Background information and definitions
Restoring or creating forest and woodland may provide important habitat for forest-dependant mammal species, particularly in disturbed or fragmented landscapes. Trees grow slowly and therefore the effects of forest restoration may not be evident for decades or even longer after restoration begins. Care must therefore be taken when interpreting the results of these studies.
Supporting evidence from individual studies
A replicated, site comparison study in 1999–2001 of riparian forest at a site in California, USA (Queheillalt & Morrison 2006) found that mammal species richness in restored riparian forest was similar to that in natural riparian forest. Mammal species richness in restored sites did not differ from that in natural sites during any season of sampling (data not reported). There was also no significant difference in species richness of small mammals (rodents and shrews) between restored (2–3 species) and natural (3–5 species) sites. Restoration, which included planting of woody riparian species, commenced between 1996 and 1998. Small mammals were surveyed between December 1999 and February 2001, using 16 Sherman live traps/ha. Other mammals were caught in larger live traps (cross section 7.6 × 8.9 cm) between November 1999 and April 2001.Study and other actions tested
A replicated, site comparison study in 2010–2012 of 16 riparian forest sites in California, USA (Derugin et al. 2016) found that restored riparian forest areas were visited more by carnivores than were remnant forests when restored areas were newly established, but not subsequently, whilst restored areas were not visited more frequently by black-tailed deer Odocoileus hemionus columbianus. More mammalian carnivore species were detected in young restored forests (3.4/plot) than in remnant forests (1.8/plot) but neither figure differed from that in old restored forests (2.1/plot). Coyotes Canis latrans made more visits to young restored forests than to remnant forests (data not presented). No differences were detected between visit rates to the three forest stages for raccoon Procyon lotor, bobcat Felis rufus or black-tailed deer. Five young restored forests (restored in 2003–2007), six old restored forests (restored in 1991–2001) and five natural forest remnants were sampled. Camera traps were operated over two consecutive years in December–March and May–July, starting in December 2010 and finishing in July 2012.Study and other actions tested
A site comparison study in 2013–2014 in a forest in Caldas department, Colombia (Ramírez-Mejía & Sánchez 2016) found that mammal species richness was similar in an area reforested with flooded gum Eucalyptus grandis compared to native forest, though there were differences in occurrence rates of individual species between forest types. Mammal species richness did not differ significantly between the reforested (9 species) and native forest (11 species) areas. Nine-banded armadillos Dasypus novemcinctus were recorded less frequently in the reforested site (10 records) than in native forest (30 records) as were South American coatis Nasua nasua (23 vs 48 records). Western mountain coatis Nasuella olivacea was recorded more frequently (43 records) in the reforested site than in native forest (10 records). There were no differences in the number of records of red-tailed squirrel Sciurus granatensis or dwarf red brocket Mazama rufina between forest types (data not reported). A 93-ha area, reforested in the 1960s, was compared with a 146-ha native forest block. Mammals were surveyed using four camera traps each in the two forest blocks, from September 2013 to February 2014.Study and other actions tested
A replicated study in 2010–2012 of 10 deciduous woodland sites in a protected area in central Italy (Sozio et al. 2016) found that forest regrowing on previously cultivated and/or grazed land had a greater abundance of hazel dormice Muscardinus avellanarius, and they had greater survival rates, than in coppiced forest. Peak abundance was higher in regrowing forest plots (17 dormice/plot) than in recent coppice (0–1/plot) and old coppice (1–7/plot). Monthly survival probability in regrowing forest (0.75) was higher than in old coppice (0.43). Too few dormice were recorded in young coppice to calculate survival. Forest type did not affect average litter size (regrowing forest: 4.5 young/litter; old coppice: 4.8 young/litter; no litters found in new coppice). Hazel dormice were surveyed within a grid of 36 tree-mounted wooden nest boxes/plot. Two recently coppiced plots (1–5 years since coppicing), three old coppice plots (20–30 years since coppicing) and two regrowing plots (formerly cultivated and/or grazed areas, unmanaged for 20 years) were sampled.Study and other actions tested
A replicated, site comparison study in 2002–2011 of 137 forest sites in New South Wales, Australia (Lindenmayer et al. 2017) found that replanted forest supported few common brushtail possums Trichosurus vulpecula or common ringtail possums Pseudocheirus peregrinus by 7–30 years after planting. The probability of a replanted site holding brushtail possums when surveyed 7–30 years after planting (0.02) was lower than that in old growth forest (0.44). For ringtail possums, the probability of occupancy in replanted forest 7–30 years after planting (0.07) was also lower than that in old growth forest (0.75). Greater tree cover in the surrounding area did not increase the probability of subsequent colonisation for either species (result presented as model coefficient). Sixty-five replanted forests and 72 old growth forests were surveyed. Most replanted forests were 7–30 years old and comprised local and exotic Australian plant species. Old growth forests were ≥200 years old. Marsupials were surveyed by spotlight, whilst walking at an average 3 km/h, 1–5 hours after dusk. At each site a 200-m transect was surveyed for 20 min. Sites were surveyed in winter 2002, 2003, 2008, 2009 and 2011.Study and other actions tested