Birds feed on a wide variety of organisms and methods to measure the density of all of them are beyond the scope of this chapter. Practical methods for estimating density or an index of density for many taxonomic groups can be found in Sutherland (1996). Rough measures of density may often be sufficient to answer the required questions. We want to emphasize here that studies of the food of birds require information not only on the density of food item, but also on their availability to birds, which is affected by prey activity, protective attributes (such as thorns, camouflage or poisonous compounds), depth in the substrate or height above ground in vegetation. The researcher should find out enough about the basic ecology of the bird to make choices about how to combine the measurement of food abundance with measures of habitat that influence food availability. One simple precaution is to ensure that the sampling of food abundance is being done in the type of habitat in which the birds can forage. It is often the case that birds have strong preferences for particular types of vegetation architecture in the areas where they forage. Samples taken in places that have vegetation cover that prevents birds from foraging their may give a misleading picture of food availability. Hence, pilot studies of the bird's foraging behavior and diet are recommended before large-scale sampling of food abundance and availability begins.
In some cases it may be appropriate to use a method that measures a combination of abundance and activity. For example, the catch per trap per day of ground-living invertebrates in pitfall traps is influenced by the activity of the animals as well as their abundance. This would be a disadvantage in a study of invertebrate population dynamics, but in a study of the availability of food for birds that locate prey when it is active on the surface, pitfall trap catches could be a useful measure. However, the researcher should be careful in deciding which invertebrates from those obtained in pitfall to include as being available to the study's focal bird species. A high proportion of arthropods caught in pitfall traps are nocturnally active and thought should be given to whether these are available to a diurnally foraging bird. This will depend on the foraging behavior of the bird and the resting location by day of the prey.
The availability of flying insects to birds that catch them on the wing can be assessed by powered suction nets with intakes placed high above the ground (Woiwod and Harrington 1994). Catches from even a single trap have been shown to be a good predictor of breeding parameters of local aerial feeding birds.
Suction traps often yield large quantities of insects, which would be difficult to sort and identify, but the daily total volume of the catch can be a useful measure (Bryant 1975). Insects flying near the ground can be sampled by flight interception traps (Ausden 1996) or by counting insects seen through binoculars during a watch of a standard volume of air for a fixed time (Flaspohler 1998). Visual counts of the moths and dipteran flies active during the night that are the prey of Nightjars Caprimulgus europaeus have be made using a vertically oriented spot-lamp that is switched on briefly to minimize the attraction of insects to the light (Bowden and Green 1994).
Sweep nets can provide a useful quick way of assessing relative invertebrate abundance in dense vegetation. It is essential they are standardized, for example, 10 sweeps of constant strength and then the invertebrates counted.
The abundance of soil invertebrates can be assessed by hand sorting or otherwise separating them from the soil cores. However, this may include animals that are inactive or too deep in the soil to be available to the birds. Chemical extraction of soil invertebrates by applying a solution of an irritant chemical such as mustard to a quadrat in the field (Ausden 1996), appears to measure a combination of abundance and activity or proximity to the surface because the number of earthworms extracted per unit area shows short-term variations that are correlated with soil moisture levels that affect earthworm behavior (Green et al. 2000).
The availability of plant foods such as leaves and seeds can be measured by counting plants or their parts in quadrats or along transects. This approach is appropriate for seed-eating bird species that mainly take seed while it is still on the parent plant, but for species that take seed from the soil surface it is necessary to scrape a layer of soil from a measured area and separate the seeds by sieving. The seeds can then be counted and weighed. The contents of many seeds in the soil seed bank may have rotted away or have been removed by soil invertebrates and been replaced by soil. Hence, each seed (or a sample) should be crushed to check that it contains endosperm.
For ground-layer plants that are the food of grazing birds, counts of leaves or seedling cotyledons can be done in quadrats or along transects. It may be useful also to score plants for signs of damage from grazing birds to give a measure of utilization as well as availability. This is especially useful when the plant structures being eaten vary little in their number per plant, size, and shape so that the researcher can easily judge what is missing. For example, by carefully scoring damage to the paired cotyledons of seedlings being grazed by Skylarks Alauda arvensis, it was possible to estimate the species composition of the diet and the dry weight of cotyledon material of each species being eaten per day. The diet species composition results agreed closely with an independent assessment based upon the identification of fragments of cotyledon epidermis in Skylark droppings (Green 1980).
The main approaches for measuring fruit abundance are
1. Visit the same plants regularly and either use a quantitative measure of fruit abundance and ripeness or mark individual branches and count fruits at different stages. This shows differences in phenology between species, individuals, and years (seasonal patterns are reasonably constant in temperate regions but show considerable annual variation in the tropics).
2. Assess fruit density on the ground. This will pick up broad differences between sites and indicate timings but is obviously crude and the persistence on the ground clearly depends upon the abundance of ground dwelling frugivores.
3. Place fruit traps (e.g. suspended bags) under the canopy and count the fruit (usually ignoring aborted fruit) that has fallen. This can provide qualitative data even where the fruiting trees cannot be seen. However, to obtain sufficient data for a forest, a large number of traps are needed (75—300) (Blake et al. 1990). A smaller number of traps are needed if under individual trees. The abundance of fallen fruit is not the same as the abundance of fruit available to tree-dwelling birds.
4. Using transects or point counts (applying methods for bird censuses in Chapter 2) to assess fruit abundance of a range of plants (Blake et al. 1990).
The availability of food for nectar-feeding birds can be assessed by a combination of counts of flowers and measurements of the volume and concentration of nectar in a sample of flowers. Nectar is removed from the flower by probing it with a microcapillary tube. The volume of nectar is estimated by measuring the length of the column of liquid in the tube and its sugar concentration can be measured with a refractometer (Prys-Jones and Corbet 1987).
Was this article helpful?
Do You Want To Learn More About Green Living That Can Save You Money? Discover How To Create A Worm Farm From Scratch! Recycling has caught on with a more people as the years go by. Well, now theres another way to recycle that may seem unconventional at first, but it can save you money down the road.