Beneath the glass and polyethylene that blanket entire Dutch provinces and Kenyan lakeshores, an industry has built its own climate — warm, humid, precisely controlled, and increasingly at odds with the one outside. The story of how we grow flowers in winter is also, inescapably, the story of what that costs the planet.
Drive south from Amsterdam on the A4 motorway on a grey November morning and, somewhere around the town of Naaldwijk, the landscape begins to glow. It is not the sun — there is no sun. It is the Westland greenhouse district, one of the most intensively cultivated agricultural zones on earth, radiating artificial light into the overcast Dutch sky. Some 80 square kilometres of glass cover this corner of the Netherlands. Viewed from the air at night, the cluster of greenhouses pulses with a sodium-orange light that outshines nearby cities. Seen from ground level, the effect is stranger still: an unbroken plain of glass, stretching to the horizon, sweating in the cold air, enclosing a perpetual, constructed summer.
Inside, roses are in full bloom. So are gerberas, chrysanthemums, and a dozen other species that have no business flowering in the Northern European winter. The temperature is 18 degrees Celsius. Outside, it is four. Between those two numbers lies one of the most consequential and underexamined environmental questions in modern agriculture: what does it cost — in energy, in carbon, in ecological disruption — to grow a flower out of season? And as the world warms around them, what happens to the greenhouses that helped warm it?
Glass and Heat: The Original Forcing Technology
The greenhouse is not a modern invention. The Romans grew cucumbers under sheets of transparent stone called lapis specularis; Renaissance Italian botanists built the first giardini botanici with heated shelters for exotic plants. But the industrial greenhouse — vast, systematically heated, dedicated to commercial production — is a product of the nineteenth and twentieth centuries, and nowhere did it take root more completely than in the Netherlands.
The Dutch advantage was partly geographical: a flat, low-lying delta landscape with easy water access and proximity to major European markets. But it was also cultural and institutional — a centuries-old tradition of horticultural precision, combined with post-war investment in research and infrastructure, that turned a small nation into the world’s largest flower exporter. Today the Netherlands is responsible for roughly 65 percent of global flower exports by value, almost all of it grown or transshipped through facilities that depend on artificial heat.
The basic physics of a greenhouse is simple. Glass (or modern polycarbonate and polyethylene films) transmits shortwave solar radiation but blocks the longwave thermal radiation that would otherwise allow heat to escape. The sun warms the interior; the casing traps that warmth. On a summer day, this is so effective that ventilation is the primary challenge. In a northern winter, however, solar input is insufficient to maintain the temperatures most flowering plants require, and the gap must be filled by active heating.
In the Westland district and the larger Greenport Holland cluster, that heating has historically come from natural gas. Enormous combined heat and power (CHP) units — gas engines that simultaneously generate electricity and capture waste heat — feed warmth into the greenhouse through networks of hot-water pipes running along the floor and sometimes through elevated rail systems. The waste CO₂ from combustion is, in many modern facilities, piped directly into the greenhouse atmosphere, where elevated carbon dioxide levels accelerate plant growth. The Dutch system, in other words, has been engineered to extract maximum utility from every joule of combustion: the heat warms the air, the CO₂ feeds the plants, the electricity either runs the facility or is sold back to the grid.
It is an impressively integrated system. It is also, by any reasonable measure, a significant source of greenhouse gas emissions.
The Carbon Bill
The Dutch horticultural sector consumed approximately 120 petajoules of natural gas annually in the years before the energy crisis of 2021–2022 — a figure comparable to the total energy consumption of a medium-sized European country. Flower and ornamental plant growing accounts for a substantial share of this, alongside vegetable production. Converting that gas consumption to carbon dioxide equivalent emissions produces numbers that dwarf those generated by the air freight that gets cut flowers from equatorial farms to European consumers.
This is the central paradox of the “local versus imported” debate in cut flowers. Environmental advocates have long questioned the sustainability of flying roses from Kenya or carnations from Colombia to European markets. But the comparative life-cycle analyses tell a more complicated story. A study published in the International Journal of Life Cycle Assessment found that a rose grown in a heated Dutch greenhouse and sold in the Netherlands had a carbon footprint roughly three to four times higher than a rose grown in Kenya and flown to Amsterdam. The fuel burned heating glass against a Dutch winter dwarfs the jet fuel burned transporting flowers from the equator.
“People have a very strong intuition that local equals sustainable,” says one environmental economist who has studied agricultural supply chains extensively. “With flowers, that intuition is almost entirely wrong. The worst thing you can do, from a carbon perspective, is heat a greenhouse in Northern Europe to grow something that would grow perfectly well outdoors in East Africa.”
The situation is not static. Dutch growers, facing both regulatory pressure and the brutal economics of the 2021–2022 energy crisis — when natural gas prices increased tenfold and many greenhouse operators faced bankruptcy — have been investing rapidly in alternatives. Geothermal energy, abundant in the volcanic substrata beneath parts of the Netherlands, is being accessed through deep boreholes to provide low-carbon heat. Residual heat from data centres and industrial processes is being piped to greenhouse clusters. Large-scale heat pumps powered by renewable electricity are being installed. The industry has committed, under Dutch government pressure, to substantial reductions in fossil fuel dependence by 2040.
But the transition is slow and expensive, and in the meantime, enormous quantities of gas continue to burn.
Kenya’s Altitude Advantage — and Its Water Problem
The contrast with flower production in East Africa could hardly be more stark. Around Lake Naivasha, a rift valley lake at 1,884 metres above sea level, an entire flower industry has grown up that barely needs heating at all.
The lake’s altitude moderates temperatures naturally, providing the cool nights that many cut flower varieties require to develop properly. Daytime temperatures are warm but rarely excessive. Rainfall, supplemented by lake irrigation, is generally adequate. The climate, in short, does a large part of the greenhouse’s job for free. Most Kenyan flower farms use simple polyethylene-covered structures — not for heating, but for protection from hail, pests, and excessive rain. Some use no cover at all.
The result is that Kenyan roses have a carbon footprint that, even accounting for air freight to European markets, remains substantially lower than Dutch-grown equivalents. This is why the environmental case for Kenyan flowers is stronger than it might initially appear, and why the “food miles” framing, imported from the food debate, can mislead when applied to flowers.
But the East African industry has its own environmental ledger, and it makes for uncomfortable reading in different ways. Lake Naivasha has shrunk dramatically since intensive flower farming began in the 1980s. The lake level has fluctuated significantly, with periods of severe decline linked to irrigation withdrawal, though the relationship between flower farming and lake health is contested — other factors, including changes in rainfall and land use in the catchment, also play a role. What is clear is that the flower industry extracts enormous volumes of water from a closed-basin lake with no outlet, and that this withdrawal has ecological consequences for the papyrus beds, fish populations, and waterbird colonies that depend on it.
Heavy pesticide use has also been documented in Kenyan growing regions. The warm, humid microclimate inside growing structures — even unheated ones — creates ideal conditions for fungal diseases and insect pests, requiring regular chemical intervention. Worker exposure to pesticides has been a persistent concern raised by labour rights organisations, even as the industry has made genuine improvements in health and safety standards over the past two decades.
The lesson is that removing the carbon problem of heating does not remove the environmental problem of growing flowers at industrial scale. It relocates and transforms it.
The Netherlands’ Geothermal Gamble
Return to Westland, and to a greenhouse complex near the town of Monster — the name is a coincidence of Dutch etymology — where geothermal drilling has produced something remarkable: a pair of boreholes descending nearly 2,500 metres into the earth, tapping water at 75 degrees Celsius that flows continuously to the surface, surrenders its heat to the greenhouse network through a heat exchanger, and returns, cooled, to the earth via a second borehole. No combustion. No CO₂ from heating. Just the slow, inexhaustible warmth of the planet’s interior.
Dutch growers have drilled dozens of such doublet systems since the mid-2000s, and the technology has matured considerably. The geology of the Netherlands — thick sedimentary layers of porous sandstone saturated with ancient seawater — is well-suited to geothermal extraction. The main limitations are cost (a doublet system requires capital investment of several million euros) and the uneven distribution of suitable geology across the growing regions.
Where geothermal heat is available, the transformation can be dramatic. Growers who have made the transition report heating cost reductions of 80 to 90 percent compared to gas, and carbon reductions to match. But geothermal cannot heat the entire Dutch industry. Estimates suggest it can ultimately supply perhaps 40 percent of horticultural heat demand in the areas where geology permits. The remainder will require other solutions.
One increasingly attractive option is aquifer thermal energy storage, or ATES — a technology that stores summer heat underground and retrieves it in winter. Greenhouses naturally overheat in summer and require cooling; aquifer systems can absorb this excess heat into groundwater layers and return it six months later. Many Dutch greenhouse clusters now operate ATES systems alongside geothermal, creating a remarkably efficient seasonal energy exchange. The greenhouse, in effect, borrows from summer to pay for winter — a form of thermal banking that requires no combustion at all.
LEDs and the Second Energy Revolution
Heating is only part of the greenhouse energy equation. Light — or rather, the absence of sufficient natural light during northern winters — is the other great challenge, and historically the other great energy drain.
Traditional high-pressure sodium (HPS) grow lights, which have illuminated Dutch greenhouses since the 1970s, consume electricity at a rate that makes them one of the more significant power draws in the industrial world. The Westland district, at full winter production, draws electricity that is visible from orbit. HPS lamps emit light across a broad spectrum, much of which plants cannot use effectively, and they generate substantial waste heat — useful in winter, problematic in summer.
The transition to LED grow lights, which has accelerated dramatically since the mid-2010s, represents a genuine efficiency revolution. Modern horticultural LEDs can be tuned to emit precisely the wavelengths — primarily red and blue — that plant photosynthesis uses most efficiently. They consume 40 to 60 percent less electricity than equivalent HPS installations. They generate far less heat, reducing the cooling burden in summer. And their longer lifespan reduces replacement costs and waste.
The environmental benefit depends entirely on the source of the electricity, of course. Dutch greenhouse electricity has historically been generated largely by natural gas — including, in an ironic loop, by the CHP units in the greenhouses themselves. As the Dutch grid decarbonises with offshore wind and solar, the LED transition becomes genuinely meaningful in climate terms. But until the grid is clean, switching from HPS to LED simply shifts the emissions from the greenhouse to the power station.
Colombia’s Plastic Plateau
Between the highland extremes of Dutch glass and Kenyan open-air culture lies a third model: the Colombian plateau greenhouse, which has become the world’s most productive environment for cut rose growing.
The Sabana de Bogotá sits at 2,600 metres, in a climate that is cool and stable but rarely cold enough to damage plants. Colombian growers use greenhouse structures primarily to control rain, humidity, and pests — not, for the most part, to add heat. But the scale of the infrastructure is extraordinary: hundreds of thousands of hectares covered in polyethylene film, creating an artificial landscape that has profoundly altered the hydrology and ecology of the savanna.
The original Sabana was a high-altitude wetland ecosystem — a mosaic of lakes, marshes, and native grassland that was one of the most biodiverse highland plains in South America. Much of it was drained for agriculture long before the flower industry arrived, but the expansion of greenhouse coverage since the 1980s has accelerated habitat loss and complicated watershed management. The plastic covering intercepts rainfall, altering runoff patterns. Irrigation wells have drawn down aquifers. The iconic Bogotá River, which drains the savanna, carries a cocktail of agricultural chemical residues.
Plastic itself poses a disposal problem that the industry has been slow to address. Polyethylene greenhouse film has a lifespan of three to five years before UV degradation and mechanical wear require replacement. In Colombia and many other producing countries, collection and recycling infrastructure for agricultural plastics is limited. Burning or burying used film remains common. The volume of plastic waste generated by the global flower industry is not well documented, but informed estimates are sobering.
When the Greenhouse Meets a Warming World
There is a profound irony embedded in the greenhouse flower industry’s situation. The technology that allows it to warm air and grow plants out of season contributes, at scale, to the warming of the planetary atmosphere — and that warming is now beginning to feed back into the industry’s own growing conditions in ways that are difficult to manage.
In the Netherlands, milder winters have reduced heating requirements — a short-term economic benefit — but also increased the incidence of fungal diseases that thrive in warmer, damper conditions, and created new pest populations that no longer die back in cold winters. In Kenya, increased variability in rainfall patterns has made irrigation management more complex and lake levels less predictable. In Colombia, unusual temperature events on the Sabana — increasingly common as regional climate patterns shift — can damage crops in facilities designed for a narrower range of conditions.
Growers are adapting. Dutch greenhouse architects are now designing facilities with greater passive solar capacity, better insulation, and dynamic ventilation systems that can manage a wider range of external temperatures. In Kenya and Ethiopia, new growing regions at even higher altitudes are being developed as insurance against warming lowlands. In all regions, investment in climate monitoring and precision environment control is accelerating.
But the central tension remains unresolved. The world’s appetite for flowers — and for the symbolic gesture of giving them — is not diminishing. The infrastructure required to satisfy that appetite at global scale, in every season, continues to impose costs on the same planetary systems that keep the air warm, the rain coming, and the soil alive. The greenhouse, that elegant trick of glass and warmth, sits uncomfortably at the intersection of nature and its own undoing.
A Reckoning Deferred
On a winter afternoon in Westland, with the light fading early and the greenhouses beginning their nightly illumination, a rose grower walks a visiting journalist through rows of flowers that have no business existing in January. The heating pipes along the floor are warm to the touch. The air smells of damp soil and faintly of fertiliser. Outside, the temperature has dropped to two degrees.
“People ask me if I feel guilty,” she says, pausing near a spray of orange roses in full, spectacular bloom. “I feel complicated. My family has done this for three generations. We employ two hundred people. We are investing everything we have in geothermal. And still — yes. Some days I look at the gas meter and I feel complicated.”
The roses do not look complicated. They look, as they always do, perfect: coloured objects of desire, engineered for beauty and engineered for endurance, carrying within their petals no record of the gas burned, the water drawn, the plastic spread across the plateau to bring them here. That record exists, somewhere, in the aggregate arithmetic of an industry that has learned to manufacture spring. Reading it clearly, and acting on what it says, may be the more pressing kind of love letter the world needs to write.