The effect of fire on flower visitation rate and fruit set in four core-species in east Mediterranean scrubland

Summary

The recovery of vegetation following fire has been extensively studied in Mediterranean shrublands. However, almost no data have been collected on the effects of fire on subsequent pollinator activity and plant fruit-set. This paper reports the effects of fire on flower visitation rates, and the possibly related fruit-set, 5-7 years post-burn. Visitation rates of the main pollinators (bumblebees Bombus spp. and solitary bees: Hymenoptera) on four core-species (branched asphodel Asphodelus ramosus, sage Salvia fruticosa, thyme-leaved savory Satureja thymbra and Phlomis viscosa) in burned and adjacent unburned areas were recorded.

Study area: The study was carried out on Mount Carmel, 3-7 km east of Haifa, Israel. Two study sites (c. 300 m elevation) were established, one in unburned garigue and phrygana and one in an adjacent area (burned in September 1989), where the four study species (branched asphodel Asphodelus ramosus, sage Salvia fruticosa, thyme-leaved savory Satureja thymbra, Phlomis viscosa) occurred; individuals were often widely dispersed. The sites were about 2 km apart and were of similar slope, aspect, Rendzina soils and vegetation structure. The study species monitored in the burned area were located about 1 km from the fire border to limit edge effects (the size of the burned area is not given in the original paper).

Pollinator observations: Observations were carried out in spring 1994, and on S.thymbra also in 1995. A random grid point was assigned within each site from which a 100 m transect was established. All medium and large sized flowering individuals along transects were observed at the peak flowering season for each species during the daily period of main insect activity (10:00 to 14:00; above 18ºC). For each plant species, all individual insects that visited the flowers were counted in units of 10-30 min, with between 5-20 h total observation time per species. Bees were caught as required for identification purposes, and ranked according to their size into three categories: large (>13 mm in length), medium (12-9 mm) and small (<8 mm). Honeybees Apis mellifera and buff-tailed bumblebees Bombus terrestris (the only Bombus recorded) were counted separately.

Fruiting: Fruits were counted (in 1994, 1995 and 1996) on 10 medium to large individuals that were randomly chosen. Fruits were counted on 6 random branches of each individual plant, and average percent fruit-set was calculated.

Asphodelus ramosus Solitary bees of all size classes visited flowers in the unburned area. Only large solitary bees did so in the burned area where the average number of visits/flower/10 min was higher (8 visits) than in the unburned area (5 visits). However, when all solitary bee size classes combined the number of visits was much higher in the unburned area (18 visits) than the burned area (8 visits). As A.ramosus flowers in early spring (February–March) and B.terrestris activity on Mt. Carmel commences at the end of March, it was not observed visiting flowers; likewise A.mellifera

Average percent fruit-set was higher in the unburned area, but significant so only in 1994 (1994: burned - c.30%; unburned - c.48%. 1995: burned - c.29%; unburned - c.35%). There was a positive trend between the higher visitation rate and the higher fruit-set in the unburned vs the burned area.

Salvia fruticosa B.terrestris were the only visitors to the flowers in the burned area, and their visitation was significantly higher (c.3 visits/flower/10 min) than in the unburned area (1 visit). Honeybees and large, medium and small solitary bees also visited the flowers in the unburned area (all at visitation rates of <1 visit/flower/10 min). Overall total pollinator activity was slightly higher in the burned (c.3 visits/flower/10 min) than unburned (2 visits) area. Average fruit-set was higher in the unburned area (16-20%) than the (13-15%) burned area; therefore there was a negative trend between pollinator activity and fruit set.

Satureja thymbra B.terrestris was almost the only flower visitor in 1994. In 1995 B.terrestris visitation was much lower, but visits of other bees were slightly more frequent than in 1994 (but still <1 visit/flower/10 min for honey bees and each solitary bee size class). Bumblebee visits were far more frequent in the unburned than in the burned area in 1994 (c.15 visits/flower/10 min and <0.5 visit respectively) and 1995 (c.3 visits/flower/10 min and <1 visit respectively); likewise total visits of all bees. In both 1994 and 1995, fruit-set was higher in the unburned area (1994: unburned - 90%; burned - 45%. 1995: unburned - 54% burned - 30%); this trend is in accordance with the higher flower visitation rate.

Phlomis viscosa B.terrestris and solitary bees (large only, smaller classes absent) visited flowers more frequent in the unburned area (c.1 and 0.6 visits/flower/10 min respectively) than in the burned area (c.0.3 visit and <0.1 visits/flower/10 min respectively). Solitary bees were almost entirely absent from the burned area. In each of the three year (1994-96), fruit-set was consistently higher in the unburned area (><50-60 %) than in the burned area (35-44%); this trend is in accordance with the higher flower visitation rate.

(Note: all above visitation rate and fruit set values are read from graphs in original paper).

Conclusions: Fruit-set was higher in the unburned area for all four species in the study years 1994-96 (and significantly so for all species except for A.ramosus in 1995). The lower fruit-set in the burned area was possibly the result of low solitary bee activity – solitary bees are the main ollinators of the plant species investigated - but were almost absent from the burned area. The authors suggest that populations of solitary bees may have been diminished or extirpated by the fire, or subsequent to it, by the scarcity of nectar in the early post-fire years; the typically short foraging range of solitary bees and their slow invasion rate into the burned area may explain this.


Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at:

http://www.springerlink.com/content/k142943l12238772/fulltext.pdf