Conservation planning in a changing world
185
Another approach to dealing with threats that
cannot be stopped by reserves and reserve management is to build a robust reserve system that buffers
species against those threats. One way of buffering a
single reserve against catastrophe is to ensure that any
chance it is completely affected by a single catastrophe
is very low. However, given that many catastrophes,
such as fi re and oil spills, have size distributions with
long tails (there is still a small chance of huge catastrophes), the chance of having a single reserve big
enough to withstand every catastrophe is negligible.
We therefore need to choose enough reserves to spread
the risk and deliver adequate persistence (Quinn &
Hastings, 1987 ; Gilpin, 1988 ; Pressey et al. , 1993 ;
McCarthy et al. , 2005 ). While there are many
approaches to deal with threats that cannot be mitigated, their implementation in actual conservation
decision - making has so far been limited.
Species change their distribution in both predictable
and unpredictable ways. Where the changes are part
of predictable seasonal migrations, then, in principle,
we can have moving reserve systems (Botsford et al. ,
2003 ; Grantham et al. , 2008 , 2009 ; Klaassen et al. ,
2008 ; Game et al. , 2009 ). In principle, moving reserves
can deliver better conservation outcomes, though
frequently this might not be feasible from a socio -
economic and practical political perspective. For a discussion of dynamic conservation planning in marine
systems, see Section 5.4.8 .
7.4.2 Changes in s ocio - e conomic f actors
Proper conservation prioritization requires a consideration of the costs of different actions from a human
perspective. Conservation budgets are limited, and
every decision we make about nature conservation is
necessarily a trade - off with other human aspirations
such as providing shelter, food and water security for
people (Ando et al. , 1998 ; Wilson et al. , 2006 ; Polasky
et al. , 2008 ). However, human aspirations are themselves dynamic. For example, what was considered
high - quality (and hence high - cost) agricultural land in
the past may or may not be suitable in the future as a
consequence of technological change, changing
markets or both (Naidoo et al. , 2006 ).
Bearing this in mind, optimal spatial conservation
planning should factor in likely changes in costs and
opportunities. One way to incorporate these dynamic
costs is to formulate the conservation planning problem
as a Markov Decision Process and solve that problem
using stochastic dynamic programming (Possingham
et al. , 1993 ). That is to say, the system is modelling
using inputs and decisions partly under the control of
a decision - maker and partly randomly determined, in
a step - by - step iterative process designed to simulate a
dynamic real - world scenario. This has been done on a
few occasions, but so far the approach is so unwieldy
that it is largely of only theoretical interest (Costello &
Polasky, 2004 ; Meir et al. , 2004 ; Drechsler, 2005 ;
Strange et al. , 2006 ).
Despite this, it has been possible to develop rules of
thumb to help us decide where to buy conservation
reserves. Such approaches can be applied to issues
beyond reserve acquisition to include any sort of land
use management (Hof et al. , 1994 ; Hof & Raphael,
1997 ; Ö nal & Briers, 2003 ; Westphal et al. , 2003 ).
Fully incorporating socio - economic dynamics with
conservation decision - making will require more interaction with economists and social scientists (cf. Box
7.5 ). While we regard the prediction of species persistence to be a major challenge, predicting changes in the
economy and human preferences is probably even
harder. Reducing this uncertainty and having better
predictive models is important, but ultimately we may
have to live with a high level of risk and uncertainty and
factor it into our decision - making. Fortunately, there is
a variety of tools we can employ to assist in making
robust decisions under risk and uncertainty (for a full
recent review of this topic see Regan et al. , 2009 ).
7.4.3 Climate c hange, c onservation
p lanning and a ssisted m igration
Climate change is one all - pervasive threat that protected areas cannot stop. Ever since the recognition by
Peters & Darling ( 1985 ; and see Figure 7.3 ) that some
protected areas could be rendered obsolete by climate
change, there has been interest in developing spatial
conservation plans that deliver species persistence in
the face of climate change.
One approach to conservation planning under
climate change is to protect every species in all its
ranges – present and future (Hannah et al. , 2007 ). If
we have predictive models of future species distribution
models for each species, then these future distributions
can be added as features in any conservation plan. This
simplest of approaches ignores the question of how a
species might shift its range. However, that can be
185
Another approach to dealing with threats that
cannot be stopped by reserves and reserve management is to build a robust reserve system that buffers
species against those threats. One way of buffering a
single reserve against catastrophe is to ensure that any
chance it is completely affected by a single catastrophe
is very low. However, given that many catastrophes,
such as fi re and oil spills, have size distributions with
long tails (there is still a small chance of huge catastrophes), the chance of having a single reserve big
enough to withstand every catastrophe is negligible.
We therefore need to choose enough reserves to spread
the risk and deliver adequate persistence (Quinn &
Hastings, 1987 ; Gilpin, 1988 ; Pressey et al. , 1993 ;
McCarthy et al. , 2005 ). While there are many
approaches to deal with threats that cannot be mitigated, their implementation in actual conservation
decision - making has so far been limited.
Species change their distribution in both predictable
and unpredictable ways. Where the changes are part
of predictable seasonal migrations, then, in principle,
we can have moving reserve systems (Botsford et al. ,
2003 ; Grantham et al. , 2008 , 2009 ; Klaassen et al. ,
2008 ; Game et al. , 2009 ). In principle, moving reserves
can deliver better conservation outcomes, though
frequently this might not be feasible from a socio -
economic and practical political perspective. For a discussion of dynamic conservation planning in marine
systems, see Section 5.4.8 .
7.4.2 Changes in s ocio - e conomic f actors
Proper conservation prioritization requires a consideration of the costs of different actions from a human
perspective. Conservation budgets are limited, and
every decision we make about nature conservation is
necessarily a trade - off with other human aspirations
such as providing shelter, food and water security for
people (Ando et al. , 1998 ; Wilson et al. , 2006 ; Polasky
et al. , 2008 ). However, human aspirations are themselves dynamic. For example, what was considered
high - quality (and hence high - cost) agricultural land in
the past may or may not be suitable in the future as a
consequence of technological change, changing
markets or both (Naidoo et al. , 2006 ).
Bearing this in mind, optimal spatial conservation
planning should factor in likely changes in costs and
opportunities. One way to incorporate these dynamic
costs is to formulate the conservation planning problem
as a Markov Decision Process and solve that problem
using stochastic dynamic programming (Possingham
et al. , 1993 ). That is to say, the system is modelling
using inputs and decisions partly under the control of
a decision - maker and partly randomly determined, in
a step - by - step iterative process designed to simulate a
dynamic real - world scenario. This has been done on a
few occasions, but so far the approach is so unwieldy
that it is largely of only theoretical interest (Costello &
Polasky, 2004 ; Meir et al. , 2004 ; Drechsler, 2005 ;
Strange et al. , 2006 ).
Despite this, it has been possible to develop rules of
thumb to help us decide where to buy conservation
reserves. Such approaches can be applied to issues
beyond reserve acquisition to include any sort of land
use management (Hof et al. , 1994 ; Hof & Raphael,
1997 ; Ö nal & Briers, 2003 ; Westphal et al. , 2003 ).
Fully incorporating socio - economic dynamics with
conservation decision - making will require more interaction with economists and social scientists (cf. Box
7.5 ). While we regard the prediction of species persistence to be a major challenge, predicting changes in the
economy and human preferences is probably even
harder. Reducing this uncertainty and having better
predictive models is important, but ultimately we may
have to live with a high level of risk and uncertainty and
factor it into our decision - making. Fortunately, there is
a variety of tools we can employ to assist in making
robust decisions under risk and uncertainty (for a full
recent review of this topic see Regan et al. , 2009 ).
7.4.3 Climate c hange, c onservation
p lanning and a ssisted m igration
Climate change is one all - pervasive threat that protected areas cannot stop. Ever since the recognition by
Peters & Darling ( 1985 ; and see Figure 7.3 ) that some
protected areas could be rendered obsolete by climate
change, there has been interest in developing spatial
conservation plans that deliver species persistence in
the face of climate change.
One approach to conservation planning under
climate change is to protect every species in all its
ranges – present and future (Hannah et al. , 2007 ). If
we have predictive models of future species distribution
models for each species, then these future distributions
can be added as features in any conservation plan. This
simplest of approaches ignores the question of how a
species might shift its range. However, that can be