Trees and Precipitation in Permaculture

Permaculture Designers Manual




Section 6.5 –

Trees and Precipitation in Permaculture


Trees have helped to create both our soils and atmosphere.

The first by mechanical (root pressure) and chemical (humic acid) breakdown of rock, adding life processes as humus and myriad decomposers.

The second by gaseous exchange, establishing and maintaining an oxygenated atmosphere and an active water vapor cycle essential to life.

The composition of the atmosphere is the result of reactive processes, and forests may be doing about 80% of the work, with the rest due to oceanic or aquatic exchange.

Many cities, and most deforested areas such as Greece, no longer produce the oxygen they use.


The basic effects of trees on water vapor and wind streams are:

Compression of streamlines and induced turbulence in air flows;

Condensation phenomena, especially at night; Re-humidification by the cycling of water to air;

Snow and melt water effects;

Provision of nuclei for rain.

We can deal with each of these in turn (realizing that they also interact).



Wind streams now across a forest. The streamlines that impinge on the forest edge are partly deflected over the forest (almost 60% of the air) and partly absorbed into the trees (about 40% of the air).

Within 1000m (3,300 feet) the air entering the forest, with its tonnages of water and dust, is brought to a standstill. The forest has swallowed these great energies, and the result is an almost imperceptible warming of the air within the forest, a generally increased humidity in the trees (averaging 15-18% higher than the ambient air), and air in which no dust is detectable. (Figure 6.3)

Under the forest canopy, negative ions produced by life processes cause dust particles (++) to dump or adhere each to the other, and a fallout of dispersed dust results.

At the forest edge, thick-stemmed and especially wind-adapted trees buffer the front-line attack of the wind. If we cut a windward forest edge, and remove these defenses, windburn by salt, dust abrasion, or just plain wind force may well kill or throw down the inner forest of weaker stems and less resistant species.

This is a commonly observed phenomenon, which I have called “edge break“.

Conversely, we can set up a forest by planting tough, resistant trees as windbreak and so protect subsequent downwind plantings.

Forest edges are therefore to be regarded as essential and permanent protection and should never be cut or removed.

If dry hot air enters the forest, it is shaded, cooled, and humidified.

If cold humid air enters the forest, it is warmed, dehumidified, and slowly released via the crown of the trees.

We may see this warm humid air as misty spirals ascending from the forest. The trees modify extremes of heat and humidity to a life enhancing and tolerable level.

The winds deflected over the forest cause compression in the streamlines of the wind, an effect extending to twenty times the tree height, so that a 12m (40 foot) high line of trees compresses the air to 244m (800 feet) above, thus creating more water vapor per unit volume and also cooling the ascending air stream.

Both conditions are conducive to rain.

As these effects occur at the forest EDGE, a single hedgerow of 40% permeability wilt causes similar compression.

In flat country and especially in the path of onshore winds, fine grid placements of rain gauges in such countries as Holland and Sweden reveal that 40% of the rainfall measured downwind of trees and mounds 12m (40 feet) or more in height is caused by these compression phenomena. If wind speeds are higher (32 km/h or more), the streamlines may be preserved, and rain falls perpendicular to the windbreak.

However, at lower wind speeds (the normal winds), turbulence and overturn occur.

Wind streaming over the hedgerow or forest edge describes a spiral section, repeated 58 times downwind, so that a series of compression fronts, this time parallel to the windbreak, are created in the atmosphere.

This phenomenon was first described by Ekman for the compression fronts created over waves at sea.

The Ekman spirals over trees or bluffs may result in a ranked series of clouds, often very regular in their rows. They are not perfectly in line ahead, but are deflected by drag and the Coriolus force to change the wind direction, so that the wind after the hedgerow may blow 5-15 degrees off the previous course.

(One can imagine that ranks of hedgerows placed to take advantage of this effect would eventually bring the wind around in a great ground spiral.)

Winds at sea do in fact form great circuses, and bring cyclonic rains to the westerly oceanic coasts of all continents.

These cyclones themselves create warm and cold fronts which ridge up air masses to create rain.

In total, hedgerows across wind systems have a profound effect on the airstreams passing over them, and a sub­ sequent effect on local climate and rainfall.



On the sea-facing coasts of islands and continents, the relatively warmer land surface creates quiet inshore airflows towards evening and too many areas cooler water laden air flows inland.

Where this humid air flows over the rapidly cooling surfaces of glass, metal, rocks or the thin laminae of leaves, condensation occurs and droplets of water form.

On leaves, this may be greatly aided by the colonies of bacteria (Pseudomonas) which also serve as nuclei for frost crystals to settle on leaves.

These saturated airstreams produce seaward-facing mosses and lichens on the rocks of fresh basalt rows, but more importantly condense in trees to create a copious soft condensation which, in such conditions, may far exceed the precipitation caused by rainfall.

Condensation drip can be as high as 80-86% of total precipitation of the upland slopes of islands or sea coasts, and eventually produces the dense rainforests of Tasmania, Chile, Hawaii, Washington/Oregon and Scandinavia.

It produced the redwood forests of California and the giant laurel forests of pre-conquest Canary Islands (now an arid area due to almost complete deforestation by the Spanish).

A single tree such as a giant Til (Ocotea foetens) may present 16 ha of laminate leaf surface to the sea air and there can be 100 or so such trees per surface hectare, so that trees enormously magnify the available condensation surface.

The taller the trees, as for example the giant redwoods and white pines, the larger the volume of moist air intercepted and the greater the precipitation that follows.

All types of trees act as condensers; examples are Canary Island pines, laurels, holm oaks, redwoods, eucalypts, and Oregon pines.

Evergreens work all year, but even deciduous trees catch moisture in winter.

Who has not stood under a great tree which “rains” softly and continuously at night, on a clear and cloudless evening?

Some gardens, created in these conditions, quietly catch their own water while neighbors suffer drought.

The effects of condensation of trees can be quickly destroyed.

Felling of the forests causes rivers to dry up, swamps to evaporate, shallow water to dry out, and drought to grip the land.

All this can occur in the lifetime of a person.

Precipitation from clear air is much less than that from fog, from which the precipitation by condensation often exceeds the local rainfall.

Advection fogs are most noticeable where cold currents such as the Oya Shio off East Asia and the Labrador Current off northeast America cause humid inland airstreams in spring and summer.

South facing coasts near Newfoundland get 158 days of fog per year. Wherever mountains or their foothills face onshore night winds, fog condensation will probably exceed rainfall.

On Table Mountain (South Africa) and on Lanai (Hawaii), fog drip has been measured at 130-330 cm and in both cases condensation exceeds rainfall.

Redwoods in California were once restricted to the fog belt, but will grow well in areas of higher rainfall without fogs (Chang. 1968).

In Sweden “… wooded hills rising only 3050m (9,500 feet) above the surrounding plains may cause precipitation (rain only) during cyclonic spells (fronts) to be increased by 50-80% compared with average falls over the lowland.”

In most countries, however, the rain gauge net is too coarse to detect such small variations (Chorley and Berry, 1971).



If it rains again, and again, the clouds that move inland carry water mostly evaporated from forests and less and less water evaporated from the sea.

Forests are cloud-makers both from water vapor evaporated from the leaves by day, and water transpired as part of life processes on high islands, standing clouds cap the forested peaks, but disappear if the forests are cut.

The great bridging cloud that reached from the forests of Maui to the Island of Kahoolawe, remembered by the fathers of the present Hawaiian settlers, has disappeared as cutting and cattle destroyed the upper forests on Maui and so lifted the cloud cap from Kahoolawe, leaving this lower island naked to the sun.

With the cloud forests gone, and the rivers dry, Kahoolawe is a true desert island, now used as a bombing range for the U.S. Air Force.

A large evergreen tree such as Eucalyptus globulus may pump out 3,600-4,500l of water a day, which is how Mussolini pumped dry the Pontine Marshes of Italy. With sixty or so of these trees to the hectare, many tens of thousands of liters of water are returned to the air to become clouds.

A forest can return (unlike the sea) 75% of its water to air, “in large enough amounts to form new rain clouds.” (Bayard Webster, “Forests’ Role in Weather Documented in the Amazon“, New York Times (Science Section), 5 July ’83).

Forested areas return, ten times as much moisture as bare ground does and twice as much as grasslands.

In fact, as far as the atmosphere itself is concerned, “the release of water from trees and other plants accounts for half, or even more, of all moisture returned to air.” (Webster, ibid.) This is a critical finding that adds even more data to the relationship of desertification by deforestation.

It is data that no government can ignore. Drought in one area may relate directly to deforestation in an upwind direction.

This study “clearly shows that natural vegetation must play an important role in the forming of weather patterns” (quote from Thomas E. Lovejoy, Vice-president of Science, World Wild life Fund).

Clouds form above forests, and such clouds are now mixtures of oceanic and forest water vapor, clearly distinguishable by careful isotope analysis.

The water vapor from forests; contain more organic nuclei and plant nutrients than does the “pure” oceanic water. Oxygen isotopes are measured to determine the forests’ contribution, which can be done for any cloud system.

Of the 75% of water returned by trees to air, 25% is evaporated from leaf surfaces, and 50% transpired. The remaining 25% of rainfall infiltrates the soil and eventually reaches the streams.

The Amazon discharges 44% of all rain falling, thus the remainder is either locked into the forest tissue or returns to air.

Moreover, over the forests, twice as much rain falls then is available from the incoming air, so that the forest is continually recycling water to air and rain, producing 50% of its own rain (Webster, ibid.).

These findings forever put an end to the fallacy that trees and weather are unrelated.

Vogel (1981), applying the “principle of continuity” of fluids to a tree, calculates that sap may rise, in a young oak, fifty times as fast as the leaves transpire (needing only 7% of the total trunk area as conductive tissue, with an actual sap speed of 1 cm/sec).

It is thus certain that only perhaps one-fiftieth of the xylem is conducting sap upwards at any one time, and that most xylem cells contain either air or sap at standstill.

Perhaps too, the tree moves water up in pulsed stages rather than as a universal or continuous stream flow. (Figure 6.5 A and B)

With such rapid sap flows, however, we can easily imagine the water recycled to atmosphere by a large tree, or a clump of smaller trees.

It is a wonder to me that we have any water available after we cut the forests or any soil.

There are dozens of case histories in modern and ancient times of such desiccation as we find on the Canary Islands following deforestation, where rivers once ran and springs flowed.

Design strategies are obvious and urgent – “save all forest that remains, and plant trees for increased condensation on the hills that face the sea.”



Although trees intercept some snow, the effect of shrubs and trees is to entrap snow at the edges of clumps, and hold 75-95% of snowfall in shade.

Melting is delayed for 210 days compared with bare ground, so that release of snowmelt is a more gradual process. Of the trapped snow within trees, most is melted, while on open ground snow may sublime directly to air.

Thus, the beneficial effect of trees on high slopes is not confined to humid coasts. On high cold uplands such as we find in the continental interiors of the U.S.A. or Turkey near Mt. Ararat, the thin skeins of winter snow either blow off the bald uplands, to disappear in warmer air, or else they sublime directly to water vapor in the bright sun of winter.

In neither case does the snow melt to round water, but is gone without productive effect and no streams result on the lower slopes.

Even a thin belt of trees entraps large quantities of driven snow in drifts. The result is a protracted release of melt water to river sources in the highlands and stream flow at lower altitudes.

When the forests were cleared for mine timber in 1846 at Pyramid Lake, Nevada, the streams ceased to flow and the lake levels fell.

Add to this effect that of river diversion and irrigation and whole lakes rich with fish and waterfowl have become dustbowls, as has Lake Winnemucca.

The Cuiuidika’s Indians (Paiute) who live there lost their fish, waterfowl and freshwater in less than 100 years.

The cowboys have won the day, but ruined the future to do so.



The upward spirals of humid air coming up from the forest carry insects, pollen, and bacteria aloft.

This is best seen as nights of gulls, swifts and Ibis spiraling up with the warm air and actively catching insects lifted from the forest; their gastric pellets consist of insect remains.

It is these organic aerial particles (pollen, leaf dust, and bacteria mainly) that create the nuclei for rain.

The violent hailstorms that plague Kenya tea plantings may well be caused by tea dust stirred up by the local winds and the feet of pickers, and “once above the ground the particles are easily drawn up into thunderheads to help form the hailstorms that bombard the tea-growing areas in astounding numbers… Kenyan organic tea leaf litter caused water to freeze in a test chamber at only -5°C, in comparison with freezing points of -11°C for eucalyptus grove leaf litter, and -80°C for the litter from the local indigenous forests” (of Colorado).

That is, tea litter “Is a much better seeding agent than silver iodide, which requires -80°C to -100°C to clouds.” (New Scientist, 22 Mar 79).

Thus, the materials given up by vegetation may be a critical factor in the rainfall inland from forests.

All of these factors are clear enough for any person to understand.To doubt the connection between forests and the water cycle is to doubt that milk flows from the breast of the mother, which is just the analogy given to water by tribal peoples.

Trees were “the hair of the earth” which caught the mists and made the rivers flow.

Such metaphors are dear allegorical guides to sensible conduct and caused the Hawaiians (who had them brought on earlier environmental catastrophes) to “taboo” forest cutting or even to make tracks on high slopes and to place mountain trees in a sacred or protected category.

Now that we begin to understand the reasons for these beliefs, we could ourselves look on trees as our essential companions, giving us all the needs of life, and deserving of our care and respect.

It is our strategies on-site that make water a scarce or plentiful resource.

To start with, we must examine ways to increase local precipitation. Unless there is absolutely no free water in the air and earth about us (and there always is some), we can usually increase it on-site.


Here are some basic strategies of water capture from air:

We can root the air by shade or by providing cold surfaces for it to flow over, using trees and shrubs, or metals, including glass.

We can cool air by forcing it to higher altitudes, by providing windbreaks, or providing up draughts from heated or bare surfaces (large concreted areas), or by mechanical means (big industrial fans).

We can provide condensation nuclei for raindrops to form on, from pollen, bacteria and organic particles.

We can compress air to make water more plentiful per unit volume of air, by forcing streamlines to converge over trees and objects, or forcing turbulent flow in airstreams (Ekman spirals).


If by any strategy we can cool air, and provide suitable condensation surfaces or nuclei, we can increase precipitation locally.

Trees, especially crosswind belts of tall trees, meet all of the criteria in one integrated system. They also store water for local climatic modification.

Thus, we can clearly see trees as a strategy for creating more water for local use.

In summary, we do not need to accept “rainfall” as having everything to do with total local precipitation, especially if we live within 30-100 km of coasts (as much of the world does), and we do not need to accept that total precipitation cannot be changed (in either direction) by our action and designs on site.


Image result for go to next page button