajira and part of the Eastern Plains are among the highest-irradiation in the country, while Andean and cloud-heavy zones yield less.

For an agribusiness project this is no detail: much of the country's banana, palm, rice, cattle and export production happens precisely in the regions with the best solar resource. The same installed capacity produces more on a coastal farm than on a highland one, and that difference passes straight through to the savings.

The practical consequence is that the assessment must start from the site's real irradiation, with its source stated, not from a national average. A proposal that does not say which irradiation it used is hiding the assumption that most moves the result.

Not every agribusiness load behaves the same toward solar. It is worth looking at them separately, because the fit depends on when they consume.

Ley 1715 de 2014, amended by Ley 2099 de 2021, sets out four mechanisms for generation projects using non-conventional renewable energy sources, and Decreto 829 de 2020 regulates the procedure over a project certified with UPME.

For a formal agribusiness with taxable income to apply them against, these incentives noticeably change the after-tax cost. But they carry a requirement worth stating plainly: the income-tax deduction and accelerated depreciation are only worth something if there is enough tax base, and everything runs through certifying the project with UPME. It is not a discount the installer applies to the invoice.

That is why one should be wary of a proposal that presents the incentives already netted off the price as if they were certain. They are not; they depend on the year's tax situation and on a procedure with its own calendar, and presenting them as automatic distorts the comparison between quotes.

The first particularity is the grid. Many rural farms and plants sit at the end of a weak line, or with no stable connection at all. That changes the design: where an urban industry installs a grid-tied system, a rural operation may need a hybrid system with backup, or even an off-grid one, to sustain critical loads such as the cold chain during an outage. That is more equipment and more investment, and it must be assessed from the start, not discovered later.

The second is seasonality. A crop with a concentrated harvest has a consumption peak a few weeks a year and little the rest. Sizing for that peak produces an oversized system that exports too much the rest of the time, and the surplus is compensated below the tariff. The right reference is the sustained yearly consumption, not the campaign peak.

The third is the environment. Dust from field work, particulate from milling and proximity to agrochemicals soil the modules faster than an urban roof, and some coastal zones add salt corrosion. That means more frequent cleaning and components suited to the setting: a real operating cost the proposal must acknowledge.

The fourth, as in any business project, is ownership of the premises and the horizon of stay. On leased or sharecropped land, who invests and who keeps the asset must be agreed in writing before anything is mounted.

When there is plenty of land and roof, the temptation to install big appears. In farming a second trap is added: sizing for the busiest season of a seasonal crop.

Both lead to the same result: a system that generates too much in the hours and months of lower consumption, exports that surplus and sees it compensated at a value that does not equal the avoided tariff. The investment rises with size; the savings do not, in the same proportion.

Correct sizing starts from real consumption data —ideally the retailer's hourly curve and, failing that, twelve months of bills that capture the full seasonality— and looks for the point where self-consumption is maximised without a systematic surplus. If a proposal arrives with a capacity it cannot explain, that is the first question to ask.

An agribusiness operation about to invest needs a proposal that can be audited, not a loose savings figure. Asking for the right things up front saves a whole review cycle.