 |
|
|
|
|
Download full report (780kb - HINT: right click and save the file to your hard disk before opening)
by David C. Paton
May 2008
RIRDC Publication No08/087 RIRDC Project No UA-66A
Executive Summary
What the report is about
This report documents the
floral resources used by the honeybee industry in rural landscapes of South
Australia, particularly the Mt Lofty Ranges. Various species of eucalypt
are important sources of nectar for honeybees in this region. A simple
method is developed to document the flowering of these plants, allowing
spatial and temporal variation in flowering to be documented.
The research focussed on documenting nectar production and flowering of key eucalypts and comparing the performances of these plants in remnant native vegetation, with those of trees in paddocks and revegetation. The extent to which paddock trees, trees in revegetation and trees in intact native vegetation were used by honeybees and native fauna particularly native birds is assessed using visitation rates to flowers and from the quantities of nectar left unexploited in flowers at the end of the day. The influences of drought, frost and fire on floral production were assessed at several long-term monitoring sites and used as a basis for predicting likely changes in floral resources for the honeybee industry with climate change.
Additional long-term monitoring programs were established on three farms that are being set aside for revegetation to document changes in floral resources through time. Some recommendations on planting densities and species compositions in revegetation programs are made to improve the likelihood that these revegetation programs will provide resources of value to the honeybee industry.
Who is the report targeted
at?
The report aims primarily
to provide information on the use of different floral resources by the
honeybee industry, and how the flowering levels of key plant species changes
over time and following perturbations like fire and drought. Since fire
and drought are more likely in southern Australia due to climate change,
monitoring the performances of the plants during and following the current
drought, and following recent fires, provides a strong basis for predicting
future changes to floral resources.
Background
The honeybee industry uses
three types of floral resources: a diversity of native plants, a suite
of introduced plants that are environmental weeds, and various annual and
perennial crops. The value of native plants as floral resources has been
eroded in recent years, and is likely to continue to decline, because of
vegetation clearance, rural tree decline, and reduced access to crown land.
Changes in agriculture practices and intensification may also reduce floral
resources to honeybees by eliminating some of the important weed species.
Climate change also has the potential to alter flowering levels and performances
of these plants. To be able to plan for the future, the honeybee industry
needs to better document the floral resources that are currently being
used and to establish baselines from which to monitor changes in these
resources over time.
Aims/Objectives
The objectives of this project
were to:
Methods used
Observations of the use
of floral resources by honeybees were undertaken at six sites established
in the Mt Lofty Ranges of South Australia. The sites included rural areas
with scattered trees, areas of intact native vegetation, and areas of revegetation.
Counts of honeybees and other insects visiting flowers on marked branches
at hourly intervals throughout the day coupled with the time taken to process
flowers were used to determine daily visitation rates to flowers. Visitation
rates to flowers by birds were determined from the amount of time birds
spent foraging on the flowers of marked branches during 30 minute observation
periods spread evenly over the day, combined with the time the birds took
to probe flowers. Nectar was measured using calibrated capillary tubes
to extract and determine the volume of nectar in a flower and a hand-held
sugar refractometer used to measure the sugar concentration. The quantities
of nectar produced per day were determined by covering flowers with fine
voile bags for 24 hours, measuring the accumulated nectar in these flowers,
and subtracting the quantity present in samples of flowers taken at the
time the flowers were bagged.
A simple method of estimating the numbers of flowers produced per unit of canopy, the handspan method, was developed and used for documenting the quantities of flowers produced by plants. The quantities of flowers per handspan were estimated for 25 to 50 individual plants to represent the level of flowering for plant populations at different locations and different times. Abundances of flowers and responses to fire and drought were assessed for Banksia ornata in Ngarkat annually using a series of permanent 50 m x 4 m quadrats. Smaller sized quadrats were used to document floral abundances of other species.
The establishment of baseline levels of floral resources in farming areas was done using a GPS to map all the trees. Line transects and 25m x 25m cells were used to document per cent cover or numbers of individuals for understorey species.
Results/Key findings
The quantities of nectar
produced over 24 hours were determined for eighteen species of eucalypts,
including one natural hybrid. Of the species sampled, four consistently
produced more nectar per flower per day than the other species. These species
were Eucalyptus angulosa (7.4 – 10.6 mg sugar/flower/day), E.
torquata (7.6 – 15.4 mg/fl/d), E. leucoxylon (7.3 – 15.6 mg/fl/d)
and E. cosmophylla (9.7 - 18.8 mg/fl/d). Eucalyptus albopurpurea,
E cladocalyx, E. diversifolia, E incrassata, a natural E.
porosa -E. leucoxylon hybrid, and E. platypus typically produced
2-5 mg/fl/d, while E. baxteri, E. calycogona, E. fasciculosa,
E. odorata, E. socialis and E. viminalis typically
produced between 0.5 – 2 mg/fl/d. Eucalyptus remota and E. porosa
produced <0.5mg/fl/d. The quantities of nectar produced by plants
in revegetation, in remnant vegetation or by plants scattered across paddocks
were similar. Plants in revegetation programs, therefore, are just as productive
as those in intact native vegetation, so revegetation has the potential
to replace losses in floral resources.
The quantities of nectar remaining in flowers late in the day were often substantial, particularly for Eucalyptus cosmophylla, E. leucoxylon, E. angulosa, E incrassata and E. torquata where there was often more than 1 mg/fl left unexploited. For some species like E. leucoxylon and E. cosmophylla there was often over 5 mg/fl of nectar, even when a commercially-managed apiary was within a few hundred metres. At times these high volumes of nectar were such that nectar dripped from the flowers.
This apparent over-production of nectar, however, was not caused by the plants producing more nectar in some years than others but reflected low visitation rates to flowers particularly by native birds. The high quantities of unexploited nectar in flowers indicate that honeybees are unlikely to be affecting the ability of native fauna to harvest nectar from these flowers. Beekeepers, therefore, should be allowed to have continued access to these resources.
The rates at which flowers were visited by native fauna and honeybees were assessed for ten species of eucalypts on 33 occasions, across three different settings: remnant vegetation, paddock trees and revegetation. For most species and on most occasions honeybees were more frequent floral visitors than other invertebrates and birds, accounting for more than 100 visits per flower per day for at least three species of eucalypt (E. cosmophylla, E. odorata and E. torquata) on some occasions. For another four species of eucalypts, honeybees made at least five visits per flower per day on most occasions.
The other major visitors to flowers were various species of honeyeaters, lorikeets, Silvereyes and Crimson Rosellas, the latter often removing flowers to harvest nectar from them.
The rates at which birds visited flowers varied dramatically both within a species and between species, ranging from 0 - 23.4 visits/flower/day. On more than half of the occasions, birds made less than one visit/flower/day suggesting that floral resources often exceeded the numbers of birds available to exploit them consistent with measurements of nectar in flowers. In general, those eucalypts (e.g. E. calycogona, E. fasciculosa, E. odorata) with small flowers and producing low to moderate quantities of nectar (<2mg/fl/d) received 0 - 1.4 visits/flower/day from birds. Those species that had larger flowers and produced higher amounts of nectar (E. cosmophylla, E. leucoxylon, E. torquata) typically had visitation rates ranging from 0.1 - 10 visits/flower/day. For Eucalyptus leucoxylon visitation rates were generally higher in remnant vegetation than in paddock trees or revegetation, particularly in 2004 and 2006 when E. leucoxylon flowered more extensively. This suggests that any potential impact of honeybees on native flora and fauna will be lower for scattered paddock trees. This pattern of reduced use of paddock trees, however, was not seen for either Eucalyptus diversifolia or E. cosmophylla.
On two occasions, both involving eucalypts in remnant vegetation, the rates at which birds visited flowers exceeded 20 times per flower per day. These both coincided with times when there were few other plants in flower.
Sources of pollen used by honeybees when exploiting floral resources at the study sites were largely confined to one or two species at a site. Key pollen sources being used by the bees were Echium plantagineum, and Hypochoeris glabra, both environmental weeds. Eucalyptus camaldulensis, Gonocarpos elatus and Lissanthe strigosa were the native plants that provided some pollen but only at 1 – 2 study sites each.
The numbers of flowers produced by different species of Eucalyptus varied dramatically. Some species produced moderately low maximum numbers of flowers (e.g. E. angulosa, E. astringens, E. cosmophylla, E. incrassata and E. torquata) and these tended to be species with larger flowers. Other species with smaller flowers had maximums in excess of 200 flowers per handspan of canopy (e.g. E. baxteri, E. calycogona, E. camaldulensis, E. cladocalyx, E. diversifolia, E. fasciculosa, E. obliqua, E. odorata, E. porosa, and E. socialis). One species, E. leptophylla had over 800 floral units per handspan of canopy.
There were also substantial changes.
Considerable variation was found in flowering levels (flowers/handspan) between individual plants within the same population and between populations of the same species. For example, the numbers of flowers produced by individual E. camaldulensis varied from <5 to >250 per handspan, an almost 50 fold spread. Despite the broad range in flowering levels there were differences in the spread of flowering levels between years within a site and also between sites in any one year. Eucalyptus leucoxylon flowered extensively in spring 2006 despite a severe drought but most populations did not flower or flowered poorly in 2007, suggesting that droughts may affect subsequent flowering rather than the flowering at the time of the drought.
The complexity or variability in the flowering of these plants was increased further by the wide range in the timing of flowering of individual plants within a population. For all of the species of eucalypts that were monitored there were some individual plants within a population that had almost finished flowering while others had barely commenced flowering. This pattern suggests that the timing of flowering within these populations of plants is influenced at least to some extent by genetic differences rather than local environments, since plants a few metres apart should experience similar environments. One outcome of this variability is that moderate to large sample sizes are needed to document flowering seasons adequately.
The handspan method that has been developed in this study provides a relatively simple way of estimating floral production for a range of eucalypts that is capable of documenting not just changes in the quantities of flowers being produced from one year to the next, but also the variability that exists within and between populations in floral production. Since the numbers of flowers and not the quantity of nectar produced by flowers is the major variable affecting resource production monitoring the quantities of flowers produced each year over extended periods is required to document changes in resources. South Australian beekeepers will commence monitoring some of their floral resources using this technique in 2008.
Long-term monitoring of Banksia ornata at Ngarkat revealed that this species produced far fewer flowers following drought years. Recovery of floral production following a fire in January 1999 was much slower (~ 9 years) compared to recovery following earlier fires in December 1990 and November 1997, suggesting that recent droughts were also affecting post-fire recovery. Low seedling recruitment for Banksia ornata following the January 2006 fire suggests floral resources for this reserve will be even further reduced following this fire.
The distributions and abundances of key floral resources being used by honeybees on three farms being set aside for re-establishment of native vegetation were mapped and provide a baseline from which to judge how floral resources change as the new vegetation is established. Trees planted densely (>250 trees/ha) may not be as productive as more widely spaced trees. More widely spaced trees have larger canopy dimensions and may produce more flowers per unit area of canopy, and so be a more valuable resource to the beekeeping industry. The weeds that currently provide significant quantities of pollen to the honeybee industry in these rural landscapes are likely to be eliminated as part of the revegetation works. Suitable native plants that can provide pollen still need to be identified and incorporated into the planting programs.
Implications for relevant
stakeholders
The floral resources used
by honeybees are sensitive to drought, fire and cold temperatures, and
these can affect the quantities of flowers produced for more than just
a single flowering season. Given that droughts, fires and extremes in temperature
are all likely to increase with climate change, floral resource availability
is likely to decline in the future. The honeybee industry is therefore
likely to be significantly affected by predicted climate change.
Given that significant quantities of nectar were left unexploited in the flowers of key eucalypts even close to apiaries, is consistent with honeybees not having a detrimental effect on native nectarivores.
This was particularly true for scattered paddock trees that were used less frequently by nectar-feeding birds than trees in intact native vegetation. In general the low visitation rates by honeyeaters to flowers may indicate that these birds are also vulnerable to the impacts of climate change, and sounds a warning for wildlife managers.
Recommendations
The honeybee industry needs
to invest in understanding how the floral resources that they currently
depend on will respond to climate change. The relatively simple method
developed for scoring the flowering performances of native plants, particularly
eucalypts should allow individual beekeepers to commence monitoring key
assets and so provide a broader base on which changes to resource levels
are assessed. There is considerable potential, and a commitment from a
biodiversity perspective, at least in South Australia, to re-establish
extensive areas of native vegetation that if planted at appropriate densities
with an appropriate mix of plants may provide not only biodiversity outcomes
but critical additional floral resources for the honeybee industry. Identification
of potential pollen sources for honeybees from amongst the local native
flora is still deficient.
|