at random locations, or better still, locations placed on a regular grid covering the
study area. One frequent error is to intensively sample one or a few small plots as
being a measure of the entire area. However, when relating habitat to point counts
it is usual and sensible to collect data just around the point.
Much of the skill in measuring habitat considers of devising an efficient and
effective protocol that can easily be repeated by different observers or by yourself
after a long interval. For example, the protocol for habitat sampling around a
point might be to count the number of trees within a 50-m radius circle around
the point, to measure the soil pH at the central location and to measure the sward
height at five locations at 20-m intervals along a north–south transect through
the central location.
We know of many examples in which overenthusiastic habitat measurement has
weakened an otherwise perfectly good field study. A major problem is balancing
effort against precision. If habitat assessment is too time consuming then this
may result in too few sites being visited, so estimates of habitat attributes have low
precision because of low sample size. However, if the assessment is too quick and
crude then the essential variables relevant to the birds may not be measured accu-
rately, though precision is good because of the large sample size.
There are a number of possible means of planning a sensible programme
including:
- Estimate how long the habitat measures will take. This can even be done
before setting out to the field site. One approach is to try them out in the
most similar habitat nearby. - Estimate the tradeoff between the number and quality of data points. For
example, how many more sites could be visited if the trees were not identified
to species? - Consider the balance between the time and accuracy of the bird and habitat
data. For example, if the habitat measures are time consuming (e.g. 10-min
point counts followed by 3 h measuring habitat) then it might be better to
repeat each bird count twice to make the point count more accurate at the
cost of slightly reducing the sample size. - Consider when time is most limited. One useful trick to maximize the use
of the early morning, when birds are most active, is to collect bird data
along transects or points while walking out and when returning collect the
habitat data from locations that have been marked.
Think carefully about the questions and plan the research design that is most
likely to answer these. If there is insufficient time to answer the question then do
something else.
Protocols| 253