Action

Other biodiversity: Restore habitat along watercourses

How is the evidence assessed?
  • Effectiveness
    60%
  • Certainty
    50%
  • Harms
    15%

Study locations

Key messages

Amphibians (1 study): One replicated site comparison from the USA found similar numbers of amphibian species in restored and remnant sites.

Birds (8 studies): Two replicated site comparisons from Spain and the USA found similar numbers of bird species in restored and remnant sites. Two replicated site comparisons from the USA found fewer bird species in restored riparian sites, compared to remnant sites. One replicated site comparison from Spain found similar numbers of birds and bird species in restored contaminated sites and uncontaminated sites. One replicated site comparison from the USA found that an endangered bird nested in restored sites, and had similar nesting success in restored and remnant sites. One replicated site comparison from the USA found that bird populations increased with the area of restored habitat in the landscape, in some comparisons. One replicated site comparison from the USA found similar levels of nest parasitism in restored and remnant sites.

Fish (1 study): One before-and-after site comparison from the USA found differences in fish communities, before and after changing river flow.

Invertebrates (3 studies): One replicated site comparison from the USA found fewer native ants, but similar numbers of invasive ants, in restored sites, compared to remnant sites. One before-and-after site comparison from the USA found similar numbers of freshwater invertebrates in restored and reference sites, after restoration. One replicated, before-and-after study from the USA found more invertebrates and invertebrate species in plots with added gravel, compared to plots without added gravel, in some comparisons. One replicated before-and-after study from France found relatively more alien species after restoring river flow.

Mammals (2 studies): Two replicated site comparisons from the USA found similar numbers of mammal species in restored and remnant sites.

Plants (11 studies)

  • Abundance (6 studies): Four replicated site comparisons from Spain and the USA found lower plant cover in restored sites, compared to remnant sites. One of these studies also found higher cover of exotic plants, but another one did not. One replicated, paired site comparison from the USA found similar numbers of flowers in restored and remnant sites. One replicated site comparison from the USA found more seeds, but fewer native seed, in orchards next to restored riparian habitats, compared to orchards next to remnant habitats. One replicated site comparison from the USA found similar exotic plant cover in remnant and restored forests.
  • Diversity (6 studies): Two replicated studies from the USA found fewer native plant species in restored forests, compared to remnant forests. One of these studies also found more exotic species, but another one did not. One replicated site comparison from the USA found more plant species in restored sites, compared to remnant sites. One replicated, paired site comparison from the USA found similar numbers of flower species in restored and remnant sites. One replicated site comparison from the USA found fewer seed species and native seed species in orchards next to restored riparian habitats, compared to remnant riparian habitats. One controlled study from the USA found different plant communities in restored and unrestored habitats.
  • Survival (2 studies): One replicated study from the USA found that about one-third of planted willows survived for one year. One site comparison from the USA found that some species survived after planting, as part of riparian restoration, but others did not.
  • Habitat suitability (1 study): One replicated site comparison from the USA found that vegetation at one of five sites met the criteria for Bell’s Vireo nesting habitat.
  • Size (1 study): One replicated site comparison from the USA found smaller elderberry plants in restored sites.

Reptiles (1 study): One replicated site comparison from the USA found similar numbers of reptile species in remnant and restored sites.

Implementation options (7 studies)

  • Plants (3 studies): One study from the USA found more tree, shrub, vine, and perennial species, higher canopy cover, and higher native tree cover, in older restored plots, compared to younger restored plots, but this study also found fewer annual plant species, lower vegetation cover, lower annual forb cover, and lower grass cover. One study from the USA found an increase in native species and overstorey cover in restored sites, over time, but it found similar numbers of species and overstorey cover in sites planted at different densities. One study from the USA found that willow cuttings planted on the stream bottom had a higher survival rate than those planted on the streambank or terrace.
  • Birds (3 studies): Three studies from the USA found more birds or bird species in older restored plots, compared to younger restored plots. One of these studies also found that the populations of some bird species increased with tree-planting density.

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

  1. A replicated study in 1987–1988 in three riparian meadows in the northern Sierra Nevada, California, USA, found that Geyer willow Salix geyeriana cuttings planted on the stream bottom had a higher survival rate than those planted in the streambank or terrace. Plants: Out of 2,700 cuttings, 32% survived in 1987 and 26% survived in 1988. Implementation options: Willow cuttings planted on the stream bottom had a higher survival rate (82%), compared to those planted on the streambank (34%) or the stream terrace (3%). Methods: Geyer willow cuttings were planted (30 cm depth, in May 1987) in three locations (stream bottom, streambank, and stream terrace). At each site, cuttings (over two years old; diameter: 10.5 mm; length: 42.3 cm) were planted along thirty transects perpendicular to the stream, crossing the stream, and extending 10 m from the top of both banks. Survival was measured in September 1987 and 1988.

    Study and other actions tested
  2. A replicated site comparison in 1989–1993 in riparian forests along the San Luis Rey and San Diego Rivers, California, USA, found that the endangered Bell’s Vireo Vireo bellii pusillus nested in restored sites, and similar numbers of fledglings were found in restored and remnant sites, even though four of five sites did not meet modelled criteria for nesting habitat. Birds: Similar numbers of fledglings were found in restored and remnant sites, at least along the San Luis Rey River (1.9–4 vs 1–1.4 fledglings/nest; 3–4 vs 1.6–2.4 fledglings/breeding pair), and nests were established within 1–5 years. Plants: By 1993, the vegetation at one of five sites met the modelled criteria (e.g., height and cover) for Bell’s Vireo nesting habitat. Methods: Five restored sites (3–13 ha) were surveyed along the San Luis Rey River (three sites, established in 1989) and the San Diego River (two sites, established in 1990). The sites were planted with willows and/or other species, based on the natural habitat of Bell’s Vireo. These sites were compared to natural habitats along these rivers. Birds were surveyed every 1–2 weeks. Nests were surveyed between mid-March and August.

    Study and other actions tested
  3. A replicated site comparison in 1990–1995 in restored riparian forests along the Sacramento River, California, USA, found that some species survived after planting, as part of riparian restoration, but others did not. Plants: Box elder Acer negundo had 75–100% survival after two years (planted in five sites). Oregon ash Fraxinus latifolia had 0% survival in two of three sites but had 100% survival in one of three sites. Western sycamore Platanus racemosa had 53–100% survival in four of five sites and 0% in one of five sites. Fremont’s cotton wood Populus fremontii had 14–66% survival in three of four sites and 0% in one of four sites. Valley oak Quercus labata had 18–100% survival (planted in five sites). Californian Rose Rosa californica had 21–100% survival (planted in three sites). Sandbar willow Salix exigua had 5–88% survival (planted in two sites). Goodding’s willow Salix gooddingii had 17–94% survival (planted in two sites). Arroyo willow Salix lasiolepis had 26–100% survival (planted in four sites). Blue elderberry Sambucus mexicana had 8–68% survival (planted in five sites). Methods: Sites were on flood plains and were ≥200 ha. Seven sites were selected, with varying planting dates: Lohman (1994), Princeton (1992), River (1990), Sam (1991), Vista 1 (1992), Vista 2 (1993), and Vista 3 (1994). All sites had previously been cleared of vegetation. All species were collected from natural stands. Plants were protected by sleeves (35 cm height). Sites were irrigated and weeds were controlled through monthly spraying. Survival and height were measured in 405 m2 plots (the number of plots varied to cover 5–10% of each site). Planting rows were sampled in Lohman and Vista 3 plots. Plants were sampled at the end of each growing season.

    Study and other actions tested
  4. A replicated site comparison in 1991–1999 along rivers in southeast Spain found lower herb and shrub cover, and lower liana frequency, in restored sites, compared to undisturbed sites. Plants: Lower shrub and herb cover was found in restored plots, compared to undisturbed plots (shrubs: 20–39% vs 79%; herbs: 0–2% vs 4–5%), but no difference was found in the cover of trees, lianas, or annuals (trees: 1.3–1.5% vs 1.5–2.3%; lianas: 0–0.05% vs 0.5%; annuals: 0.03–0.15% vs 0.09–0.12%). Tree frequency was higher in restored sites, compared to undisturbed sites, in 1991 (8% vs 4%), but not in 1993–1999 (4–7% vs 4%). Liana frequency was lower in restored sites (0–2% vs 5–7%). The frequency of shrubs, herbs, and annuals was similar in restored and undisturbed sites (shrubs: 60–71% vs 60%; herbs: 1–2% vs 2–3%; annuals: 2–5% vs 2–3%). Methods: Two riparian forest sites had two plots each: one restored, and one undisturbed (120 x 3 m each). Restored plots were planted with root cuttings and seeds (broom Retama sphaerocarpa) from undisturbed forests in December 1991, and plants were monitored and irrigated weekly (in the first summer) for a few months. Plants were surveyed on transects (October: 1993, 1995, 1999; September: 1997).

    Study and other actions tested
  5. A replicated site comparison in 1989–2001 in 23 riparian forest sites along the Sacramento River, California, USA, found that vegetation cover, native species cover, and the number of native species was lower in restored sites, compared to remnant sites. Plants: In the understorey, the cover of total and native vegetation was lower in restored sites, compared to remnant sites (total: 40–50% vs 73%; native: 2–10% vs 51%), as was the number of native species (3–5 vs 12), but the cover and number of exotic species was no different (15–17 vs 10 species; 40–42% vs 21% cover). In the overstorey, total cover was lower in remnant sites (0–29% vs 82%). Methods: Native riparian tree and shrub species were planted (520–1,300 tree/ha) on former farmland along 150 km of the Sacramento River (15 sites restored in 1989–1996; three sites restored in 2000). Plants were surveyed in quadrats (1 x 1 m) in spring 2001 at restored sites and five remnant sites.

    Study and other actions tested
  6. A replicated, before-and-after study in 1996–2000 at seven salmon-spawning sites along the lower Mokelumne River, in the Central Valley, California, USA, found more macroinvertebrates, 10 weeks after gravel addition, compared to before gravel addition. Invertebrates: Similar numbers of macroinvertebrate species and individuals were found in sites with or without gravel addition, by week 6 (0–30 vs 21 species; 350,000 vs 300,000 individuals/m3). However, by week 10, more individuals were found in sites with gravel addition (290,000 vs 150,000 individuals/m3). Methods: Gravel was added to seven sites, between 15 August and 15 September, in 1996–2000 (different sites in different years, approximately 30 x 65 m sites, each of which was approximately 1–2% of remaining Chinook salmon Oncorhynchus tschawytscha spawning habitat). Macroinvertebrates were collected from the substrate (15 cm depth) with a stream sampler (bottom open area: 0.086 m2) and a dolphin bucket (368 μm) in flowing water (0.25–1 m/s, <60 cm depth), one week before and every two weeks after gravel addition, in September–January 1996–2000.

    Study and other actions tested
  7. A replicated site comparison in 1993–2003 on ten sites along the Sacramento River, California, USA, found that 13 of 20 bird species were increasing on plots revegetated as part of riparian reforestation, although abundances did not reach that of plots of remnant forest. Birds: Thirteen of 20 bird species were increasing on plots revegetated as part of riparian reforestation, although abundances did not reach that of plots of remnant forest. Nine of these were also increasing on the remnant plots, with a further three only increasing in remnants. Three species were stable on both plot types and one, lazuli bunting Passerina amoena, declined on both (mirroring a regional trend). Methods: Restoration focused on revegetating with native trees, shrubs and understory plants, and restoring natural river processes.

    Study and other actions tested
  8. A replicated site comparison in 1996–2001 in five riparian sites in Carmel-by-the-Sea, California, USA, found more plant species, fewer bird species, and similar numbers of amphibian, mammal, and reptile species, in restored forests, compared to mature forests. Amphibians, Mammals, and Reptiles: Similar numbers of amphibian (3), mammal (16), and reptile (4) species were found in restored plots and mature plots. Birds: Fewer bird species were found in restored plots, compared to mature plots, in summer (26–29 vs 48–52), but not in spring (53–56 vs 62–69), fall (17–23 vs 26–32), or winter (22–33 vs 40–41). Restored sites had fewer breeding bird species (4–7 vs 28–33). Plants: More plant species were found in restored forests, compared to mature forests (15–26 vs 8–11). Methods: In 1996–1998, 15 ha of woody riparian species and 2.4 ha of freshwater wetland species were planted. Three restored sites (17,400 m2, 28,000 m2, 65,000 m2) were compared to two mature riparian forest sites (47,420 m2 and 24,780 m2). Vegetation was sampled using transects (30 m) in April, August, October, and January 1999–2000. Amphibians and reptiles were sampled using pitfall traps (May–August 2000) and visual surveys (25 x 25 m area). Bird species were identified in ten-minute point counts (25 m radius, twice/season, March 2000–February 2001) and on transects (1.5 km/hr for 1–2.5 hours). Mammals were captured in live traps (7.6 x 8.9 x 22.9 cm and 7.6 x 8.93 x 30.5 cm), marked, and released (November 1999–April 2001, except spring 2000).

    Study and other actions tested
  9. A replicated site comparison in 2005–2006 in 46 riparian sites in the Central Valley, California, USA (same study as (10)), found smaller elderberry plants Sambucus mexicana, and fewer native ants, but similar numbers of non-native ants, in restored sites, compared to natural sites. Invertebrates: Fewer native ants but similar numbers of Argentine ants were found in restored sites. Plants: Smaller elderberry plants were found in restored sites. Implementation options: The number of Longhorn beetles Desmocerus californicus increased with site size and age. Elderberry seedlings grew faster than plant transplants. Elderberry plants grew slower in older sites. Methods: Thirty restored sites (with <30 planted elderberry plants) were compared with 16 natural sites (within 20 km). Restored sites were surveyed in July–early November 2005 and February–April 2006 and natural sites in April–September 2006. Restored sites were 24% of the size of natural sites.

    Study and other actions tested
  10. A replicated site comparison in 2005–2006 in 46 riparian sites in the Central Valley, California, USA (same study as (9)), found that restored sites had lower canopy cover, stem diameter, and height than natural sites. Plants: Elderberry canopy size (400 vs 272 cm), stem diameter (8 vs 5 cm), and height (428 cm vs 320 cm) were larger in natural sites, compared to restored sites. Methods: Thirty restored sites (urban: 19; agricultural: 11; all with <30 planted elderberry plants; 2–15 years old) and 16 natural sites (within 20 km of restored sites) were compared. Restored sites were surveyed in July–early November 2005 and August–October 2006 and natural sites in April–September 2006. Restored sites were 24% of the size of natural sites. Growth rate was measured for 30 shrubs at each restored site (growth rate for natural sites came from a previous study).

    Study and other actions tested
  11. A before-and-after site comparison in 2001–2003 in a grazed riparian meadow in Bagley Valley Creek, Sierra Nevada, California, USA, found different communities of freshwater invertebrates in a restored site, compared to two reference sites. Invertebrates: Before it was restored, the restored site had fewer mayfly, stonefly, and caddisfly taxa (2000: 6–9 fewer taxa) than two reference sites (a 10-year-old restored site and a similar site with less disturbance). After it was restored, it had similar or higher numbers than the two reference sites (no data provided). Methods: To restore the site, a new channel was constructed (rocks, erosion control fabric), Salix spp. willow trees were planted, and gullies and roads in the meadow and its watershed were rehabilitated, in 2001. Invertebrates were collected from randomly selected riffle habitats in the water (three 30 x 30 cm sampling areas; D-frame net; 250 μm mesh; 30 cm width; three samples/site). Samples were collected before restoration (1999 and 2000) and after (2002 and 2003).

    Study and other actions tested
  12. A replicated site comparison in 1998–2003 in 17 riparian sites along the Sacramento River in California, USA, found that the abundance of some bird species increased with the cover of restored area in the landscape. Birds: Abundance increased with the cover of restored habitat in the landscape, for six of seven species. Implementation options: Species abundance increased (15–51%) with age. The abundance of Nuttall’s woodpecker, ash-throated flycatcher, and black-headed grosbeak increased with the number of tree species planted. For three species, abundance increased with the planting density of tree species (western wood-pewee: abundance increased by a factor of 1.9 for each additional 100 Salix willows planted/ha; Bewick’s wren and spotted towhee: abundance increased by a factor of about 1.04 for each additional 100 valley oak trees Quercus lobata/ha). However, Bewick’s wren abundance decreased by a factor of 0.84 for each additional 100 cottonwood Populus fremontii trees/ha. Methods: Restored sites (4–74 ha each; former farmland; adjacent to remnant riparian forest) were disked, burned, furrowed, levelled, and sprayed with herbicides, and trees and shrubs were planted. Birds were surveyed with point counts (45 points, 50 m radius, 200 m apart, during the breeding season, from dawn until 4 h after sunrise, twice/year for 3–10 years; 5 minutes/point).

    Study and other actions tested
  13. A replicated site comparison in 1991–2004 in 26 riparian sites along the Sacramento River, California, USA, found fewer seed species and native seed species, more seeds in total, but fewer native seeds, in orchards next to restored habitats, compared to orchards next to remnant habitats. Plants: Fewer species (36–50 vs 68) and native species (6–10 vs 13) were found in orchards next to restored habitats. More seeds were found next to restored habitats (data reported in log units), but fewer native seeds were found, and there was no difference in invasive seeds. Implementation options: More seeds were found next to older restored habitats, compared to younger (data not reported). Methods: Soil samples were collected from 26 walnut plots, 0–5.6 km from restored riparian, remnant riparian, or agricultural habitats. Restored sites were formerly farmland. Restoration included disking, burning, furrowing, levelling, and spraying with herbicide, and replanting. On each walnut farm, soil samples (10 cm depth) were collected from seven points adjacent to restored or remnant forest and nine points within the walnut orchard, in March 2004. Seeds were germinated and identified in a greenhouse.

    Study and other actions tested
  14. A replicated, before-and-after site comparison in 2011 in 102 riparian forest sites in California, USA, found that riparian vegetation changed over time in restored sites, for 16 of 21 measurements. Implementation options: The following metrics increased over time: species of trees, perennial plants, and shrubs and vines; density of woody vegetation, native trees, and native and exotic shrubs and vines; absolute cover of the total canopy, native tree canopy, ground cover, exposed roots, and litter (data reported as model results). The following metrics decreased over time: absolute cover of total vegetation; relative cover of annual grasses and forbs; species of annual herbaceous plants. The number of exotic tree species and the cover of native and exotic perennial grasses and forbs did not change over time. Methods: A total of 102 riparian sites from three coastal counties (Marin, Mendocino, and Sonoma) were surveyed (restored: 89 sites, 0–39 years after restoration; non-restored: 13 sites). Restoration involved willow Salix planting. Vegetation cover was estimated using a Daubenmire Frame (20 x 50 cm). Canopy density was measured with a spherical densitometer.

    Study and other actions tested
  15. A replicated, before-and-after site comparison in 1989–2008 in riparian forests along the Sacramento River, California, USA, found fewer native species, with lower ground and canopy cover, and more exotic species in restored sites, compared to remnant sites. Plants: Fewer native species (5–7 vs 10 species; 21–32% vs 65% relative cover, 48–56% vs 87% frequency) and more exotic species (15–16 vs 9 exotic species; 79–67% vs 34% relative cover; 91–84% vs 56% frequency) were found in the 15 sites planted with overstorey species, compared to forest remnants. Fewer native species (4–6 vs 10 species; 16–19% vs 65% relative cover; 41–42% vs 87% frequency) and lower overstorey cover (31–39% vs 79%) were found in the 20 sites planted in 1997–2003, compared to forest remnants, in some comparisons. Implementation options: Between 2001 and 2007, increases in native species (5 vs 7 species) and overstorey cover (29% vs 60%) were found in the 15 sites planted with overstorey species in 1989–1996. Similar numbers of native species (5–6 species, 16–29% relative cover; 41–65% frequency) and similar overstorey cover (31–39%) were found in plots planted with understorey species at high or low densities in 1997–2003. Lower overstorey cover (31% vs 64) was found in plots planted with both overstorey and understorey species, compared to overstorey species, in one of two comparisons (31% vs 64%). Methods: Native overstorey species (trees and shrubs) were planted on 15 sites in 1989–1996 and 20 sites (14 of which were also planted with understory species, at high or low densities: herbs, vines, grasses, and low shrubs) in 1997–2003 (5–60 ha sites; 530–1,300 plants/ha; disked, planted, mown, irrigated, and weeds controlled for three years). Plants were surveyed in plots planted with overstorey species (in 2001 and 2007) or non-woody vegetation (2007) and in 10 forest remnants (15–20 ha, five sites in 2001, five sites in 2008). Vegetation was surveyed in quadrats (1 x 1 m, 20–80 quadrats/site) along a transect (part of a 40 x 80 m grid).

    Study and other actions tested
  16. A replicated, paired site comparison in 2003 in 10 riparian sites along the Sacramento River in California, USA, found similar numbers of flowers and flower species in restored and remnant forest sites. Plants: Similar numbers of flowers (401–2,458 vs 317–1,668) and flower species (37–47 vs 21–36) were found in restored and remnant sites, but different communities were found (data reported as relative Sørensen index). Methods: Each of five restored sites was paired with a remnant site (5.5–10 km apart). Plots within sites (1 ha) were 0.5–3.7 km apart. Restored sites were previously walnut and almond orchards (6 years before sampling) and were planted with similar vegetation to remnant sites (maple Acer spp., oak Quercus spp., willow Salix spp., and grass). Flowers were sampled in each plot, in February-August 2003 (60 quadrats, 0.25 x 4 m).

    Study and other actions tested
  17. A controlled study in 2000–2008 along a stream on a farm in the Central Valley, California, USA, found different plant communities in a restored area, compared to an unrestored area. Plants: Different plant communities were found in the restored and unrestored areas (data reported as ordination results: restoration explained 12% of the variation in plant communities). Methods: Part of the streambank was graded to create a floodplain (4 m width) and planted with native perennial grasses, sedges, forbs, shrubs, and trees. Herbaceous biomass was collected in the restored area and the unrestored area in October 2007 and April–May 2008 (0.25 x 0.5 m plots).

    Study and other actions tested
  18. A before-and-after site comparison in 1991–2008 in six sites along the Putah Creek, California, USA, found different fish communities before and after a change in river flow. Fish: Different communities were found 6 km and 21 km from a dam, after the change, but no differences were found 0, 16, 25, and 30 km from the dam. Methods: Habitat was restored through a change in hydrology. Three-day pulses, between 15 February and 31 March, followed by a month of higher flows, were used to initiate spawning. Five-day pulses (in November or December) were used to promote salmon Oncorhynchus tshawytscha migration. Six sites at different distances from a dam (0, 6, 16, 21, 25, or 30 km) were sampled annually in September and October for eight years before and nine years after the change in the flow (1991–2008). Fish were captured through electroshocking.

    Study and other actions tested
  19. A replicated site comparison in 1989–2008 in riparian forests in the Sacramento and San Joaquin river valleys, California, USA, found fewer bird species, but similar numbers of birds, in restored riparian forests, compared to remnant riparian forests. Birds: Fewer bird species were found in restored sites, compared to remnant sites (permanent species: 18 vs 45; all species: 33 vs 76), but similar numbers of overwintering bird species were found (15 vs 23). Similar numbers of birds were found in restored and remnant sites (all species: 39–40; permanent: 12–18; overwintering: 28–29). Methods: Forest was restored on three sites (Sacramento: two sites, 66–86 ha, planted in 1989–1992; San Joaquin: one site, 8 ha, planted in 2002–2003), by planting trees, shrubs, and grasses. Remnant forest was used for comparison (Sacramento: two sites, 25–45 ha; San Joaquin: three sites, 6–7 ha). Birds were captured in mist nests (12 x 2.5 m, 30 mm mesh; in November, December–January, and January–February 2003–2008). Sites were sampled for at least three years. Captured birds were ringed and re-sighted in November–February (Sacramento: 2004–2005 and 2006–2007; San Joaquin: 2004–2005). Bird abundance (birds captured/100 mist net hours), diversity (Shannon index), evenness, and richness were recorded.

    Study and other actions tested
  20. A replicated site comparison in 2001–2009 in riparian forest along the Guadiamar and Alcarayón Rivers, Spain, found a similar number of birds and bird species in restored sites and uncontaminated sites. Birds: Similar numbers of birds and bird species were found in restored and uncontaminated sites (24 vs 20 species; 76 vs 87 individuals). Implementation options: More birds were found eight years after restoration, compared to one year (2009 vs 2001: 102 vs 67 individuals/10 ha), but there was not a significant difference in the number of bird species (71 vs 45). Methods: Riparian areas along the Guadiamar River were contaminated by heavy metals from a mine accident in 1998. Sites were restored from 1998–2001 by planting native trees and shrubs and removing exotic and cultivated plants. Birds were surveyed along seven transects (8 km long and 2 km apart) in 2001–2006 and 2009 (winter: 15 October–15 February; breeding season: 15 April–15 June) in restored sites and a non-contaminated site (10 km away, along the Alcarayón River, in 2004–2006). These surveys were compared with data from before the contamination (bird atlas).

    Study and other actions tested
  21. A replicated, before-and-after study in 2003–2008 in 36 riparian sites in France found relatively more alien species, but fewer still-water species, after restoring river flow, compared to before. Invertebrates: Relatively more alien species were found after restoration, compared to before (4% vs 2% of functional diversity), but there were relatively fewer still-water (lentic) species (64% vs 72%) and similar numbers of flowing-water (lotic) species (31% vs 26%). Methods: There were three types of restoration: increasing flow (6 sites), dredging (6 sites), and reconnecting sites to the main river through dredging, (8 sites). Another sixteen sites had no restoration activities. Macroinvertebrates were sampled with four 0.25 x 0.25 m quadrats and nets (500 μm mesh) along a 30 m stretch in each site in spring and summer, one year before restoration and two years after.

    Study and other actions tested
  22. A replicated site comparison in 2002–2012 in 21 riparian sites in the Central Valley, California, USA, found similar amounts of parasitism by the cowbird Molothrus ater in restored or remnant forest sites. Birds: Similar amounts of parasitism were found in restored or remnant forest sites. Lower parasitism rates were found for spotted towhee Pipilo maculatus (Sacramento River: 26% vs 47%) and red-winged blackbird Agelaius phoeniceus (San Joaquin River: 0% vs 25%) in restored sites, compared to remnant sites, in some comparisons. Methods: Restored sites were formerly farmland, and restoration included planting. Bird nests were observed every three days in April–July (Sacramento River: 1993–2003; San Joaquin River: 2007–2009).

    Study and other actions tested
  23. A replicated site comparison in 2004–2009 in 33 restored riparian sites in the Central Valley and North Coast, California, USA, found more bird species in older restored plots, compared to younger restored plots. Implementation options: The number of bird species increased as restoration sites matured. For each year after restoration, the number of bird species increased by 0.4 in the Central Valley and 0.5 in the North Coast regions. Methods: Bird surveys were conducted in April–June in restored riparian plots (0.33–10 acres; 0–20 years after restoration; Central Valley: 18 sites; North Coast: 15 sites). Restoration included excluding grazers and planting native riparian vegetation. In the Central Valley, there were two surveys (152 points, 200 m apart) in the breeding seasons in 2004–2008. Birds heard and seen within 50 m of the points were recorded. On the North Coast, there were 2–3 surveys (area searches, 0.33–10 acres), at least 10 days apart, for 20 mins each, in 2001–2002, 2004–2005, and 2009.

    Study and other actions tested
  24. A replicated site comparison in 2010–2012 in riparian forests along the Sacramento River, California, USA, found similar numbers of birds and mammals in restored and remnant forests. Birds and Mammals: Overall, similar numbers of species were found in restored or remnant forests (4–5 vs 4). More species were found in restored forests, compared to remnant forests, in the wet seasons (4–5 vs 2), but not in the dry seasons (3–5 vs 3). More predator species were found in young restored forests, compared to remnant forests (1.9 times as many species as in remnant forests). Most animals were black-tailed deer Odocoileus hemionus columbianus (66% of all observations) or wild turkeys Meleagris gallopavo (21%). Most predators were raccoons Procyon lotor (8% of all observations), coyotes Canis latrans (4%), or bobcats Felis rufus (1%). Methods: All sites were part of The Nature Conservancy’s Sacramento River Project. Camera traps were set along a 100 km section of the river in restored forests (young: restored in 2003–2007, five sites; old: restored in 1991–2000, six sites) or remnant forests (five sites). Sites were 5.34 km apart, on average. Camera traps (2.1 m height) were placed on trees at each site, near animal signs (tracks, scat, or scratch marks). Cameras were visited every 1–1.5 months.

    Study and other actions tested
Please cite as:

Shackelford, G. E., Kelsey, R., Robertson, R. J., Williams, D. R. & Dicks, L. V. (2017) Sustainable Agriculture in California and Mediterranean Climates: Evidence for the effects of selected interventions. Synopses of Conservation Evidence Series. University of Cambridge, Cambridge, UK.

Where has this evidence come from?

List of journals searched by synopsis

All the journals searched for all synopses

Mediterranean Farmland

This Action forms part of the Action Synopsis:

Mediterranean Farmland
Mediterranean Farmland

Mediterranean Farmland - Published 2017

Mediterranean Farmland synopsis

What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 21

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.


Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape ProgrammeRed List Champion - Arc Kent Wildlife Trust The Rufford Foundation Save the Frogs - Ghana Mauritian Wildlife Supporting Conservation Leaders
Sustainability Dashboard National Biodiversity Network Frog Life The international journey of Conservation - Oryx Cool Farm Alliance UNEP AWFA Bat Conservation InternationalPeople trust for endangered species Vincet Wildlife Trust