What is ANR?
ANR is any set of activities, short of tree planting, which enhance the natural processes of forest regeneration. It includes the protective measures to remove barriers to natural forest regeneration (e.g. fire and livestock), along with additional actions to i) “assist” or “accelerate” the growth of natural regenerants that are already established in the restoration site (seedlings, saplings and live stumps of indigenous forest tree species) and ii) encourage seed dispersal into the restoration site.
The UN Food and Agriculture Organization (FAO) supported much of the research that helped to transform ANR from a concept into an effective and practicable technique, in partnership with the Philippines’ government and community-based NGO’s. FAO now promotes ANR as a method for enhancing the establishment of secondary forests by protecting and nurturing seed trees and wildlings already present in the area. With ANR, secondary and degraded forests grow faster than they would naturally. Because this method merely enhances already existing natural processes, it requires less labour than tree planting and there are no tree nursery costs. It can, therefore, be a low-cost way to restore forest ecosystems.

ANR and tree planting should not be regarded as mutually exclusive alternatives to forest restoration. More often than not, forest restoration combines protection and ANR together with some tree planting. A site survey can be used to determine if protection + ANR are sufficient to achieve restoration goals, whether they should be complemented with tree planting and, if so, how many trees should be planted.

Where is ANR appropriate?
Protection + ANR may be sufficient to bring about rapid and substantial forest restoration and biodiversity recovery where the density of natural regenerants exceeds 3,100 per hectare, representing more than 30 common tree species typical of the target forest type. Where the density of natural regenerants is lower or fewer tree species are represented, ANR should be used in combination with tree planting. Furthermore, intact forest should remain within a few kilometres of the proposed restoration site, to provide a seed source for re-establishment of climax forest tree species and seed-dispersing animals should remain fairly common.

ANR techniques
How can competition with weeds be reduced?

Weeding reduces competition between trees and herbaceous vegetation, increases tree survival and accelerates growth. Before weeding, clearly mark tree seedlings or saplings with brightly coloured poles, to make them more visible. This prevents accidentally trampling or cutting during weeding.

Ring weeding
Remove weeds, including their roots, with hand tools within a circle of 50 cm radius from around the base of all natural seedlings and saplings. Hand-pull weeds (wear gloves) close to small seedlings and saplings, as digging out weed roots with hand tools can damage their root system. Then, lay a thick mulch of the cut weeds around each seedling and sapling, leaving a gap of at least 3 cm between the mulch and the stem, as this helps prevent fungal infection. Where cut weeds do not yield sufficient mass of mulching material, use corrugated cardboard as mulch.

Weed pressing or lodging

Flatten all remaining herbaceous vegetation between the exposed natural regenerants, to remove shade, using a wooden board (130 x 15 cm). Attach a sturdy rope to both ends of the plank, making a loop, long enough to pass over your shoulders (attach shoulder pads for comfort). Lift the plank onto the weed canopy and step on it with full body weight to fold over the stems of grasses and herbs near the base. Repeat this action, moving forward in short steps. The weight of the plants should keep them bent down. This is particularly effective where the vegetation is dominated by soft grasses such as Imperata. Old, robust or cane-stemmed grasses (e.g. Phragmites, Saccharum, Thysanolaena spp. etc.) should not be pressed, because they can readily re-sprout from nodes along their stems. Pressing weeds is much easier than slashing them; one experienced person can press about 1,000 sq m per day.

Pressing is best carried out when the weeds are about 1 m tall or taller – shorter plants tend to spring back up shortly after pressing. Usually the best time to press grass is about two months after the rains start, because grass stems easily crimp (fold) at this time. But before pressing on a large scale, conduct a simple test on a small area. Press the grass and wait overnight. If the grass is starts to spring back up by the morning, then wait a few more weeks before trying again. Always press the weeds in the same direction. On slopes, press grasses downhill. If plants are pressed when they are wet, water on the leaves helps them to stick together, so that they are less likely to spring back up.

Pressing effectively uses the weeds’ own biomass to shade and kill them. Plants in the lower layers of the pressed mass of vegetation die, due to lack of light. Some plants may survive and grow back, but they do so much more slowly than if they had been slashed. Therefore, pressing does not have to be repeated as often as slashing. The pressed vegetation suppresses germination of weed seeds, by blocking light. It also protects the soil surface from erosion and adds nutrients to the soil, as the lower layers begin to decompose. Weed pressing opens up the restoration site, making it easier to move around and work with the young trees. It also helps to reduce the severity of fires. Pressed plants are a lot less flammable than erect ones, due to lack of air circulation within the pressed mass of vegetation. They do burn, but flame height is lower, so tree crowns are less likely to be scorched.
Where the density of natural regenerants is high, use of herbicides to clear weeds is not recommended, because it is very difficult to prevent the spray drifting onto the foliage of the natural regenerants.  

What about fertilizer?
Most tree seedlings and saplings up to about 1.5 m tall will respond well to fertilizer application, regardless of the soil fertility. Fertilizer application increases survival and accelerates growth and crown development. This brings about canopy closure and shades out of weeds sooner than if no fertilizer is applied and thus reduces labour costs for ring weeding and weed pressing. So although chemical fertilisers can be expensive, the costs are partly offset by long term savings in weeding costs. Organic fertilisers, such as manure, can be used as a cheaper alternative to chemical fertilisers. It is probably a waste of effort and expense to apply fertilizer to older saplings and tree stumps, since they would already have developed deep root systems.

Assisting the seed rain

Natural seed dispersal is a vital and free ecological service, which ensures re-colonization of restoration sites by climax forest tree species. So how can it be enhanced?
Artificial bird perches are, in theory, a rapid and cheap way of attracting birds and increasing the seed rain in restoration sites. Perches are usually 2-3 m-high posts, with cross-bars pointing in different directions. Although the seed rain is increased beneath perches seedling establishment increases only if the conditions for germination and seedling growth are favourable beneath the perches. Seeds can be predated, or young seedlings can be out-competed by herbaceous weeds, so weeding beneath perches is necessary, or perches should be used on sites with low weed density.
Although artificial perches attract birds, they do so less effectively than actual trees and shrubs, which provide the added benefit of shading out weeds, thus improving conditions for seedling establishment. Establishing structurally diverse vegetation, including fruiting shrubs or remnant trees, is the best way to attract seed-dispersing birds and animals, but it takes time. In the mean time, artificial bird perches can provide a decent stop-gap measure.

What are the limitations of ANR?
ANR acts solely on natural regenerants that are already present in deforested sites. Canopy cover can be achieved rapidly, but only where such regenerants are present at sufficiently high densities. Most of the trees that colonize such areas are of relatively few, common, light-demanding, pioneer species, which produce seeds that are dispersed by the wind or small birds. They represent only a small fraction of the tree species which grow in the target forest type. Where wildlife remains common, the ‘assisted’ trees will attract seed-dispersing animals, resulting in tree species recruitment. However, where large seed-dispersing animal species have become extirpated, planting large-seeded climax forest tree species may be the only way to transform the secondary forest, created by ANR, into climax forest.