Environmental Engineering Reference
In-Depth Information
2.4.2.4 Harvest mortality
The survival analysis methods described above provide estimates of overall survival
rate, which encompasses both natural and harvest mortalities. If we wish to assess
the impact of the harvest, it is useful to know what proportion of the overall mor-
tality it represents (see Chapters 4 and 5). Harvest mortality can be measured
either by monitoring directly the amount of offtake and relating this to the popu-
lation size, or by using one of several indirect approaches.
The ratio of catch to pre-harvest population size gives an obvious direct measure
of harvest mortality (Box 2.12). For example, Menu
et al
. (2002) used this
approach to estimate harvest mortality in greater snow geese
Anser caerulescens
in
North America, taking total harvest from national hunter surveys, and estimating
pre-harvest (autumn) population size from a combination of spring aerial
photo surveys and summer survival rate estimated from ring recovery data. A
major challenge with this approach is to obtain an accurate measure of the catch.
Harvest may be monitored in many ways, from accompanying harvesters in
order to observe offtake directly, through monitoring of collection points such as
hunting bases or markets, to various forms of questionnaire to provide indirect
reports on offtake (see Chapter 3 for guidelines on these field methods). When
applying these methods to measure total offtake, it is important to recognise
the potential bias that may arise. For example, the more harvesters are involved,
and the more remote the monitoring point is from the point of offtake, the more
likely it is that significant amounts of offtake will go unrecorded. In this case,
any pretence at complete coverage should be abandoned, in favour of a rigorous
sampling approach
.
Box 2.12 provides analytical methods for calculating total offtake from a sam-
ple. The crucial assumption with these methods is that the harvesters and occasions
sampled are
representative
of the overall pattern of harvest. When sampling
occasions, it is usually easy to ensure representative coverage through a carefully
designed sampling regime, for example, by picking random days of the week, or using
a regular cycle that covers all days of the week equally. However, it is often more dif-
ficult to achieve a representative sample of harvesters. This is because harvesters
that are less accessible to monitoring are likely to have different patterns of effort
or success than those that are easier to reach, particularly if the survey relies on
voluntary responses. The best option is to start by identifying the entire harvester
population, then ensure that all individuals have an equal chance of being moni-
tored through stratified or strictly random sampling. Where this is not possible, it
may be necessary to make a special effort to sample individuals not covered by
the main survey, for example, by following up voluntary returns with direct obser-
vation of non-respondents. This can make it possible to detect and correct for any
bias in the main survey.
However, it is important to bear in mind that even a well-designed, representa-
tive offtake sample may suffer from under-reporting. This is particularly true if
there are social or legal pressures against harvest, but can be a problem even in the
