Restore or create forest or woodland

Overall effectiveness category Awaiting assessment
Number of studies: 6
Share Tweet Email

View assessment score

How is the evidence assessed?

Effectiveness

not assessed
Certainty

not assessed
Harms

not assessed

Calculating overall effectiveness category

Background information and definitions

Woodland creation or restoration can be used to reverse the global loss of forest habitat, and forest-dependent species. However, restored forests can remain structurally different from older forests even 20 years after regeneration (Fuentes-Montemayor et al. 2015), which may in turn affect the butterfly and moth community which they can support. Additional management, such as thinning, may be necessary to improve the structural diversity of new forests, as part of the restoration process (Fuentes-Montemayor et al. 2015).

This action includes studies where either multiple actions have been used to restore or create forests, or where the specific action used is not clear, including natural regeneration on abandoned land. For studies of specific actions for creating or managing forests, see “Replant native vegetation”, “Replace non-native species of tree/shrub with native species”, “Clear or open patches in forests”, “Coppice woodland”, “Thin trees within forests”, “Create young plantations within mature woodland” and “Natural system modifications – Use prescribed fire to maintain or restore disturbance in forests”. For studies testing the natural regeneration of native forests after forestry/logging operations, see “Biological resource use – Encourage natural regeneration in former plantations or logged forest”.

_{Fuentes-Montemayor E., Peredo-Alvarez V.M., Watts K. & Park K.J. (2015) Are woodland creation schemes providing suitable resources for biodiversity? Woodland moths as a case study. Biodiversity and Conservation, 24, 3049–3070.}

Study locations

Key messages

Six studies evaluated the effects on butterflies and moths of restoring or creating forest or woodland. Two studies were in Brazil and one was in each of Cameroon, Mexico, Malaysia and Costa Rica.

COMMUNITY RESPONSE (6 STUDIES)

Community composition (4 studies): Three site comparison studies (including two replicated studies) in Mexico, Costa Rica and Brazil found that naturally generating or secondary forest had a different community composition of caterpillars, geometrid and arctiine moths and butterflies to replanted forest, oil palm plantations, pasture or remnant primary forest. One site comparison study in Brazil found that a 54-year-old restored forest had a higher proportion of fruit-feeding forest butterfly species than 11–22-year-old restored forests, and a similar community composition to a remnant forest.
Richness/diversity (6 studies): Three replicated, site comparison studies in Cameroon, Costa Rica and Brazil found that secondary forest had a similar species richness of butterflies and geometrid and arctiine moths to agroforestry plantations, pasture and remnant forest. Two of these studies also found that secondary forest had a greater species richness of butterflies and geometrid and arctiine moths than cropland or oil palm plantations. One of two site comparison studies (including one replicated study) in Brazil and Malaysia found that a 54-year-old restored forest had a lower species richness of fruit-feeding butterflies than 11–22-year-old restored forests. The other study found that 5–60-year-old restored forests had a greater species richness of butterflies than newly restored forests (<3-years-old), but restored forests had a lower species richness than primary forests. One site comparison study in Mexico found that a forest restored by natural regeneration had a similar diversity of caterpillars to a forest restored by planting.

POPULATION RESPONSE (3 STUDIES)

Abundance (3 studies): Two replicated, site comparison studies in Cameroon and Costa Rica found that secondary forest had a higher abundance of butterflies and geometrid and arctiine moths than cropland or oil palm plantations. One of these studies also found that secondary forest had a similar abundance of butterflies to coffee and cocoa agroforestry, and the other study also found that secondary forest had a lower abundance of geometrid and arctiine moths than primary forest. One site comparison study in Mexico found that a forest restored by natural regeneration had a similar abundance of caterpillars to a forest restored by planting.

BEHAVIOUR (0 STUDIES)

About key messages

Key messages provide a descriptive index to studies we have found that test this intervention.

Studies are not directly comparable or of equal value. When making decisions based on this evidence, you should consider factors such as study size, study design, reported metrics and relevance of the study to your situation, rather than simply counting the number of studies that support a particular interpretation.

Supporting evidence from individual studies

A replicated, site comparison study in 2003–2004 in 24 sites in a forested landscape in the Korup Region, Cameroon (Bobo et al. 2006) found that secondary forest had a similar abundance and species richness of butterflies to agroforestry sites, but a higher abundance and species richness than cropland. The abundance of butterflies was similar in secondary forest (310 individuals) and agroforestry (412 individuals), but was higher than in near-primary forest (270 individuals) and cropland (175 individuals). Butterfly species richness was similar in secondary forest (44 species), agroforestry (36 species) and near-primary forest (35 species), but higher than in cropland (17 species). Six out of 119 species were more abundant in secondary forest than in both agroforestry and cropland. However, seven species were less abundant in secondary forest than agroforestry or cropland. Two species were less abundant in secondary forest, agroforestry and cropland than in near-primary forest (see paper for details). Six 50-m radius sample sites, >500 m apart, were established in each of four habitat types: secondary forest, cocoa/coffee plantation (agroforestry), near-primary forest and annual cropland. From late December 2003–early March 2004, butterflies were caught in three baited, cylindrical gauze-traps/site, set for nine days/site.
Study and other actions tested
Referenced paper
Bobo K.S., Waltert M., Fermon H., Njokagbor J. & Muhlenberg M. (2006) From forest to farmland: butterfly diversity and habitat associations along a gradient of forest conversion in Southwestern Cameroon. Journal of Insect Conservation, 10, 29-42.
A site comparison study in 2008 in two forest sites in Jalisco, Mexico (Hernandez et al. 2014) found that a naturally regenerated forest and a forest restored by planting native trees had a similar diversity and abundance of caterpillars, but the species present at the two sites differed. In a forest which had regenerated naturally, the diversity and abundance of caterpillars (103 individuals) was similar to a forest restored by planting (119 individuals; diversity data presented as model results). However, only 27% of species were found at both sites. Three conserved forest sites had an average abundance of 159 caterpillars/plot (statistical significance not assessed). One 1-ha abandoned pasture was allowed to regenerate naturally from 1992. In 2002, a second 1-ha abandoned pasture was restored by planting 39 native tree species which were shared with the naturally regenerating site. Three conserved forest sites were also surveyed for comparison. From July–November 2008, caterpillars were sampled five times along four parallel 20 × 2-m transects/site, 20 m apart. All leaves in trees up to 2 m high were searched for caterpillars, and in trees >2 m high three branches/tree were searched. Caterpillars were reared in the laboratory to identify the adults.
Study and other actions tested
Referenced paper
Hernandez Y., Boege K., Lindig-Cisneros R. & del-Val E. (2014) Lepidopteran herbivory in restored and successional sites in a tropical dry forest. The Southwestern Naturalist, 59, 66-74.
A site comparison study in 2009 in four forest fragments in São Paulo, Brazil (Sant’Anna et al. 2014) found that an old restored forest had fewer species and a lower diversity of fruit-feeding butterflies (Nymphalidae) than younger restored forests, but the proportion of forest species and overall species community at the older site was the most similar to a remnant forest. A 54-year-old restored forest had fewer species (25) and lower diversity of butterflies than 11–22-year-old restored forests (29–35 species) or remnant forest (28 species; diversity data presented as model results). However, the proportion of forest species in the 54-year-old forest (79% of individuals; 72% of species) was higher than in the younger forests (36–46% of individuals; 60–65% of species), and more similar to the remnant forest (92% of individuals; 89% of species). The species community in the old forest was most similar to the remnant forest (data presented as model results). In 1955, 1987 and 1998, three areas of forest (30–50 ha) were restored using seedlings of >70 native tree species and some non-native species. From January–April 2009, butterflies were surveyed in three plots (200 m apart) in each restored forest, and in a 245-ha remnant forest. At each plot, five baited Van Someren-Rydon traps were placed 30 m apart. Traps were left open for 8 days/month. Every 48 hours, butterflies were identified and released, and bait was replaced. Butterfly species were classified according to habitat preference as “forest”, “edge” or “grassland”.
Study and other actions tested
Referenced paper
Sant'Anna C.L.B., Ribeiro D.B., Garcia L.C. & Freitas A.V.L. (2014) Fruit-feeding butterfly communities are influenced by restoration age in tropical forests. Restoration Ecology, 22, 480-485.
A replicated, site comparison study in 2003–2006 around a tropical rainforest reserve in Sarawak, Malaysia (Itioka et al. 2015) found that butterfly species richness was higher in older forest regeneration plots than in newly regenerating plots, but that regenerating areas had lower species richness than primary forest. Butterfly species richness was higher at sites which had been regenerating naturally for >5 years (5–13 years: 10–18 species; 20–60 years: 12–22 species) than at newly regenerating sites (<3 years: 6–7 species). However, species richness in all regenerating sites was lower than in isolated (20–40 species) or intact (48–66 species) primary rainforest. In August 2003, twenty-one open plots (two × 100 m² each) on the edge of five types of forest stand (2,772–4,917 m²) were selected. Six plots were next to old regenerating forest, where 20–60 years had passed since the land was last cultivated; three were next to young regenerating forest where 5–13 years had passed since cultivation; three were next to newly regenerating forest where one year had passed since cultivation; six were next to isolated primary forest stands, and three were next to intact primary forest. Butterflies were surveyed twice/plot in August 2003, September 2003, January 2005 and June 2006.
Study and other actions tested
Referenced paper
Itioka T., Takano K.T., Kishimoto-Yamada K., Tzuchiya T., Ohshima Y., Katsuyama R., Yago M., Yata O., Nakagawa M. & Nakashizuka T. (2015) Chronosequential changes in species richness of forest-edge-dwelling butterflies during forest restoration after swidden cultivation in a humid tropical rainforest region in Borneo. Journal of Forest Research, 20, 125-134.
A replicated, site comparison study in 2013 in 15 forest sites in southwest Costa Rica (Alonso-Rodríguez et al. 2017) found that secondary forests had a greater abundance and species richness of geometrid (Geometridae) and arctiine (Arctiinae) moths than oil palm plantations, and a similar species richness but lower abundance than old-growth forest. In young secondary forest, the species richness of both geometrid (90 species) and arctiine (96 species) moths was higher than in oil palm plantations (geometrids: 31; arctiines: 35 species), but not significantly different to old-growth forest (geometrids: 113; arctiines: 81 species). The abundance of geometrid moths was higher in secondary forest (314 individuals) than in oil palm (135 individuals), but lower than in old-growth forest (570 individuals). The abundance of arctiine moths was similar between habitat types (secondary forest: 668; oil palm: 529; old-growth forest: 581 individuals). Species composition was different in the three habitats (see paper for details). Fifteen sites, >200 m apart, were selected: five 3–10-year-old secondary forests; five even-aged oil palm plantations (>1 ha); and five old-growth primary or 80-year-old secondary forests. From February–July 2013, moths were sampled overnight once/month, using an 8 W, UV-emitting funnel trap installed 1–2 m above ground at up to three sites/night in different habitats.
Study and other actions tested
Referenced paper
Alonso-Rodriguez A., Finegan B. & Fiedler K. (2017) Neotropical moth assemblages degrade due to oil palm expansion. Biodiversity and Conservation, 26, 2295-2326.
A replicated, site comparison study in 2011 in a fragmented forest in Paraná, Brazil (Shuey et al. 2017) found that regenerating and replanted forest plots had a similar species richness of butterflies to both grazed pasture and remnant forest, but the species present differed between habitats. The number of butterfly species in regenerating (69 species) and replanted (47–102 species) forest was not significantly different from in pastures (52–59 species) or remnant forest (57–79 species). However, out of 213 butterfly species recorded, 33 were found only on restored sites (regenerating or replanted), compared to 18 species unique to pastures and 66 species unique to remnant forests. Eight sites, all >40 ha, were studied: one former pasture ungrazed for 14 years and naturally regenerating from the surrounding remnant forest, two former pastures planted with 15–20 species of native trees 12–14 years before the study, two grazed open pastures, and three intact forest remnants. In January, March and April 2011, butterflies were sampled once/month. Four baited butterfly traps were placed 1–2 m above ground, 50 m apart, in the centre of each plot, for three consecutive days/month, and checked daily. In addition, butterflies were counted on two 1-hour transects/month at each site.
Study and other actions tested
Referenced paper
Shuey J., Labus P., Carneiro E., Silva Dias F.M., Leite L.A.R. & Mielke O.H.H. (2017) Butterfly communities respond to structural changes in forest restorations and regeneration in lowland Atlantic Forest, Parana, Brazil. Journal of Insect Conservation, 21, 545-557.

Please cite as:

Bladon A.J., Smith R.K. & Sutherland W.J. (2022) Butterfly and Moth Conservation: Global Evidence for the Effects of Interventions for butterflies and moths. Conservation Evidence Series Synopsis. University of Cambridge, Cambridge, UK.