Shelter is measured in tree heights, and that changes what you plant
Everything about windbreak reach scales with mature height. A belt of 40-foot conifers at a leeward multiple of 10 shelters 400 feet downwind; the same multiple on a 12-foot shrub row reaches 120. That is why a hedge planted to protect a farmyard fails and a proper belt succeeds, and why the cost per foot of row is nearly irrelevant next to the mature height you will eventually get.
It also means the honest input is your local mature height, not the nursery tag. A species that makes 60 feet in a river bottom may top out at 30 on a dry exposed ridge, and shelter follows the 30. Walk round and look at 40-year-old specimens of whatever you are considering, on ground like yours.
Why the multiple is a field and not a constant
You will find leeward reach quoted at various multiples of height, and the spread between the figures is not sloppiness — it is a real dependence on density. A very dense belt cuts wind speed sharply just behind it and the effect dies away quickly. A moderately open belt cuts less right behind it but the reduction carries much further downwind, because air passing through the belt keeps the flow attached instead of letting it tumble back down.
Which of those you want depends on what you are protecting. A stockyard needing shelter in the first hundred feet is a different problem from a field wanting reduced evaporation over a quarter mile. Since the density is a design decision you have not made yet, this page will not pick the multiple for you.
The ends leak, so the belt runs past the thing it protects
Wind that meets a barrier goes over it, through it, or round it, and the air going round the ends speeds up. That is why a belt exactly as long as the yard it protects leaves the corners windier than they were. Extending the belt past the protected area at each end — the end-extension field, expressed in heights — moves the accelerated flow out past what you care about.
The same effect explains why belts are laid out for the prevailing damaging wind rather than the average wind. If the wind that matters is the January northwest one, orientation follows that, and the summer breeze is a secondary consideration.
Spacing in the row is tighter than you would plant anything else
A yard tree wants room to reach its full spread. A windbreak tree does not: the row has to close into a continuous line, so it is planted closer than mature spread and the trees grow into each other on purpose. That is the point where windbreak layout diverges completely from the arithmetic on the tree spacing calculator, which is built around mature canopy spread and open grids — the right approach for an orchard or a yard, and the wrong one here.
Between rows is different again, and wider: rows too close shade each other out from the side, and you want mower or sprayer access for the first several years. For measuring the height of trees you already have, the tree height calculator does it from a tape and an angle, and the tree watering calculator covers the establishment years, which is when belts are actually lost.
What this does not cover
Species selection, soil, disease exposure, wildlife value, snow behaviour and any local ordinance about planting near a road or a boundary. Snow deposition in particular is its own subject with its own siting rules — a belt placed to catch drifting snow is deliberately set well back from whatever it protects, and getting that distance wrong puts the drift on the drive instead of short of it. If what you are planning for is plowed snow rather than drifted snow, the snow stacking space calculator covers where the piles go.
Questions people ask
How far downwind does a windbreak really protect?
It depends on the density of the belt and on what you mean by protect, which is exactly why this page takes the multiple as an input instead of asserting one. A dense planting produces a large reduction in wind speed close behind it that fades quickly. A more porous one produces a smaller reduction that carries much further. Published figures for leeward reach vary substantially between sources and between densities, and none of them is a fact about your site. Take a figure from your conservation district or extension service, who can see the local conditions, and put it in the field.
How many rows should a windbreak have?
More rows give redundancy and a deeper belt, and the classic farmstead arrangement puts a shrub row on the windward side, one or more middle rows and a tall row behind. Fewer rows cost less land and less money and are far more vulnerable to a single loss. A one-row belt is workable where land is tight, but a gap in it is a hole the wind finds immediately, and single rows thin at the base with age. What matters more than the count is that the belt closes into a continuous barrier and stays closed for decades.
Why plant closer together than the mature spread?
Because a windbreak is trying to be a wall and an orchard is trying to be a collection of individual trees. Planted at mature spread, the row is a line of separate canopies with permanent gaps at the trunks, and the wind funnels through them faster than it goes over the top. Planted tighter, the crowns knit and the row functions as one object. The cost is that individual trees are more crowded, grow narrower and can be shorter-lived, which is a trade windbreak planting accepts deliberately.
Which direction should the belt run?
At right angles to the wind that causes the problem, which is usually not the average wind. On many sites the damaging wind is a winter one from a single quarter, and a belt laid out for the annual average protects poorly on the days that matter. Work out which direction the trouble comes from — drifting snow, drying summer wind, winter chill on stock — and lay the belt across that. If two directions matter, an L-shaped arrangement is common, which is why some farmsteads are sheltered on two sides.
How long before it does anything?
Years, and there is no way round that. Shelter scales with height, height takes time, and the first several years are the ones where the planting is most likely to fail, from drought, competition from grass, rodents or deer. Practically, that means the establishment plan matters more than the layout: weed control, watering through the first dry summers and protection from browsing. Order spares, because gaps replanted in year five never catch up with their neighbours and stay as holes in the belt.