References | Country | NP | Area | Pollinator | Matrix | Isolation | Key results |
---|---|---|---|---|---|---|---|
[4] | Argentina | 8 | 0.5–480 | Native and honeybee | Cornfield | – | By decreasing patch size, the frequency and richness of native bees declined but the relative numbers of honeybees increased. Honeybees visiting were negatively correlated with individual trees |
[120] | Brazil | 2 | 50 | Euglossine | – | – | Euglossine bees moved over cleared areas in search of fragrances |
[78] | Panama | 10 | 1–1500 | Euglossine | Mixed | 100–500 | Euglossine bees recorded on islands were visitors from mainland sites and were equally frequent in fragments and continuous forest |
[68] | Brazil | 8 | 1–50,000 | Bombus brasiliensis | Farm | 4 | Patches maintain greater richness and frequency of floral visitors than continuous sites |
[20] | Costa Rica | 22 | 0.25–230 | Native and honeybee | Farm; pasture | 500 | Tree-nesting Meliponines were correlated with larger patches, smaller edge: area ratios. Honeybees showed opposite patterns |
[122] | Chile | 5 | 2–600 | Native bees | Pine plantation | – | Small patches had higher species richness than continuous forests |
[18] | Costa Rica | 22 | 0.25–230 | Euglossine | Pasture | 0.5–19 | Euglossine bees’ abundance was significantly positively correlated to forest patch size, negatively related to forest shape. Richness was negatively related to fragment area, and not related to fragment isolation |
[47] | Spain | 6 | 2–140 | Native and honeybee | Farm | 1–20 | Large patches supported a greater flower visitor diversity, but small patches tended to have higher insect visitation rates |
[79] | Brazil | 9 | 1–354 | Euglossine | Farm | > 100 | Fragment size or ratio area/perimeter did not affect the abundance and richness of euglossine bees but the size of core areas positively affected them |
[131] | Switzerland | – | – | Chelostoma florisomne; Hoplitis adunca | Farm | – | Forests covering a distance of up to 480 m were crossed by Chelostoma florisomne |
[23] | Mexico | 14 | 0.07–24.9 | Native bees | Pasture | 2 | Patch size positively affected bee richness and diversity |
[130] | Mexico | – | – | Euglossa dilemma | Farm | 130–200 | Bee populations forest remnants were neither differentiated from nor had less genetic diversity than, populations in near-continuous forest separated from 130 km of agricultural lands |
[1] | Brazil | 9 | 2–18 | Euglossine | Pasture; tomato | 0.05–135 | The smallest forest patch had the highest abundance of bees |
[61] | Brazil | 4 | 287–94,000 | Euglossine | Coffee | – | Forest shape index negatively affected euglossine abundance |
[5] | Brazil | 3 | 100–280 | Euglossine | Pasture | 3 | The largest fragment was the main source of the observed variation in species richness and abundance |
[82] | Tanzania | 6 | – | Megachile | Tea | – | Continuous fragments had a higher diversity of pollinators than forest patches |
[48] | Australia | 4 | 0.15–30 | Honeybee | Farm | – | Honeybee abundance and pollen deposition were lower in small patches |
[36] | New Zealand | 15 | 0.01–1,000,000 | Bumblebee | Grassland | – | Patch area did not affect variation in the abundance or biomass of bumblebees |
[45] | Brazil | 5 | 3–484 | Native bees | Soybean | 20 | Patch size positively affected the abundance of Apinae and oligolectic bees and negatively affected the richness of Augochlorini bees |
[114] | Canada | 3 | 7–350 | Andrena | Forest | – | Two small fragments had higher reductions in reproductive output than the continuous (350 ha) fragment |
[10] | Ethiopia | – | 4–100,000 | Honeybee | Coffee | – | Forest fragmentation increased the relative abundance of honeybees |
[98] | Brazil | – | – | Eulaema Athletica | Oil palm; rubber tree |  | Fragment size and isolation did not affect genetic diversity |
[110] | Costa Rica | 12 | 0.9–16 | Native bees | Farm | 2 | Fragment size did not affect bee abundance, diversity, and parasitism, and mortality rates in trap nests. Total bee abundance did not vary from edge to center. Species diversity was higher in the forest center |
[13] | Ecuador | 19 | 2.5–3500 | Euglossine | Farm | 0.3–17 | Fragments area and isolation did not affect bee abundance, richness, or evenness |
[80] | Brazil | 30 | 15–25 | Native bees | Mixed | – | Open areas had higher bee richness and diversity than forest patches |
[106] | Australia | 14 | > 5,  < 20 | Native bees | – | – | Large forest fragments had higher taxonomic diversity of bees visiting flowers of trees than small fragments. Small fragments had higher mean body sizes than those in larger fragments. The abundance of stingless bees decreased in small fragments compared to large fragments |
[118] | Japan | 13 | 1.3–10 | Native bees | Farm | – | Small patches can have the same potential in maintaining as large patches. Bee richness quickly increased at the small range of the area (< 3 ha) |
[65] | USA | 14 | 5–164 | Solitary bees | Mixed | 0.6–19 | Forest patch size did not affect bee community structure or individual family occupancy |
[76] | USA | – | – | Bombus vosnesenskii; Bombus bifarius | Forest | – | Forests did not act as barriers to the fine-scale movement for either species |
[44] | Norway | 24 | 0.11–72 | Bumblebee | Farm | 0–428 | Patch isolation negatively affected bumblebee abundance Forest fragmentation reduced the abundance of forest specialists while increasing the abundance of open-habitat species |
[66] | Norway | 24 | – | Native bees | Farm | – | Patch complexity negatively affected the total number of pollinators |