Climate, Water, and the Growing Season

TL;DR. Where a crop can grow is decided by four numbers: how cold it gets in winter, how much heat accumulates in summer, how long the frost-free window is, and how much water arrives and when. Temperate fruit trees have an additional and counterintuitive requirement: they need a specific amount of cold each winter, and without it they will not flower properly, which is the single biggest reason apples and cherries do not grow in the tropics. Almost every fact about where your food comes from traces back to these constraints, and climate change is currently moving all of them.

Key takeaways

• Chill hours are a hard requirement, not a preference. Most apple and cherry varieties need 800 to 1,200 hours below about 7 °C (45 °F) each winter or they break dormancy erratically and crop poorly.
• Growing degree days predict when a crop will be ready far better than the calendar does, and they are how commercial growers actually schedule harvest and spraying.
• Frost during flowering is the single most destructive weather event in fruit growing, and it lasts one night.
• About 70 percent of all freshwater withdrawal on Earth goes to agriculture. Irrigation is why California, Spain, Egypt, and the Indian Punjab produce what they do, and it is why several aquifers are being emptied.
• Latitude, altitude, and proximity to water are the three levers that let one crop grow in surprisingly different places. A high plateau near the equator can behave like a temperate climate.

Temperature: four different numbers

In short: Cold tolerance, heat accumulation, chilling requirement, and frost timing are separate constraints, and a crop must satisfy all four.

1. How cold it gets, at the worst moment

Every plant has a temperature below which it dies. For a lemon tree that is around -3 °C (27 °F). For an olive, around -10 °C (14 °F). For a hardy apple rootstock, below -35 °C (-31 °F). This is what hardiness zones measure: the USDA system divides the world into zones by average annual minimum temperature, each zone 5.6 °C (10 °F) wide. Zone 5 averages a coldest night of about -29 to -23 °C; zone 9 about -7 to -1 °C; zone 11 never freezes.

Hardiness zones are useful and routinely over-interpreted. They tell you about winter lows and nothing about summer heat, humidity, rainfall, or soil, so a plant can be "zone-appropriate" and still fail. Britain and the US Pacific Northwest share zones with places whose summers are ten degrees hotter, and the crops that thrive differ completely.

2. How much heat accumulates over the season

In short: Plants develop according to accumulated warmth, not elapsed days, which is why the same variety ripens six weeks apart in two regions.

A plant's rate of development depends on temperature, so growers count growing degree days (GDD) rather than calendar days. The arithmetic is simple:

$$\mathrm{GDD} = \sum_{\text{days}} \max!\left(0,; \frac{T_{\max} + T_{\min}}{2} - T_{\text{base}}\right)$$

The base temperature is the threshold below which that crop effectively stops developing: about 10 °C for maize, grapes, and most warm-season crops, and about 4 to 5 °C for many cool-season crops and fruit trees. So a day with a high of 24 °C and a low of 12 °C contributes $(24+12)/2 - 10 = 8$ growing degree days for maize.

Crops carry GDD requirements the way recipes carry cooking times:

Two consequences. First, a short-season variety is not a marketing term; it is a genuinely different plant, bred to complete its cycle in fewer accumulated degree days, usually at a yield cost. This is why northern regions grow different maize varieties than southern ones. Second, GDD explains why the same apple variety is picked in early August in one valley and late September a hundred kilometres away.

3. Chilling requirement: the cold that fruit trees cannot do without

In short: Temperate trees measure winter by accumulating hours in a specific cold band, and refuse to wake up until the count is met.

This is the fact that most surprises people, and it is why you cannot grow apples in Singapore.

Temperate deciduous trees enter dormancy in autumn. Dormancy is not simply a response to cold, it is an internally enforced shutdown with a built-in safety catch: the tree will not resume growth until it has accumulated a set number of chill hours, conventionally hours between roughly 0 and 7 °C (32 to 45 °F). The evolutionary logic is sound. A warm spell in December is not spring, and a tree that flowered in response to it would lose everything to the next frost. The chill requirement is a counter that says "winter has genuinely happened."

If chilling is not met, the consequences are specific and bad: bud break is delayed and scattered, flowering is spread over weeks so pollination fails, leaves emerge patchily, and yield collapses. The tree does not die. It just does not produce.

Plant breeding has produced low-chill cultivars of apple, peach, and blueberry specifically so subtropical regions can grow them, and this is a live and commercially important area. It is also why climate change is a direct threat to fruit growing in places like southern Spain, California's Central Valley, and parts of Australia and South Africa: warming winters mean chill accumulation is falling, and some regions are approaching the point where their existing orchards will no longer perform.

Don't be confused: vernalisation is a different thing with a similar flavour. Chilling requirement releases dormancy in woody perennials. Vernalisation is the cold requirement that lets certain annuals and biennials flower at all: winter wheat sown in autumn will not produce grain unless it experiences winter, and carrots, beets, and cabbages will not bolt to seed in their first year. Same theme, different mechanism, and it is why winter wheat and spring wheat are different products.

4. Frost, and the one night that ruins a year

In short: Open flowers and small fruitlets die at about -2 °C, so a single late spring frost can eliminate a crop that took a year to build.

Cold hardiness is not a fixed property of a tree; it changes dramatically through the year. A dormant apple tree in January shrugs off -25 °C. The same tree in full bloom in April is destroyed by -2 °C. The critical temperatures for apple are roughly:

The vulnerability window is short, a few weeks, and the loss is total: no flowers, no fruit, and no second chance until next year. This is why fruit-growing regions cluster near large bodies of water. Lake Michigan, Lake Ontario, the Great Lakes generally, the Rhine, the sea around Kent and Normandy: large water masses warm slowly in spring, which holds back bud break until the frost risk has passed, and they release heat in autumn, extending the season. Michigan's entire cherry industry exists because of the lake. So does Ontario's Niagara fruit belt.

Growers fight frost with a small arsenal of physics:

• Wind machines: large fans that mix the warmer air sitting 10 to 15 metres up back down to tree level. Works only against radiation frost on still, clear nights, which is the common kind. Useless against a cold air mass moving in.
• Overhead irrigation: counterintuitive but sound. Water releases 334 joules per gram as it freezes (latent heat of fusion), which holds the ice-coated bud at exactly 0 °C as long as you keep spraying. Stop early and you do more damage than you prevented.
• Orchard heaters and burning: expensive, polluting, still used in high-value crops.
• Site selection: cold air is dense and flows downhill, pooling in hollows. A slope is warmer at night than the flat ground below it. This is why vineyards and orchards sit on hillsides, and why the bottom of the valley grows pasture.
• Delayed pruning and evaporative cooling to push bloom later, which buys days.

Light and day length

In short: Some plants flower according to night length, which decides where they can be grown and when they can be forced.

Beyond raw energy for photosynthesis, day length acts as a calendar signal. Plants are classified by photoperiod response:

• Short-day plants (really long-night plants) flower when nights exceed a critical length: chrysanthemum, poinsettia, soybean, rice, some strawberries. Greenhouse growers force poinsettias for Christmas by covering them with blackout cloth.
• Long-day plants flower when nights are short: spinach, lettuce, onion, barley, potato tuberisation in some varieties. This is why spinach and lettuce bolt (rush to flower and turn bitter) in early summer, and why onion varieties are sold explicitly as short-day, intermediate-day, or long-day types, matched to latitude. Planting a long-day onion in Texas produces greens and no bulb.
• Day-neutral plants flower on age or size instead: tomato, cucumber, maize, and the "everbearing" strawberry varieties that made year-round supply possible.

At the equator day length barely varies, which is one reason equatorial agriculture runs on day-neutral and short-day crops and on continuous rather than seasonal cropping.

Water: the largest input by mass

In short: Agriculture uses about 70 percent of humanity's freshwater withdrawals, and the timing of water matters as much as the amount.

Plants use enormous quantities of water, and mostly not for chemistry. Photosynthesis consumes a trivial amount. The rest evaporates from the leaves through open stomata, the microscopic pores a plant must open to let carbon dioxide in. That is the fundamental trade-off of terrestrial plant life: to eat, you must open the door, and when the door is open you lose water.

The efficiency of that trade is measured as water use efficiency, and it varies enormously by photosynthetic type:

CAM plants solve the problem by opening their stomata at night, when it is cool and humid, storing the carbon dioxide as an acid, and using it the next day with the pores shut. That is why pineapple and dragon fruit thrive in places that would kill a lettuce, and it is a good illustration that "drought tolerant" is not vagueness, it is a specific biochemistry.

How much water crops actually need

Approximate seasonal requirements, expressed as depth of water over the field:

Those per-kilogram figures deserve care. A litres-per-kilogram number depends on whether you count rainfall that would have fallen anyway (green water) or only irrigation drawn from rivers and aquifers (blue water), and headlines almost always use the larger, less meaningful figure. Almonds are the famous example: the number is real, and the reason it is a political issue is not the number itself but that Californian almonds are irrigated from a stressed system, and that almond trees, unlike annual crops, cannot be fallowed in a drought year without killing a twenty-year investment.

Timing matters as much as total

Every crop has a critical period where water stress does disproportionate damage:

• Maize: silking and pollination. Drought during that two-week window can halve yield even if rain returns immediately after.
• Wheat: grain filling.
• Potato: tuber initiation and bulking.
• Grapes: deliberately stressed after fruit set, because mild water stress concentrates sugars and phenolics. This is regulated deficit irrigation and it is one of the few cases where growers withhold water on purpose.
• Apple: cell division in the first 40 days after bloom, which sets final fruit size.

And too much water at the wrong time is equally destructive: rain just before a cherry harvest causes the fruit to absorb water through the skin and split, which can destroy a crop in an afternoon. Cherry growers in some regions fly helicopters over orchards to blow the water off.

How irrigation is delivered

Drip irrigation delivers water slowly to the root zone, cutting evaporation and weed growth, and lets fertiliser be dissolved into the water (fertigation) for precise feeding. It is expensive to install and is the main reason high-value fruit in arid regions is economically possible at all.

The dark side is groundwater. The Ogallala Aquifer under the US High Plains, the North China Plain aquifer, and the aquifers under the Indian Punjab are all being drawn down faster than they recharge, in some places by metres a year. Substantial parts of current global food production are running on stored water that will not be replaced on any human timescale. And irrigating with even slightly salty water in a hot climate concentrates salt in the soil as the water evaporates, which is salinisation, the mechanism that ended agriculture in parts of ancient Mesopotamia and is currently degrading millions of hectares.

Humidity, wind, and the things nobody mentions

Humidity decides disease pressure more than almost anything else. Fungal spores need leaf wetness to germinate, so a humid climate means constant fungicide use or disease-resistant varieties. This is why grapes for wine are grown in dry-summer Mediterranean climates, why the Pacific Northwest is excellent for apples (dry summers, low disease pressure) and difficult for stone fruit, and why table grape production concentrates in arid regions with irrigation.

Wind desiccates, breaks branches, scars fruit skin against twigs (which is cosmetic damage that downgrades the crop), and disrupts bee flight, which cuts pollination. Windbreaks are standard in exposed orchards and can raise yields by double-digit percentages.

Hail is the other one-night catastrophe. High-value orchards in Europe and South America increasingly grow under permanent hail netting, which is expensive, changes the light spectrum reaching the fruit, and reduces sunburn as a side benefit.

Sunburn on fruit is a real and growing problem: apples and grapes on the exposed side of the canopy can be damaged above about 40 °C fruit surface temperature. Growers respond with overhead misting, shade netting, and kaolin clay sprays that leave a white reflective film.

The three levers: latitude, altitude, and water

In short: You can approximate a temperate climate near the equator by going up a mountain, which is why highland tropics grow coffee, apples, and strawberries.

Temperature falls with altitude at roughly 6.5 °C per 1,000 metres. That means 1,000 metres of elevation is worth something like 1,000 kilometres of latitude in temperature terms, though not in day length.

This is why:

• Coffee grows on tropical highlands (Ethiopian highlands, Colombian Andes, Kenyan slopes at 1,200 to 2,000 m). Arabica coffee wants 18 to 22 °C, which at those latitudes exists only up high. Lowland tropics grow robusta instead, which tolerates heat and tastes harsher.
• Kenya and Ethiopia grow apples and strawberries commercially at altitude.
• Ecuador and Colombia dominate cut flowers: equatorial day length is stable year-round and altitude supplies cool nights, so roses grow slowly with long, strong stems.
• Mexican and Peruvian highland valleys supply northern-hemisphere winter vegetables.

Proximity to large water bodies is the third lever, moderating both extremes. The Mediterranean basin, coastal California, coastal Chile, the Cape region of South Africa, and southwestern Australia share a Mediterranean climate: wet mild winters, hot dry summers. Those five regions, on five different continents, grow the same crops for the same reason, which is why olives, grapes, citrus, almonds, and stone fruit come overwhelmingly from that short list of places.

What climate change is doing to all of this

This is not a speculative section; the effects are already in the trade data.

• Chill hours are declining in warm fruit regions, threatening orchards that were planted for a climate that is going away. Low-chill breeding is racing this.
• Bud break is advancing faster than the last frost date is retreating in many temperate regions, so the frost risk window is getting worse, not better. Several recent European fruit crop failures follow exactly this pattern: a warm March, early bloom, an April frost.
• Growing zones are moving poleward and uphill, which sounds like a wash and is not, because soils, water, and infrastructure do not move with them.
• Water is being redistributed, with more precipitation falling in heavy events and less in gentle ones, which is worse for soil and for aquifer recharge.
• Crop nutritional quality is measurably shifting: elevated CO₂ raises yields of C3 crops somewhat while reducing their protein, zinc, and iron concentrations by a few percent, an effect confirmed in field experiments. More food, slightly thinner food.

The bottom line

• Four temperature constraints, not one: winter minimum survival, accumulated summer heat, winter chilling requirement, and frost timing at bloom. A crop must satisfy all four.
• Chill hours are why apples, cherries, and pears are temperate crops and cannot simply be moved to warm climates, and warming winters threaten existing orchards.
• One frost night at full bloom destroys a year's crop. That is why fruit regions sit beside lakes and on slopes.
• Water use is dominated by transpiration, not chemistry, and the timing of water stress matters as much as the total. Roughly 70 percent of human freshwater withdrawal goes to agriculture, some of it from aquifers being permanently drawn down.
• Altitude substitutes for latitude at about 1,000 m per 1,000 km, which is why the highland tropics grow temperate crops.

Sources and notes

Growing degree day models, chilling requirement figures, and critical frost temperatures by phenological stage follow extension service publications, particularly Washington State University and Michigan State University tree fruit programmes, which publish the bud-hardiness tables used here. Water use efficiency by photosynthetic type and crop water requirements follow FAO Irrigation and Drainage Paper 56 (Allen et al., 1998). Green and blue water accounting follows Hoekstra and Mekonnen's water footprint work. Aquifer depletion figures for the Ogallala, North China Plain, and Indian Punjab are from USGS and published groundwater assessments. Chill-hour decline under warming and its effect on fruit regions is documented in Luedeling's work on chill models. Photoperiod classes follow standard horticultural references. Elevated CO2 reducing crop protein, zinc, and iron is from Myers et al., Nature, 2014, and subsequent FACE experiments.

Open questions. Which chill model best predicts real bud break is unsettled, and different models disagree about how quickly warming will make particular regions unviable. Projections of where growing zones will move are more confident about temperature than about whether soil, water, and infrastructure will follow.

👉 Next: what a farm actually does across a year, from planting through pollination to harvest.