158
Systematic conservation planning: past, present and future
6.7.1 Conservation p lanning i s
a d ynamic p roblem
Possingham et al . (1993) provided one of the fi rst analyses that formulated the dynamic site selection problem.
In each year they assumed one site could be bought
(due to a constrained budget), sites had a fi xed probability of becoming available and sites that were unreserved
had a fi xed probability of being destroyed. At the time,
these authors found that taking a static approach was
suboptimal compared to solving the dynamic problem
using stochastic dynamic programming.
Various authors have subsequently considered and
solved larger and more realistic versions of this original
problem (e.g. Costello & Polasky, 2004 ; Meir et al .,
2004 ; Drechsler, 2005 ; Strange et al ., 2006 ). Such
advances enable systematic conservation planners to
include complexities like dynamic budgets and feedback between acquisition actions and the cost of reservation. In principle, any sort of dynamic complexity
can be included in the site selection problem; however,
the optimal solution of stochastic dynamic problems
can only be found exactly using stochastic dynamic
programming, which is computationally intractable
for any but the smallest problem. There is therefore a
need to derive simple heuristics that sequentially
choose actions through time, such as choosing the
actions that maximize the short term gain in biodiversity or minimize the short term loss of biodiversity.
6.7.2 Conservation a ssets c hange
t hrough t ime
The biodiversity assets that we would like to conserve
are continually changing: local populations become
extinct; species ’ distributions change; species evolve;
and vegetation types change through succession (as
discussed in Chapter 3 ). This adds further complexity
and uncertainty to the dynamic conservation planning problem described above, and in principle it can
be dealt with within the same approach.
However, there are some short cuts possible. Sites
with evolutionary potential can be preferred in planning (Cowling et al ., 2003 ), present and future predicted distributions can be accommodated in the plans
(Hannah & Hansen, 2005 ) and successional changes
can be predicted and allowed for in target setting
(Drechsler et al ., 2009 ).
6.7.3 A m ix of c onservation a ctions c ould
o ccur at a ny s ite
As discussed briefl y in the introduction, formal protection of habitat is just one of many conservation actions.
In many cases, especially where there are multiple
players in land ownership issues plus complex social
and cultural constraints, reservation is an unlikely
option for conservation. What we need is tools to help
us determine which package of actions to activate at
any site.
This sort of idea is effectively zoning – a common
practice in fi sheries, forestry and conservation where
there are multiple interests (Watts et al ., 2009 ). These
zoning tools are useful to guide broad policy decisions,
and other methods have been developed to systematically select among specifi c conservation actions. For
example, the Project Prioritization Protocol is a cost -
effectiveness analysis that has been demonstrated to
be useful for selecting among specifi c management
projects for threatened species in New Zealand (e.g.
Joseph et al ., 2009 ).
6.7.4 Better e conomics and
s ocio - e conomics
Ando et al . (1998) were arguably the fi rst to highlight
in the peer - reviewed literature the naivety of building
conservation plans that ignored realism in respect of
fi nancial costs. While the inclusion of the estimated
cost of conservation is now more common in conservation planning (see section 6.3.3 ) it is still a challenge
for most conservation researchers who are more familiar with the nuances of biological data (Bode et al .,
2008 ). To this end, there is a need for more real collaboration between economists and conservation
biologists.
However, it is also being recognized that using simple
cost - layer data (i.e. the price of land), without considering socio - economic factors such as a landholder ’ s
willingness to conduct a conservation action, regardless of cost, may lead to some erroneous results.
6.7.5 Dealing with u ncertainty
There is some level of uncertainty in every aspect
of conservation planning (Regan et al ., 2009 ). For
Systematic conservation planning: past, present and future
6.7.1 Conservation p lanning i s
a d ynamic p roblem
Possingham et al . (1993) provided one of the fi rst analyses that formulated the dynamic site selection problem.
In each year they assumed one site could be bought
(due to a constrained budget), sites had a fi xed probability of becoming available and sites that were unreserved
had a fi xed probability of being destroyed. At the time,
these authors found that taking a static approach was
suboptimal compared to solving the dynamic problem
using stochastic dynamic programming.
Various authors have subsequently considered and
solved larger and more realistic versions of this original
problem (e.g. Costello & Polasky, 2004 ; Meir et al .,
2004 ; Drechsler, 2005 ; Strange et al ., 2006 ). Such
advances enable systematic conservation planners to
include complexities like dynamic budgets and feedback between acquisition actions and the cost of reservation. In principle, any sort of dynamic complexity
can be included in the site selection problem; however,
the optimal solution of stochastic dynamic problems
can only be found exactly using stochastic dynamic
programming, which is computationally intractable
for any but the smallest problem. There is therefore a
need to derive simple heuristics that sequentially
choose actions through time, such as choosing the
actions that maximize the short term gain in biodiversity or minimize the short term loss of biodiversity.
6.7.2 Conservation a ssets c hange
t hrough t ime
The biodiversity assets that we would like to conserve
are continually changing: local populations become
extinct; species ’ distributions change; species evolve;
and vegetation types change through succession (as
discussed in Chapter 3 ). This adds further complexity
and uncertainty to the dynamic conservation planning problem described above, and in principle it can
be dealt with within the same approach.
However, there are some short cuts possible. Sites
with evolutionary potential can be preferred in planning (Cowling et al ., 2003 ), present and future predicted distributions can be accommodated in the plans
(Hannah & Hansen, 2005 ) and successional changes
can be predicted and allowed for in target setting
(Drechsler et al ., 2009 ).
6.7.3 A m ix of c onservation a ctions c ould
o ccur at a ny s ite
As discussed briefl y in the introduction, formal protection of habitat is just one of many conservation actions.
In many cases, especially where there are multiple
players in land ownership issues plus complex social
and cultural constraints, reservation is an unlikely
option for conservation. What we need is tools to help
us determine which package of actions to activate at
any site.
This sort of idea is effectively zoning – a common
practice in fi sheries, forestry and conservation where
there are multiple interests (Watts et al ., 2009 ). These
zoning tools are useful to guide broad policy decisions,
and other methods have been developed to systematically select among specifi c conservation actions. For
example, the Project Prioritization Protocol is a cost -
effectiveness analysis that has been demonstrated to
be useful for selecting among specifi c management
projects for threatened species in New Zealand (e.g.
Joseph et al ., 2009 ).
6.7.4 Better e conomics and
s ocio - e conomics
Ando et al . (1998) were arguably the fi rst to highlight
in the peer - reviewed literature the naivety of building
conservation plans that ignored realism in respect of
fi nancial costs. While the inclusion of the estimated
cost of conservation is now more common in conservation planning (see section 6.3.3 ) it is still a challenge
for most conservation researchers who are more familiar with the nuances of biological data (Bode et al .,
2008 ). To this end, there is a need for more real collaboration between economists and conservation
biologists.
However, it is also being recognized that using simple
cost - layer data (i.e. the price of land), without considering socio - economic factors such as a landholder ’ s
willingness to conduct a conservation action, regardless of cost, may lead to some erroneous results.
6.7.5 Dealing with u ncertainty
There is some level of uncertainty in every aspect
of conservation planning (Regan et al ., 2009 ). For