Guidance for the use of climate science to support climate change adaptation in biodiversity conservation policies

Andrew Hartley

PhD Upgrade Report

Table of Contents

1 Research problem.................................................................................................... 3

2 Literature Review..................................................................................................... 5

2.1 Quantification and Conservation of Biodiversity..............................................5

2.2 Biodiversity impacts projections............................................................................8

2.2.1 Bioclimatic envelope models...............................................................................................9

2.2.2 Mechanistic models..............................................................................................................10

2.2.3 Ecosystems models...............................................................................................................14

2.2.4 Summary of modelling approaches...............................................................................16

2.3 Climate science........................................................................................................... 17

2.4 Adaptation strategies in biodiversity conservation.......................................21

2.5 West African Climate.................................................................................................23

2.5.1 West African Monsoon........................................................................................................23

2.5.2 Land-atmosphere interactions........................................................................................24

3 Proposed Research................................................................................................ 27

3.1 Aims................................................................................................................................ 27

3.2 Key research questions............................................................................................27

4 Proposed methods of data collection and analysis.....................................30

5 Current progress.................................................................................................... 34

6 Thesis plan............................................................................................................... 38

7 Timetable.................................................................................................................. 42

8 Bibliography............................................................................................................ 43

Annex 1.............................................................................................................................. 50

Annex 2.............................................................................................................................. 50

1 Research problemInternational policy makers, land managers and conservation scientists need

reliable information from climate science to inform biodiversity impacts models

and to devise effective climate change adaptation strategies. The aim of thisis

thesis will is to address the issues surrounding the use of climate science in

developing climate change adaptation strategies for the biodiversity sector. This

includesaim can be broken down into three elements:

1. U nderstand user requirements for climate information in the biodiversity

sector,

2. A the provision ssess the results from of climate data models for use in

biodiversity impacts models, and,

3. A but also the application ofpply earth system models for advising climate

change adaptation strategies.

Consequently, the aim of this thesis is directly relevant to understanding where

the greatest impacts are expected to occur, what the potential sources of

uncertainty are, and how to best formulate climate change adaptation policies.

By taking this approach, this thesis will address how climate science can advise

the two key questions relating to biodiversity and climate change:

1. What are the impacts of climate change on biodiversity expected to be?

2. What climate change adaptation actions can be taken to reduce these

impacts?

To address question 1, cClimate change information is frequently used in

biodiversity impacts models without consideration of the reliability of climate

model projections in the locations, and for the meteorological variables, of

greatest importance. In addition, a range of spatial downscaling methods has

been applied to climate data prior to use in biodiversity impacts models, despite

there being poor knowledge of the effect different downscaling methods may

have on the projection of biodiversity impacts. Therefore, in the first part of this

thesis, I will review requirements for the use of climate data in biodiversity

impacts assessments (aim 1). Secondly, I will provide guidance on the use of

climate models in biodiversity impacts studies, by considering aspects of climate

model verification, assessment of model uncertainty and downscaling techniques

(aim 2).

Thirdly, Oonce impacts have been identified, action needs to be taken to adapt

existing conservation strategies to projected changes. Therefore, the second part

final aim of this thesis concerns the application of climate science in advising

adaptation strategies related to biodiversity and land management. Once

knowledge is gained on the types of impacts that are projected to occur, the next

issue to consider is what can be done to reduce these impacts. These decisions

frequently need to be taken at regional or local scales, therefore necessitating the

need for a specific region of interest.

This part of the thesis will focus on West Africa, because it is an important region

for globally threatened species, and it has been shown to have a strong coupling

between the land surface and the atmosphere during the monsoon period.

Recent observational studies on the West African monsoon have shown that the

existence of forest-cropland boundaries at length scales of approximately 10 to

20km can exert an influence on the initiation and distribution of precipitation in

certain parts of West Africa. This dependence between the land surface and the

atmosphere at a relatively high spatial resolution presents a challenge for

climate models to firstly represent these processes, and secondly to advise on

the how future land management might affect these processes. Since habitat

regeneration and creation are key climate change adaptation strategies in West

Africa, it is therefore necessary and timely that the effect of these strategies is

robustly assessed. Therefore, in the second part the third aim of this thesis, I will

consider how different land management practices might affect local and

regional precipitation patterns in West Africa.

2 Literature Review

The scope of this literature review is to provide a summary of the academic

literature and policy documents that are relevant to climate change impacts

studies and adaptation decisions in the field of biodiversity conservation. Given

that this covers a range of academic disciplines, such as biodiversity impacts

modelling, weather and climate science, and systematic conservation planning,

this review is not intended to be comprehensive. Relevant concepts will be

introduced, leaving the opportunity for further exploration within each chapter

of this thesis.

2.1 Quantification and Conservation of Biodiversity

The formal definition of “biodiversity” is the degree of variation of life found in a

particular location (Wilson & Peter 1988). This term usually refers to the

diversity of all plant and animal species, however it is also used to refer to

genetic diversity, or diversity at other taxonomic levels such as sub-species,

genus, family, order, class or phylum. The quantification of biodiversity is limited

by observational constraints in space and time, and as such the amount of

biodiversity that has been documented to occur on the earth is only a small

portion of that thought to exist. Mora et al. (2011) estimate that 8.1 million

species (± 1.3 million) occur on the earth, of which only 14% of species have

currently been documented on land and in the ocean. Of the species for which

taxonomic do records exist, even fewer species have documented range maps.

Despite the limitations of current knowledge on the extent of global biodiversity,

reliable measures are needed if we are to be able to predict the likely impacts of

perceived threats to biodiversity, and make effective conservation priorities.

Variables such as species richness, endemism, or richness of threatened species

for well-documented taxa (birds, mammals, amphibians) are frequently used as

surrogates for less well documented taxa. These assumptions appear to hold at

coarse spatial resolutions (~ 600,000km2), but generally, we cannot use

indicators of threatened or endemic species from one taxon as a surrogate for a

different taxon (Grenyer et al. 2006).

Figure 1. Global distribution of total species richness, endemism, and threatened species richness for birds, mammals and amphibians. Source: Grenyer et al., 2006

Biodiversity is largely protected by funding conservation projects targeted at

protecting either individual species, groups of species, habitats or ecosystems of

particular importance. However, limited resources are available for biodiversity

conservation so consequently priorities for conservation funding need to be set

(Margules & Pressey 2000). Internationally, the main sources of conservation

funding are organisations such as the World Bank, the United Nations, and the

European Union, while non-governmental organisations such as the World

Wildlife Fund, Conservation International, BirdLife International, and The Nature

Conservancy also raise funds from donations to protect biodiversity. Each of

these organisations requires robust science-based advice on how to prioritise

their conservation funding. However, each organisation has a different approach

to priority setting, some focusing on a particular taxon, such as Birds (e.g.

BirdLife International; Fishpool & Evans 2000), whilst others focus on

ecosystems via internationally recognised protected areas (e.g. European

Commission; Hartley et al. 2007). Brooks et al. (2006) have proposed that most

conservation prioritisations fit into a conceptual framework of irreplaceability

compared to vulnerability. For example, the prioritisation template used by

Conservation International (Biodiversity Hotspots; Myers et al. 2000) prioritises

highly irreplaceability and highly vulnerability locations. In contrast, the

prioritisation template used by the World Wildlife Fund prioritises highly

irreplaceable locations only (G200; Olson & Dinerstein 1998, 2002).

Figure 2. Maps of nine global biodiversity conservation priority templates: CE,crisis ecoregions(21); BH, biodiversity hot spots [(11), updated by (39)]; EBA, endemic bird areas (15); CPD, centers of plant diversity (12); MC, megadiversity countries (13); G200,global200 ecoregions [(16), updated by (54)]; HBWA, high-biodiversity wilderness areas (14); FF, frontier forests (19); LW,lastofthe wild (20). Source: Brooks et al. (2006)

In the context of climate change, the challenge for conservation policy makers is

to incorporate information from climate science into existing prioritisation

frameworks, so that future conservation challenges can be identified, and

suitable adaptation measures devised and tested. It has been suggested that this

is the most cost-effective approach to setting conservation targets for the future

(Hannah et al. 2007).

The G200 ecoregions – defined as highly irreplaceable ecosystems that represent

a disproportionately large number of the world’s species – have been assessed

for their exposure to mean monthly temperature change (Beaumont et al. 2011).

The aim of such an approach is to advise future conservation funding at a global

scale. In contrast, Hole et al. (2009) took this approach further by quantifying

inward and outward migration of bird species from sites of high conservation

priority (BirdLife International’s Important Bird Areas; Fishpool & Evans 2001).

In doing so, they allow a further step of suggesting site-specific climate change

adaptation strategies based on species turnover at each site (Hole et al. 2011).

Another approach to assessing the impacts of climate change on conservation

goals is to focus on how climate change might impact the ecosystems in which

species live, rather than the species themselves. This can bring benefits of

framing biological impacts in the context of global mean temperature targets or

greenhouse gas emissions scenarios. For example, Mahlstein et al. (2013) use a

simple climate-ecosystem classification to show how the pace of climate change

for global ecosystems increases as global mean temperature increases. An

extension of this approach is to use our understanding of plant physiology to

model how different types of ecosystems may respond to climate and

environmental stresses. The following section will review the relative merits of

all approaches to climate change.

2.2 Biodiversity impacts projections

Given the challenges of quantifying biodiversity in situ, it may be considered an

even greater challenge to estimate how climate change might affect biodiversity

and related conservation goals. Species’ are thought to respond to climate

change by adapting their behaviour (Menzel et al. 2006), developing

evolutionary adaptations (Parmesan 2006) or by shifting their range (Thuiller

2004), in accordance with the rate and magnitude of change (Huntley et al.

2010). Fossil evidence has shown that some species respond to periods of

warming and cooling by shifting their ranges to track a niche climate. For

example, evidence from the Quaternary period has shown that species range

shifts may be the most likely response to future change (Davis & Shaw 2001).

In order to answer the question of how a species will respond to climate change,

ecologists have developed a variety of approaches. These include estimates of

how a species’ climatic niche may shift, how the population dynamics of a species

may change, or the assessment of a species’ traits that might make it vulnerable

to climate change. Other approaches consider how habitats, ecosystems or

biomes may respond to climate change, as a surrogate for biodiversity. In order

to consider the role of climate science in improving biodiversity impacts

assessments, it is first necessary to describe, and understand these approaches.

2.2.1 Bioclimatic envelope models

The concept of a species occurring within defined environmental tolerances was

first described by Grinnell (1917), who suggested in his study of the Californian

Thrasher that “the nature of these critical [environmental] conditions is to be

learned through the examination of the bird’s habitat”. Bioclimatic envelope

models (BEMs) are statistical models that correlate the observed range of

occurrence of a species with climatological and other environmental variables.

These correlations are then used to estimate future shifts in a species niche

under climate change. BEMs have been used extensively in statistical ecology to

predict future conservation priorities (Hannah et al. 2002; Midgley et al. 2002)

and to infer extinction risk to species (see for example Thomas et al. 2004;

Hannah et al. 2005; Maclean & Wilson 2011).

The projection of a species’ future range is a significant advantage of the BEM

approach. This future range can then be used to form conservation management

plans under climate change. For example, Hole et al. (2011) propose site-based

climate change adaptation strategies for African Important Bird Areas (IBAs)

according to BEM projections of inward and outward migration.

A further advantage of the BEM approach is that it allows the possibility for the

assessment of uncertainty deriving from different sources in the modelling

process (Heikkinen et al. 2006; Buisson et al. 2010; Garcia et al. 2011). One such

source of uncertainty is the choice of which model to employ. A large number of

predictive statistical models have been applied in this context (often varying in

complexity), which include linear regression, additive models, machine learning

techniques (Phillips & Dudík 2008) and hierarchical Bayesian techniques

(Gelfand et al. 2006). Other sources of uncertainty in the BEM approach derive

from uncertainty in observations of the current distribution of the species;

uncertainty in future greenhouse gas emissions scenarios; and the uncertainty in

the projections of climate change from General Circulation Models (GCMs).

Despite the popularity of the BEM approach, it is also clearly has significant

limitations (Wiens et al. 2009). One such limitation is related to the

interpretation of the current range of a species. BEMs assume that the observed

occurrence of a species is a representation of the stable fundamental niche of

that species. However, in practice, the observed occurrence of a species (and

therefore the present day realisation of it’s fundamental niche) is also influenced

by factors, such as species adaptation (or acclimation), dispersal ability, biotic

interactions with other species and human disturbance, which result in the

realisation of only a portion of the fundamental niche (Hampe 2004). Therefore,

as the climate changes, the assumptions on which BEMs are based would not

account for a species realising a different portion of the fundamental niche. One

case in which this limitation may become evident is in the emergence of future

climates where no present-day analogue exists. This limitation of BEMs has been

demonstrated, for example, using fossil-pollen records from the late quaternary

period. Veloz et al. (2012) showed that for species that were abundant in areas

with no present day analogue climate, BEMs were poor predictors of the current

species distribution. They therefore imply that the species significantly shifted

their realised niches from the late glacial period to the present day.

2.2.2 Mechanistic models

The identification of the risks of species becoming extinct under future climate

change is often the motivation for many impacts studies. However, the

application of BEMs for the assessment of extinction risks has raised several

methodological concerns (e.g. Thomas et al. 2004; Thuiller et al. 2004).

Extinction risk is inferred when a species’ projected future range is either

completely dislocated or reduced in size relative to the present day range. As the

geographical area of the niche habitat for the species reduces, so too does the

assumed population of that species, resulting in an increase in probability of

extinction. This relationship between a species’ geographical area of extent and

population size is a concept that is central to biogeography (MacArthur & Wilson

1967), and has been rigorously tested in relation to contraction of habitat area

(Diamond 1972). However, in reviewing the applicability of the species-area

relationship to climate change studies, Lewis (2006) suggests that there are

significant limitations. These include the assumption that a species populates

evenly its range, uncertainty as to the influence of climate over a species current

distribution, and uncertainty as to whether the species is currently filling its

fundamental niche (discussed above).

An alternative, or potentially complementary, approach to BEMs is rooted in a

more mechanistic understanding of the extent to which weather and climate

affect the population dynamics of a species. This may include interactions

between species, between the species and suitable habitats, and the demography

of the species (Maschinski et al. 2006). Such a model has been applied spatially

for plant species populations in the South African Fynbos (Keith et al. 2008). In

this case study, annual climate was used to drive a habitat suitability model,

which was used to calculate carrying capacity of a habitat patch for a given

species for use in a stochastic population model (Figure 3). The results from

Keith et al. (2008) indicate that complex interactions between life history,

disturbance regime and distribution pattern can affect the assessment of

extinction risk. Furthermore, by addressing population mechanisms directly,

they avoid making over-simplifications of the link between habitat suitability

and species populations.

Figure 3. Schematic of a habitat suitability model coupled with a stochastic population model. Each simulation starts at step 1, and after the first cycle is complete (step 5), subsequent cycles include step 6. Source: Keith et al. (2008)

Another mechanistic approach to the relationship between a species and climate

is described in the field of biophysical ecology. Here, the principles of

thermodynamics are applied to organisms in order to develop a mechanistic

understanding of the processes affecting them, and their physiological responses

to change in these processes (Porter & Gates 1969). Specifically, biophysical

models concern the transfer of heat, biomass and momentum from the

environment to the energy budget of the organism (Kearney & Porter 2009).

Figure 4. Energy transfer from the environment to an animal species. Source: Porter & Gates (1969)

The mechanistic understanding of species’ biophysical interaction with the

environment can inform key traits such as body temperature, energy budget and

water balance. In turn, this can inform the assessment of a species survival and

reproduction rates, and as a consequence becomes a means of quantifying the

fundamental niche of a species (Kearney & Porter 2009). Furthermore, this

fundamental niche can then be mapped, and potentially combined with a more

basic correlative (BEM) approach to form a consensus view of future species

distribution. Buckley et al. (2010) tested how projections based on the

correlative approach differ from those based on mechanism for two species: the

sachem skipper (Atalopedes campestris) and the eastern fence lizard (Sceloporus

undulates). They found generally that correlative models and mechanistic models

performed similarly in estimating the current range of both species, although

correlative models had greater success in identifying the western limit of S.

undulates. Additionally, mechanistic models predicted greater range shifts under

a uniform 3°C warming scenario. Similarly congruent results have also been

found using both mechanistic and correlative models for a species of Australian

possum (Kearney et al. 2010).

The mechanistic approach to species modelling has the advantage of

incorporating physiologically based environmental constraints that influence

both the distribution and abundance of a species (Kearney & Porter 2009). These

physiological processes are strongly related to flows of mass and energy as a

species interacts with its environment. Therefore, by understanding such

processes the impacts of climate change on biodiversity can be assessed without

reliance on observations of uncertain range limits under current climatic

conditions. However, a limitation of the application of this approach to the

regional or global scale is clearly the time and effort involved in understanding a

species’ physiology and environmental constraints.

2.2.3 Ecosystems models

An alternative to modelling species responses to climate change is to model the

response of the habitat or ecosystem instead. The idea of ecosystems being

controlled by large-scale climatic factors was first developed in the late 19th, and

early 20th centuries (Von Humboldt 1867; Koeppen 1900; Geiger 1961;

Holdridge 1967) alongside theories of biogeography. The Holdridge Life Zone

system (Holdridge, 1967) is one such classification. It has the advantage of being

relatively simple to implement whilst allowing the objective relation of

temperature and precipitation variables to potential biomes, altitudinal zones or

potential vegetation types (the combination of which was termed “Life Zones” by

Holdridge). Essentially, these types of climate-vegetation classifications are

similar to the correlative approach used in BEMs. An important caveat in this

approach is acknowledged in the term 'potential'. Climate is only one of many

factors that contribute towards determining the existence of a particular

vegetation type at a given time and location. Other factors that may influence

vegetation type, such as CO2 effects, ozone, nutrient availability and soil

condition are not accounted for by the Holdridge system. Nevertheless, similar

approaches to both the Koeppen-Geiger classification and the Holdridge Life

Zone System are still used in modern climate change impacts studies (Lugo et al.

1999; Velarde et al. 2005; Kottek et al. 2006; Good et al. 2011; Metzger et al.

2013).

More recently, mechanistic models of vegetation physiology have been

developed (Box et al. 1981; Prentice et al. 1992; Sitch et al. 2003; Clark et al.

2011), in much the same way as models of animal physiology. These dynamic

global vegetation models (DGVMs) model not only physiological differences

between functional groups of plants, but also account for differences in

allometry, morphology, phenology, bioclimate and response to disturbances such

as fire. As such, DGVMs characterise how the vegetation (or land surface)

responds to the atmosphere, and how the atmosphere responds to the

vegetation cover, via fluxes of heat, moisture, carbon and momentum. DGVMs

quantify these interactions at daily, monthly and annual time steps, thus allowing

the influence of both large scale limiting factors (such as CO2 concentration,

climate, altitude and soil), and seasonally dependent factors (such as leaf

phenology, water balance, evapotranspiration, snow cover, and soil

temperature). Therefore, DGVMs are valuable tools for modelling the terrestrial

carbon and water cycles as well as the response of large-scale ecosystems to

climate change.

While DGVMs coupled with climate models are powerful tools for making

predictions of the effects of climate change on vegetation and the carbon cycle

(Cramer et al. 2001; Friedlingstein et al. 2006), their limitations include

significant divergence of projections especially under more extreme greenhouse

gas emissions scenarios (Sitch et al. 2008). One possible cause of this divergence,

which is also a significant limitation for the application of DGVMs in conservation

science, is the low thematic detail in their definition of plant functional types

(PFTs). PFTs are defined as groups of species or taxa that exhibit similar

responses to physical or biotic changes in the environment. It has been noted

that the low number of PFTs in DGVMs (typically 5 to 15), the ad hoc definition of

their parameters, and the lack of integration with current research in functional

ecology are some of the limitations of DGVMs (Harrison et al. 2010). Boulangeat

(2012) suggest a hybrid approach to identify a minimum set of plant traits to link

plant functional groups to species diversity, as well as dynamic vegetation

models at the regional scale. The approach has the potential to be applied in

other regions, however it is yet to be used to account also for vegetation

dynamics.

2.2.4 Summary of modelling approaches

The following table summarises the main advantages and disadvantages of the

approaches to modelling the impacts of climate change on biodiversity.

Modelling Approach

Brief description Advantages Disadvantages

Bioclimatic envelope models

Statistical models that correlate the observed range of occurrence of a species with climatological and other environmental variables

- Projection of future range useful for adaptation planning- Only observations of species presence in a location is required- Models can be tested by withholding a fraction of species observations- Multiple sources of uncertainty can be quantified- Quick and easy to use - Can be applied to all species with an observed geographical range map

- Assumes present-day distribution is a representation of the stable fundamental niche of a species- Assumes a species populates evenly its range, - Uncertainty as to the influence of climate over a species current distribution- Incomplete and inconsistent observations of species occurrence- Assumes that species-area relationship can be used to identify extinction risks- Does not account for species adaptation to climates with no present day analogy

Mechanistic models

Process-based models to understand the extent to which weather and climate affect the population dynamics of a species.

- Incorporates physiologically based environmental constraints that influence both the distribution and abundance of a species- Incorporates dispersal ability and competition between species- Can be based on in-situ and laboratory observations of a species

- Requires a detailed understanding of a species, consequently more labour intensive than BEMs- Adequate data only available for a few species, therefore not viable for assessing large scale impacts on species- Despite a more holistic approach, comparisons

population response to change- Quantifies complex interactions between life history, disturbance regime and distribution pattern that can affect the assessment of extinction risk

with BEMs show largely congruent results for predicting species’ current ranges

Ecosystem models

Mechanistic models of vegetation physiology for understanding the response of ecosystems to change.

- Models changes in ecosystem or habitat distribution which may have a greater influence on where a species occurs- Fully coupled with climate models- Changes in biomass and other vegetation related indices may affect herbivores and consequently other aspects of the food chain

- Direct links between biomass production and biodiversity are difficult to prove at local and regional scales- Low thematic detail of plant functional types means that it is difficult to relate results to habitats- projections diverge especially under more extreme greenhouse gas emissions scenarios

Table 1 Summary of the advantages and disadvantages of the modelling approaches to the impacts of climate change on biodiversity.

2.3 Climate science

General Circulation Models (GCMs; also termed Global Climate Models) are the

main tools employed in climate science. GCMs are numerical models of physical

processes that occur in the dynamical earth system. This can involve processes in

the atmosphere, ocean, land surface and cryosphere. GCMs are used to run

experiments on the response of the earth system to different potential drivers of

change. The ultimate aim of a GCM is to provide a realistic simulation of the

major global scale physical processes that occur in the earth system. GCMs

typically are designed to run at horizontal grid resolutions of approximately 100

to 300km2, with 10 to 20 vertical layers in the atmosphere, and up to 30 vertical

layers in the ocean. The can be initiated from observations of historical climate,

and whilst running can be forced with different global concentrations of

greenhouse gases.

A key factor for determining future climate change will be the quantity of

greenhouse gas emissions. These will depend on the global population; it’s

lifestyle, and the way this is supported by the production of energy and the use of

the land. A large population whose lifestyle demands high energy consumption

and the farming of large areas of land, in a world with its main energy source

being fossil fuel consumption, will inevitably produce more greenhouse gas

emissions than a smaller population requiring less land and energy and deriving

the latter from non-fossil sources. These factors could vary in a multitude of

ways; the international community is already examining how energy demand

and production can be modified to cause lower emissions, but the

implementation of this will depend on both the international political process

and the actions of individuals. Even if no specific action is taken to reduce

emissions, the future rates of emissions are uncertain since the future changes in

population, technology and economic state are difficult if not impossible to

forecast. Therefore, rather than make predictions of future greenhouse gas

emissions, climate science examines a range of plausible scenarios in order to

examine the implications of each scenario and inform decisions on reducing

emissions and/or dealing with their consequences.

The climate models that have contributed towards the Intergovernmental Panel

on Climate Change (IPCC) Forth Assessment Report (AR4) have generally used a

set of scenarios known as “SRES” (Special Report on Emission Scenarios;

Nakicenovic et al. 2000). These scenarios were grounded in plausible storylines

of the human socio-economic future, with differences in economy, technology

and population but no explicit inclusion of emissions reductions policies.

Developed in the mid 1990s, these scenarios extended out to 2100 and varied

widely in their projected emissions of greenhouse gases. For the next IPCC

Assessment Report, GCMs have been forced with radiative forcing from 4

different Representative Concentration Pathways (RCPs) that are described in

terms of greenhouse gas concentrations at the end of the 21st Century in

equivalent CO2 concentration values (Vuuren et al. 2011). The RCPs represent

total radiative forcing of greenhouse gases rather than a particular scenario of

emissions.

Figure 5. Radiative forcing from the 4 Representative Concentration Pathways developed for the IPCC AR5. Source: van Vuuren et al. (2011)

Biodiversity impacts model tend to use a very small subset of the variables that

are produced by GCMs. A frequently used set of bioclimatic variables, based on

temperature and precipitation indices, is described in Hijmans et al. (2005; see

table 2). These variables are derived from time averaged monthly minimum and

maximum temperature and monthly precipitation fields, and have been shown to

be biologically relevant variables in the prediction of a species' climatic niche

(Hijmans & Graham 2006). A large number of papers use these variables for

fitting bioclimatic envelope models (see for example Elith et al. 2006; Hijmans &

Graham 2006; Garcia et al. 2011).

Table 2. Bioclimatic variables frequently used in correlative species models following Hijmans et al. (2005)

Bioclimatic variable DescriptionMean annual temperature Mean annual temperature (tas)Mean diurnal range Annual mean of monthly (tasmax – tasmin)Isothermality (Mean diurnal range / Temperature annual

range) * 100Temperature seasonality Standard deviation * 100Max temperature of warmest month

Maximum monthly temperature

Min temperature of coldest month

Minimum monthly temperature

Temperature annual range Max temp of warmest month – min temp of coldest month

Mean temperature of wettest quarter

Mean temperature of wettest quarter

Mean temperature of driest quarter

Mean temperature of driest quarter

Mean temperature of warmest quarter

Mean temperature of warmest quarter

Mean temperature of coldest quarter

Mean temperature of coldest quarter

Total annual precipitation Total annual precipitationPrecipitation of wettest month Maximum monthly precipitation ratePrecipitation of driest month Minimum monthly precipitation ratePrecipitation Seasonality Coefficient of variation of monthly

precipitationPrecipitation of Wettest Quarter Maximum quarterly precipitation ratePrecipitation of Driest Quarter Minimum quarterly precipitation ratePrecipitation of Warmest Quarter

Precipitation rate in warmest month

Precipitation of Coldest Quarter Precipitation rate in coldest monthTable 3 Bioclimatic variables frequently used in correlative species models following Hijmans et al. (2005)

All of the above variables are dependent on observations of precipitation and

minimum, mean and maximum temperature being available. Long time series of

quality controlled meteorological observations are important for use in

understanding historical trends, model biases and making valid associations

between a species’ range and the above climate variables. Table 3 shows the

observational datasets that are available for West Africa, the case study area that

will be described later in this report.

Dataset Source Relevant Climate Variables

Spatial Resolution

Temporal Coverage

Temporal Resolution

GSOD Meteorological Stations (via WMO)

Temperature, wind, precipitation, snow depth

Point locations with distribution varying through time

1901 – present

Daily

CRU TS3.1 Interpolated observations from WMO

Temperature, precipitation, PET

~55km 1901-present

Monthly

WorldClim Interpolated observations from WMO

Min, mean and max temperature,

Varying from 1km to ~19km

1950 – 2000 Monthly climatology

and other national sources

precipitation

GPCP Observations and satellite

Precipitation 2.5 degree 1979 – present

Monthly

TRMM Satellite using observations for ground truthing

Precipitation 0.25 degree 1998 – present

3-hourly to monthly climatology

FEWSNET Satellite using observations for ground truthing

Precipitation 8km 1995 – present

10-daily

Table 4 Observational datasets available for applications in biodiversity impacts models for West Africa. Acronyms: GSOD = Global Summary of Day; NOAA/NCDC = US National Ocean and Atmospheric Administration / National Climate Data Center; CRU = Climate Research Unit, University of East Anglia; PET = Potential Evapo-Transpiration; GPCP = Global Precipitation Climatology Project; TRMM = Tropical Rainfall Measuring Mission

2.4 Adaptation strategies in biodiversity conservation

One of the main science-policy related issues that the new Intergovernmental

Platform on Biodiversity and Ecosystem Services (IPBES) is likely to address is

how to adapt current conservation strategies to the projected impact of climate

change on biodiversity (Perrings et al. 2011). The global network of protected

areas (henceforth PAs) has been shown to be an effective tool for the protection

of biodiversity (Bruner et al. 2001), and is considered to be one of the principal

tools for protecting biodiversity and implementing strategies to adapt to climate

change (Mawdsley et al. 2009). However, despite the majority of global

biodiversity occurring in tropical developing countries, there is limited evidence

in the academic literature of adaptation actions being undertaken in the

developing world (Ford et al. 2011). Ford et al. (2011) found that currently,

adaptation actions are most frequently being implemented in the infrastructure

and utilities sectors as opposed to ecosystem management or forestry.

In their review of climate change adaptation plans for wildlife management in

USA, Canada, England, Mexico, and South Africa Mawdsley et al. (2009) suggest

four main categories into which adaptation strategies can be grouped: land and

water protection and management; direct species management; monitoring and

planning; and law and policy. In particular, they suggest increasing the extent of

protected areas, restoring and creating new habitat in order to maximise future

resilience, and increasing landscape connectivity. Similar suggestions also put

forward by Hannah et al. (2008) for climate change adaptation in Madagascar.

They suggest that restoration and protection of the riverine corridor forests are

important for species migrations, especially between fragmented habitats with

high genetic divergence between populations. However, the costs of this strategy

are high. Restoring forests to maintain connectivity between Madagascar’s

fragmented forests would cost approximately US$0.8 billion, albeit with an

estimated extra income of US$ 72 - 144 million annually from the post-Kyoto

protocols on reducing emissions from deforestation and degradation (REDD).

Another approach to adaptation planning that is relevant for site-based

conservation management involves the integration of projections of species

range shifts from BEMs. As species ranges shift with climate change, species will

no longer occur in some sites, but will colonise new sites. Hole et al. (2011)

suggest different adaptation strategies for Important Bird Area sites

experiencing different types of inward or outward migration (Figure 6).

Figure 6. (a) Proportion of priority bird species projected to emigrate relative to proportion of priority species projected to colonize (log scale) by 2085 each of the 803 mainland sub-Saharan Africa Important Bird Areas (IBAs). Climate-change adaptation strategy (CCAS) categories into which IBAs are classified are purple, high persistence; green, increasing specialization; red, high turnover; blue, increasing value; yellow, increasing diversification. (b) Spatial distribution of IBAs in the five CCAS categories. Source: Hole et al. (2011)

Following the identification of CCAS categories, Hole et al. went on to suggest

different adaptation strategies according to each category. This has the

advantage of offering site-specific advice on how to manage habitats for expected

future migrations of globally threatened bird species.

Finally, once impacts have been identified, and adaptation strategies agreed

upon, the final stage in the process is to implement plans. Heller and Zavaletta

(2009) review the biodiversity adaptation literature, and issue several

recommendations for actions. They stress the importance of regional institutions

to coordinate adaptation projects, the incorporation of climate change into all

areas of planning and policy, and an inclusive approach to local communities.

Furthermore, they suggest that many adaptation plans lack detailed information

on who how the plan will be implemented and by whom. They suggest that

regional planning, site scale management and modification of existing

conservation plans is the best way to over come these issues.

2.5 West African Climate

Rainfall is extremely important for the livelihoods and ecosystems of West

Africa. With a rapidly growing population that is dependent on subsistence

agriculture, and intense human pressure on the remaining natural ecosystems, it

is vitally important to be able to understand and predict the dynamics of the

West African monsoon for the present day and into the future.

2.5.1 West African Monsoon

The dominant feature of the West African climate is the monsoon period

between May and October. The main driver of seasonal variations in the West

African climate is the north-south movement of the Inter-Tropical Convergence

Zone (see ITD in Figure 7). During the period of the monsoon, low-level flows of

moist air are pulled inland from the Gulf of Guinea. This is due to a pressure

gradient between the land and sea created by warm sea surface temperature

cooling as the move north to form high pressure, and the Saharan heat low

creating low pressure and anti-cyclonic winds. As the moist air from the sea

converges with hot dry northerly winds from the Sahara, convective cells form

and develop into large-scale convective cells. During the first part of the

monsoon, precipitation is concentrated over the coastal areas, until the

beginning of July when the monsoon jumps northwards by approximately 8°N

latitude (Figure 8). After remaining at a latitude of approximately 12°N for July

and August, the zone of peak rainfall gradually retreats southwards towards the

coast. Coastal areas therefore experience a second rainy season in October.

Figure 7. Three-dimensional schematic view of the West African Monsoon. ITD, inter-tropical discontinuity; TEJ, tropical easterly jet; STWJ, subtropical westerly jet; AEJ, African easterly jet. The oscillation of the AEJ yellow tube figures an African easterly wave. Source: Lafore et al. (2011).

Figure 8. The mean seasonal cycle of rainfall over West Africa through a latitude cross-section. March–November daily precipitation values (mm/day) from GPCP satellite-estimated values are averaged over 5 ◦ W – 5 ◦ E and over the period 1997 – 2006. A 7-day moving average has been applied to remove high-frequency variability. The black horizontal line at 5◦N represents the Guinean Coast. Source: Janicot et al. (2011)

2.5.2 Land-atmosphere interactions

There are considerable social, economic and environmental stresses on land use

in West Africa. In situ measurements (Garcia-Carreras et al. 2010) and

observational studies (Taylor et al. 2007) have shown that during certain times

of day, the land surface is strongly coupled to the planetary boundary layer. As a

consequence, the land surface (in particular soil moisture gradients) has been

shown to have an important role in the initiation of convection (Taylor et al.

2011) and in modulating where precipitation falls in the region (Taylor et al.

2012). In addition to soil moisture, there is growing evidence that forest-

cropland or forest-grassland gradients can create similar gradients of heat,

moisture and momentum fluxes in the atmosphere, thus also having an influence

on the initiation of convective rainfall. Therefore, it would be reasonable to

hypothesize that land use policy has the potential to influence local and regional

scale precipitation patterns in West Africa.

Figure 9. The mechanism by which convection is proposed to initiate over forest boundaries. Source: Garcia-Carreras et al. (2010)

summarises the mechanism by which convection initiates over forest

boundaries. As cool moist air flows at low levels over the forest, when it reaches

the forest boundary it meets dry northerly winds from the Sahara, and warm

land with a high albedo and subsequently rapidly rising air mass. This

convergence results in a high convective available potential energy (CAPE),

which effectively results in the vertical movement of moist air that condenses to

form cumulus and cumulus congestus clouds on the southern side of the warm

grassland.

3 Proposed Research

3.1 Aims

To provide guidance on the use of climate science for advising climate change

adaptation strategies in biodiversity conservation policy.

3.2 Key research questions

Since the aim of this thesis is potentially very broad, I intend to frame my

research questions in the context of the different stages involved in developing a

conservation adaptation policy (Figure 10). Before addressing the research

questions that this thesis will attempt to answer, it is helpful to understand the

process by which biodiversity conservation adaptation decisions are taken. In

the final thesis, these assumptions will be verified by consulting with

conservation policy makers (see Thesis plan, Chapter 1).

Figure 10. Generalised schematic of the key questions in the process of setting future conservation priorities and adaptation strategies. Blue boxes denote key policy and practical conservation questions; green boxes denote key science questions; red boxes denote how this thesis will contribute to each stage of the above processes.

In general, policy makers and conservation practitioners need to know what the

impact of climate change will be on the species of greatest conservation value,

and what actions to take to reduce these impacts The key questions that arise in

conservation policy related to climate change can be broken down into the

following (shown in blue in Figure 10):. The key research questions that this

thesis will address are as follows:

[1.] What is the impact of climate change on biodiversity likely to be?

[2.] What adaptation actions can be taken to reduce these impacts?

1. What are user requirements for climate information in the biodiversity

sector?

2. How suitable are climate model outputs for use in biodiversity impacts

models?

3. How can earth system models advise climate change adaptation strategies in

West Africa?

In relation to question 1, international conservation funding organisations, such

as the United Nations, World Bank or European Commission, might require this

climate change information to set funding priorities. The aim of these

organisations is to target resources to the habitats and species that are projected

to be most severely impacted by climate change. National governments may also

require this information in order to advise setting global emissions reductions

targets. For example, quantification of the impacts of different greenhouse gas

emissions scenarios on global biodiversity may provide useful information on

what level of climate change is considered ‘dangerous’.

The question of how biodiversity might be impacted by climate change is also

relevant for policy makers at regional or national scales. Information on

biodiversity impacts of climate change may provide information on the direction

of expected species migrations, or the expected persistence of unique habitats.

Currently, scientists in the fields of conservation biology and climate science are

the main providers of information to inform these decisions (Figure 10, green

boxes). While a large amount of the literature addresses modelling approaches

and uncertainties in biodiversity impacts, little attention has been paid to the

most appropriate use of climate science and the effect of different methods of

downscaling climate information on conservation decisions (question 2).

Therefore, this thesis will address the following questions:

1. How reliable are climate models for applications in biodiversity impacts

studies?

2. How do uncertainties in climate projections propagate into conservation

priorities in West Africa?

Once global, regional and national scale priorities for biodiversity conservation

have been set, it is then necessary to develop effective adaptation strategies

(question 3). The aim of adaptation in biodiversity conservation is either to

improve the resilience of species and habitats or to assist their response to

climate change in order to prevent extinctions. Typically, climate science has not

been involved in advising adaptation strategies. However, this is now becoming

feasible with the inclusion of land surface processes in many established climate

models, and advances in the understanding of land-atmosphere interactions. The

case study of West Africa provides particularly interesting challenges and

opportunities for assessing climate change adaptation options. Therefore, the

third research question addressed in this thesis will be:

3. Do forest related climate change adaptation strategies affect local and regional

precipitation patterns in West Africa?

4 Proposed methods of data collection and analysis

Firstly, I will collect information to guide this study by canvasing expert opinion

on the key challenges for policy makers and conservation scientists with regard

to the impacts of climate change on biodiversity. The main source of data for this

thesis however, comes from climate and meteorological models. As this thesis

progresses through the process of translating climatological information into

adaptation advice, I will employ models ranging from coarse resolution global

models to fine resolution meteorological models. Each data source is discussed

in more detail below.

Expert survey of key issues

For chapter 1, I intend to devise a questionnaire to be distributed amongst

attendees at an international climate change and nature conservation conference

to be held in Bonn on 25th to 27th June 2013. The questionnaire will ask

participants for their experiences in using climate change information in the field

of conservation science. The aim of the questionnaire will be to understand what

are the challenges and gaps in climate data currently available, and to identify

potential applications for climate science in developing climate change

adaptation policies. The methodological issues that are relevant to this approach

are developing a questionnaire that can be both quantitative, and allows for the

collection of expert opinion. This might be overcome by creating preliminary

questions to understand the role of the participant, and their level of

understanding of key issues. The questionnaire is being developed with input

from conservation scientists and policy experts from the Royal Society for the

Protection of Birds (RSPB) and The British Trust for Ornithology (BTO).

General circulation models

In order to assess the reliability of General Circulation Models (GCMs) for

chapter 2, I compared all the models published under the Fifth Coupled Model

Inter-Comparison Project (CMIP5) to the CRU_TS3.1 observations for the

historical period between 1950 and 2000. The methodological steps involved

firstly calculating 19 bioclimatic variables commonly used in biodiversity

impacts projections that are based on monthly minimum and maximum

temperature and monthly total precipitation. I then tested the ability of each

model to simulate the observed inter-annual variability of each variable. As an

additional step, I also tested the ability of each GCM to simulate the observed

mean seasonal cycle of a subset of variables. A full explanation of the

methodology can be found in Annex 1.

In order to show the impacts of climate change mitigation on global ecosystems, I

used a perturbed physics ensemble of projections from the HadCM3 GCM. For

each model ensemble member, I calculated a simple metric (Hdistance) using

bio-temperature and annual precipitation, based on the Holdridge Life Zone

classification system. The Hdistance metric provides a measure of change that is

relevant for global ecosystems, using variables for which GCMs are largely

considered reliable. Then, using the Hdistance, I identified critical thresholds in

the rate and magnitude of change that may be relevant for large scale

ecosystems. These critical thresholds were assessed for a climate change

mitigation scenario (RCP2.6) and a business as usual scenario (A1B) to identify

how climate change mitigation may reduce damaging impacts on the worlds

ecosystems.

Regional climate models

The model simulations were run from December 1949 to December 2100 using

the HadRM3 regional climate model with the MOSES2.2 tiled land-surface

scheme and the A1B SRES scenario, on the 50km resolution Africa CORDEX

domain (Giorgi et al. 2009). They provide a comprehensive dataset of surface

and atmospheric climate variables including minimum and maximum

temperatures and precipitation at the daily and monthly timescale and at a

spatial resolution of 50km.

The lateral boundary conditions for the simulations were taken from a subset of

5 ensemble members sampled from a perturbed physics ensemble based on the

HadCM3 GCM. The model selection was primarily based on regional analysis of

the GCMs for Africa and its sub-regions with a focus on several regions including

West and Central Africa. Members of the ensemble were selected in order to

capture the spread in outcomes produced by the full ensemble, whilst excluding

any members that do not represent the African climate realistically.

The RCM simulation for Africa have already been run (by others within the Met

Office), and pre-processed by myself. In chapter 4, I attend to assess the skill of

these RCM simulations over West Africa, and discuss this suitability for use in

BEMs. In chapter 5, the results of the dynamical downscaling method will be

compared to less computationally expensive and more frequently used

downscaling methods, in order to provide guidance on the advantages and

limitations of each approach to downscaling.

High spatial and temporal resolution limited area models

The advantage of using limited area models (LAMs) at 4km spatial resolution,

with time steps every 10 seconds, is that they are sufficiently high resolution to

model convective rainfall. Coarse resolution models such as GCMs and RCMs are

typically run for grid resolutions of between 200km and 50km at between 3-

hour and 1-hour time intervals. The resolution of these models is not sufficient to

model the typical scale of this process. Therefore, GCMs and RCMs must employ a

statistical parameterisation of convection that estimates the amount of

convective rainfall based on the values of other more variables in the

atmosphere.

A nested LAM will be used in this thesis to firstly identify interactions between

land cover and the atmosphere during a period of 4 days during the West African

monsoon. This model has already been run, and will contribute towards chapter

6. For chapter 7, the same LAM will be run for one full season of the West African

monsoon period (approximately 3 months). In this second experiment, different

land cover configurations will be tested for their influence on local and regional

scale precipitation patterns. The land cover configurations will represent

different climate change adaptation options for the region as follows:

- a “great green wall” across the Sahelian belt,

- a policy of increasing protected area connectivity by forest regrowth and

planting policies

- an increasing degradation of forest around the boundaries of current

protected areas.

Climate observations

For all chapters, there will be an element of model evaluation and assessment. In

order for this to occur, robust observations datasets are needed, covering both

the global extent at coarse spatial resolutions and the West African domain at

higher spatial resolutions. The challenges of working in this area are limited

availability of meteorological station observations, variable quality of

observations, and inconsistent recording. For this reason, in addition to station

based gridded observations, I will also compare model results to observations

from satellites (such as data from the Tropical Rainfall Measuring Mission, and

FEWSNET), to model based reanalyses of surface pressure observations (such as

the NCEP 20th Century Reanalysis), and to hybrid products combining

observations and reanalysis (such as the CPC Merged Analysis of Precipitation).

5 Current progress

Currently, I have submitted two articles to peer reviewed journals. I have

provided the abstracts below, and the manuscripts are uploaded as separate

documents to this upgrade report. These two articles will be chapter 2 and

chapter 3 of my thesis respectively.

For each chapter, I outline below progress and possible problems that I may

encounter.

Chapter 1

I currently have a basic draft of the questionnaire that will be distributed at a

climate change and nature conservation policy conference in later June. The aim

of the questionnaire is to gain insight into the issues that are currently relevant

to policy makers and conservation scientists relating to climate change, and to

understand where climate science can better advise such decisions. The

questionnaire will be both printed, and distributed via the SurveyMonkey

website. Potential problems may include a low response rate and questions that

do not address the main issues in the field.

Chapter 2: Submitted article

Title: The reliability of the CMIP5 GCM ensemble for assessing the impacts

of climate change on biodiversity

Authors: Andrew Hartley, Jon Olav Skøien, Gregoire Dubois

Abstract: Following the inception of the Intergovernmental Platform on

Biodiversity and Ecosystem Services (IPBES), there is a renewed focus on the

adaptation of existing biodiversity conservation strategies to climate change, and

consequently the validity of General Circulation Models (GCMs) for such

applications. Here, we assess the ability of the Coupled Model Inter-Comparison

Project 5 (CMIP5) GCM ensemble to simulate historical observations of the

bioclimatic variables that are frequently used in biodiversity and ecosystem

impacts studies. We analyse the inter-annual variability of 19 bioclimatic

variables, mean seasonal cycle and ability to reproduce observed seasonal

minima and maxima for 24 GCMs from the CMIP5 ensemble. Our findings show

that for most of the world, temperature variables such as mean annual

temperature have the highest GCM agreement with observations. Lower

agreement is found for temperature and precipitation variables with a seasonal

component, especially in arid and semi-arid locations. The seasonality of

monthly precipitation was found to have low model agreement in Central and

Eastern Europe, East Africa, Southern Australia and parts of Asia. This was found

to be due in part to GCMs not simulating the months of either maximum or

minimum precipitation reliably. These results show that in general the CMIP5

ensemble reliably simulates bioclimatic variables, but care needs to be taken in

its use for certain parts of the world and certain variables. These results will be

made available to conservation practitioners via a protected area information

system. We propose that conservation scientists may reduce uncertainties in

biodiversity projections by selecting a subset of the most reliable models.

Chapter 3: Submitted article

Title: Climate change mitigation policies reduce the rate and magnitude of

ecosystem impacts

Authors: Andrew Hartley, Richard J. J. Gilham, Carlo Buontempo and Richard A.

Betts

Abstract:

Aim

To show the impacts of climate change mitigation on the rate and magnitude of

change in the climate that influences large-scale ecosystems.

Location

Results are calculated for all terrestrial land areas free of ice, and summarized

for 35 places of high conservation priority. We focus on 6 areas of high

conservation priority: Altai-Sayan Montane Forests, Orinoco River and Flooded

Forests, Chihuahuan Deserts, Congo Basin, Southwest Australia, and Coastal

West Africa.

Methods

We use a simple metric of change based on statistical distance within the

Holdridge Life Zone classification space (Hdistance) to quantify ecosystem-

relevant change in climate between a baseline average climate (1961-1990) and

each year in a 150 year time series (1950-2099). We apply this metric to a 58

member ensemble of GCM projections, for a business as usual scenario and an

aggressive climate change mitigation scenario. The rate and magnitude of change

in the Hdistance is calculated for each ensemble member.

Results

We find that more than 50% of high conservation priority areas show divergence

in the rate and magnitude of change in the Hdistance metric when comparing a

business as usual emissions scenario (A1B) with an aggressive carbon dioxide

mitigation scenario (RCP2.6). In other high priority areas we find that potentially

important thresholds are exceeded even with small changes in the Hdistance

under scenario A1B.

Main conclusions

We conclude that potentially dangerous impacts to high priority ecosystems can

be avoided in many parts of the world by a global policy of aggressive climate

change mitigation. Even though in some cases, the long term magnitude of

change threshold is exceeded under RCP2.6, this generally occurs later in the

century, allowing more time for ecosystems to adapt.

Chapter 4

The 5-member RCM has already been run, and some observational datasets

assembled. Problems may be encountered with the observational datasets, due

to the low number of observational stations and issues of data quality.

Chapter 5

I will be second author on a paper addressing this research question. Dr. David

Baker, a post-doctoral researcher at Durham University, has drafted a basic plan

for this paper. My role will be to apply different downscaling methods to the 5-

member GCM ensemble, and compare each method to the dynamical

downscaling approach discussed in chapter 4. Dr. Baker will use the downscaled

climate data to build Bioclimatic Envelope Models for restricted range bird

species in West Africa. We will both analyse the results.

Chapter 6

As mentioned in section 4, the LAM has already been run, and the majority of the

analysis completed. Results currently show that there is an increased likelihood

of convection to initiate on forest-grass boundaries during the afternoon period.

Analysis is still ongoing to establish a link between forest patches and meso-scale

convective systems.

Chapter 7

This chapter has yet to be started, although the same model setup will be used as

in chapter 6.

6 Thesis plan

Each of the following research questions will form a chapter of the thesis. Every

chapter will constitute an individual paper in a peer-reviewed journal, with the

exception of chapter 1.

Introduction and review of relevant literature

The introduction will present issues in global conservation planning, and assess

the approaches for quantifying the impact of climate change on biodiversity. The

second section of this report will form the basis of the introduction to the thesis.

Chapter 1: How are climate change adaptation policies formulated, and what

information is required by policy makers to decide on conservation priorities?

This chapter will identify the main challenges facing policy makers and scientists

in the field of conservation prioritisation with regard to climate change. From

reviewing the academic literature on this subject, it is expected that the

challenges will be related to understanding uncertainty in models, and

implementing and assessing climate change adaptation actions. I will also

directly address the use of climate data in the field, and scope opportunities for

extending the use of climate models.

Chapter 2: How reliable are climate change projections for applications in

biodiversity impacts projections?

This chapter will test the models in the Coupled Model Intercomparison Project

5 (CMIP5) for their ability to simulate observed climatology of variables that are

frequently used in biodiversity impacts studies. The full chapter can be found in

Annex 1.

Chapter 3: What is the impact of climate change mitigation policies on the rate

and magnitude of ecosystem change and how is this affected by climate model

uncertainty?

One of the relevant issues for policy makers is to quantify the benefits of

different climate change mitigation strategies. This paper will compare a

scenario of aggressive climate change mitigation to a business as usual scenario

in the context of impacts on global ecosystems. I will show how the rate of

climate change can be reduced below rates that may be dangerous for the

ecosystems in most locations. The full chapter can be found in Annex 2.

Chapter 4: What is the skill of the HadRM3 regional model for simulating the

climate of West Africa?

Regional climate models provide detailed dynamic projections of how large scale

climatic changes affect local and regional climate. They are an important tool for

use in climate impacts studies, and are an especially important for biodiversity

conservation planning at scales approaching the size of protected areas and

species ranges. Prior to their use in impacts models, the skill of regional models

needs to be assessed in terms of indicators for large scale processes such as

surface pressure seasonality, winds and precipitation. Once we have confidence

that the model can simulate processes such as the monsoon, then we can use

projections with more confidence in impacts studies.

Chapter 5: How does choice of downscaling methodology affect biodiversity

impacts projections in West Africa?

High resolution information on climate is required to make projections of species

future ranges under climate change. Despite several reviews of appropriate

scales in climate impacts studies on biodiversity (Pearson & Dawson 2003;

Wiens & Bachelet 2010), there has been little effort to quantify the effects of

different downscaling techniques on the results of bioclimatic envelope models.

This paper will quantify the effects of three different approaches to downscaling

GCM data for use in a regional climate impacts study: dynamical downscaling

with a regional climate model; empirically based statistical downscaling; and a

change factor approach.

Chapter 6: In a high-resolution limited area model, how does the land surface

interact with the atmosphere during the West African monsoon?

This paper will use a limited area model to explore the relationship between the

land surface and the planetary boundary layer in numerical model with explicit

convection over West Africa. I will examine the model results for evidence of

initiations of convection over forest-grass boundaries, and for evidence of the

influence of forest cover on local and regional rainfall patterns. This will involve

examining local and regional scale precipitation patterns, latent and sensible

heat fluxes before, during and after precipitation, and local winds.

Chapter 7: How do forest related adaptation strategies affect local and regional

precipitation patterns in West Africa?

Following on from chapter 6, I will run the same model for the whole period of

the monsoon (approximately 3 months). Having established a potential

mechanism for interactions between the vegetation cover and the atmosphere, I

will test the affects of different anthropogenic modifications to the land surface

on local and regional scale precipitation. I aim to test the following land cover

scenarios:

- Great Green Wall across the whole of the Sahara. This is a plan that has

been put into place by governments across the whole of the Sahel with the

intention to stop desertification in these countries. The assumption is that

a forest barrier will retain water in local environments, thus benefitting

local communities. With such a large scale project, the potential for

negative affects on precipitation over dry areas needs to be assessed.

- Increased protected area size and connectivity. This is a climate change

adaptation that is often suggested will benefit biodiversity, without

consideration of feedbacks to local and regional climate.

- Increased degradation of forest lands and reduction in forest patches.

While such a scenario may have a detrimental impact on biodiversity, the

increase in number of forest patches may have a positive impact on local

precipitation initiations.

Each of these scenarios will be mapped, and tested in a nested limited area

model. This represents the first study of its kind, and could reveal interesting

insights into climate change adaptation in West Africa.

Chapter 8: Discussion of findings and final conclusions

In this final chapter, I will discuss the implications of this thesis for conservation

and climate change adaptation, with a particular focus on West Africa.

7 Timetable

2013 2014 2015Action J J A S O N D J F M A M J J A S O N D J F M A M J J A S5A 5B 5C 6A 6B 2 3 4A 4B 7A 7B 7C 5A

5B

5C

Code Action5A GCM downscaling methodologies5B Data analysis for chapter 55C Co-write chapter 56A Finish analysis6B Write up results

2 Resubmit chapter 23 Resubmit chapter 3

4A Analysis of RCM skill4B Write-up chapter 47A Setup LAM for land cover runs7B Analyse model results7C Write up chapter 7

8 Bibliography

Beaumont LJ, Pitman A, Perkins S, Zimmermann NE, Yoccoz NG, Thuiller W (2011) Impacts of climate change on the world’s most exceptional ecoregions. Proceedings of the National Academy of Sciences of the United States of America, 108, 2306–11.

Boulangeat I, Philippe P, Abdulhak S, et al. (2012) Improving plant functional groups for dynamic models of biodiversity: at the crossroads between functional and community ecology. Global Change Biology,, n/a–n/a.

Box EO, All AM, Terms J (1981) Predicting physiognomic vegetation types with climate variables. Vegetatio, 45, 127–139.

Brooks TM, Mittermeier R a, Da Fonseca G a B, et al. (2006) Global biodiversity conservation priorities. Science (New York, N.Y.), 313, 58–61.

Bruner AG, Gullison RE, Rice RE, Da Fonseca GAB (2001) Effectiveness of parks in protecting tropical biodiversity. Science (New York, N.Y.), 291, 125–8.

Buckley LB, Urban MC, Angilletta MJ, Crozier LG, Rissler LJ, Sears MW (2010) Can mechanism inform species’ distribution models? Ecology letters, 13, 1041–54.

Buisson L, Thuiller W, Casajus N, Lek S, Grenouillet G (2010) Uncertainty in ensemble forecasting of species distribution. Global Change Biology, 16, 1145–1157.

Clark DB, Mercado LM, Sitch S, et al. (2011) The Joint UK Land Environment Simulator (JULES), model description – Part 2: Carbon fluxes and vegetation dynamics. Geoscientific Model Development, 4, 701–722.

Cramer W, Bondeau A, Woodward FI, et al. (2001) Global response of terrestrial ecosystem structure and function to CO 2 and climate change: results from six dynamic global vegetation models. Global Change Biology, 7, 357–373.

Davis MB, Shaw RG (2001) Range shifts and adaptive responses to Quaternary climate change. Science, 292, 673–9.

Diamond JM (1972) Biogeographic Kinetics : Estimation of Relaxation Times for Avifaunas of Southwest Pacific Islands. Proceedings of the National Academy of Sciences of the United States of America, 69, 3199–3203.

Elith J, H. Graham C, P. Anderson R, et al. (2006) Novel methods improve prediction of species’ distributions from occurrence data. Ecography, 29, 129–151.

Fishpool LDC, Evans M (Eds.) (2000) Important Bird Areas in Africa and Associated Islands: Priority Sites for Conservation. Cambridge, UK., Pisces Publications.

Fishpool LDC, Evans MI (2001) Important bird areas in Africa and associated islands: Priority sites for conservation. Barcelona, Lynx Edicions.

Ford JD, Berrang-Ford L, Paterson J (2011) A systematic review of observed climate change adaptation in developed nations. Climatic Change, 106, 327–336.

Friedlingstein P, Cox P, Betts R, et al. (2006) Climate–Carbon Cycle Feedback Analysis: Results from the C4MIP Model Intercomparison. Journal of Climate, 19, 3337–3353.

Garcia RA, Burgess ND, Cabeza M, Rahbek C, Araújo MB (2011) Exploring consensus in 21st century projections of climatically suitable areas for African vertebrates. Global Change Biology,.

Garcia-Carreras L, Parker DJ, Taylor CM, Reeves CE, Murphy JG (2010) Impact of mesoscale vegetation heterogeneities on the dynamical and thermodynamic properties of the planetary boundary layer. Journal of Geophysical Research, 115, D03102.

Geiger R (1961) Überarbeitete Neuausgabe von Geiger, R. Köppen-Geiger/Klima der Erde.

Gelfand AE, Silander Jr JA, Wu S, Latimer A, Lewis PO, Rebelo AG, Holder M (2006) Explaining Species Distribution Patterns through Hierarchical Modeling. Bayesian Analysis, 1, 41–92.

Giorgi F, Jones C, Asrar GR (2009) Addressing climate information needs at the regional level : the CORDEX framework. 58, 175–183.

Good P, Jones C, Lowe J, Betts R, Booth BBB, Huntingford C (2011) Quantifying Environmental Drivers of Future Tropical Forest Extent. Journal of Climate, 24, 1337–1349.

Grenyer R, Orme CDL, Jackson SF, et al. (2006) Global distribution and conservation of rare and threatened vertebrates. Nature, 444, 93–6.

Grinnell J (1917) The niche-relationships of the Californian Thrasher. Auk, 34, 427–433.

Hampe A (2004) Bioclimate envelope models: what they detect and what they hide. Global Ecology and Biogeography, 13, 469–471.

Hannah L, Dave R, Lowry PP, et al. (2008) Climate change adaptation for conservation in Madagascar. Biology letters, 4, 590–4.

Hannah L, Midgley G, Andelman S, et al. (2007) Protected area needs in a changing climate. Frontiers in Ecology and the Environment, 5, 131–138.

Hannah L, Midgley GF, Hughes G, Bomhard B (2005) The View from the Cape: Extinction Risk, Protected Areas, and Climate Change. BioScience, 55, 231.

Hannah L, Midgley GF, Millar D (2002) Climate change-integrated conservation strategies. Global Ecology and Biogeography, 11, 485–495.

Harrison SP, Prentice IC, Barboni D, Kohfeld KE, Ni J, Sutra J-P (2010) Ecophysiological and bioclimatic foundations for a global plant functional classification. Journal of Vegetation Science, 21, 300–317.

Hartley A, Nelson A, Mayaux P, Grégoire J-M (2007) The Assessment of African Protected Areas. Luxembourg, Office for Official Publications of the European Communities.

Heikkinen RK, Luoto M, Araújo MB, Virkkala R, Thuiller W, Sykes MT (2006) Methods and uncertainties in bioclimatic envelope modelling under climate change. Progress in Physical Geography, 30, 751–777.

Heller N, Zavaleta E (2009) Biodiversity management in the face of climate change: A review of 22 years of recommendations. Biological Conservation, 142, 14–32.

Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology, 25, 1965–1978.

Hijmans RJ, Graham CH (2006) The ability of climate envelope models to predict the effect of climate change on species distributions. Global Change Biology, 12, 2272–2281.

Holdridge LR (1967) Life Zone Ecology. San Jose, Costa Rica, Tropical Science Center.

Hole DG, Huntley B, Arinaitwe J, et al. (2011) Toward a management framework for networks of protected areas in the face of climate change. Conservation biology : the journal of the Society for Conservation Biology, 25, 305–15.

Hole DG, Willis SG, Pain DJ, et al. (2009) Projected impacts of climate change on a continent-wide protected area network. Ecology letters, 12, 420–31.

Von Humboldt A (1867) Ideeen zu einemGeographie der Pflan- zen nebst einem naturgemälde der Tropenländer. F. G. Cotta, Tübingen. Tübingen, F. G. Cotta.

Huntley B, Barnard P, Altwegg R, et al. (2010) Beyond bioclimatic envelopes: dynamic species’ range and abundance modelling in the context of climatic change. Ecography,, 621–626.

Kearney M, Porter W (2009) Mechanistic niche modelling: combining physiological and spatial data to predict species’ ranges. Ecology letters, 12, 334–50.

Kearney MR, Wintle B a., Porter WP (2010) Correlative and mechanistic models of species distribution provide congruent forecasts under climate change. Conservation Letters, 3, 203–213.

Keith D a, Akçakaya HR, Thuiller W, et al. (2008) Predicting extinction risks under climate change: coupling stochastic population models with dynamic bioclimatic habitat models. Biology letters, 4, 560–3.

Koeppen W (1900) Versuch einer Klassifikation der Klimate, vorzugsweise nach ihren Beziehungen zur Pflanzenwelt. Geographische Zeitschrift,, 593–611; 657–679.

Kottek M, Grieser J, Beck C, Rudolf B, Rubel F (2006) World Map of the Köppen-Geiger climate classification updated. Meteorologische Zeitschrift, 15, 259–263.

Lafore J-P, Flamant C, Guichard F, et al. (2011) Progress in understanding of weather systems in West Africa. Atmospheric Science Letters, 12, 7–12.

Lewis OT (2006) Climate change, species-area curves and the extinction crisis. Philosophical Transactions of the Royal Society of London - Series B: Biological Sciences, 361, 163–171.

Lugo a. E, Brown SL, Dodson R, Smith TS, Shugart HH (1999) The Holdridge life zones of the conterminous United States in relation to ecosystem mapping. Journal of Biogeography, 26, 1025–1038.

MacArthur RH, Wilson EO (1967) The Theory of Island Biogeography. Princeton, New Jersey.

Maclean IMD, Wilson RJ (2011) Recent ecological responses to climate change support predictions of high extinction risk. Proceedings of the National Academy of Sciences,, 1017352108–.

Mahlstein I, Daniel JS, Solomon S (2013) Pace of shifts in climate regions increases with global temperature. Nature Climate Change, 3, 1–5.

Margules CR, Pressey RL (2000) Systematic conservation planning. Nature, 405, 243–53.

Maschinski J, Baggs JE, Quintana-Ascencio PF, Menges ES (2006) Using Population Viability Analysis to Predict the Effects of Climate Change on the Extinction Risk of an Endangered Limestone Endemic Shrub, Arizona Cliffrose. Conservation Biology, 20, 218–228.

Mawdsley JR, O’Malley R, Ojima DS (2009) A review of climate-change adaptation strategies for wildlife management and biodiversity conservation. Conservation biology : the journal of the Society for Conservation Biology, 23, 1080–9.

Menzel A, Sparks TH, Estrella N, et al. (2006) European phenological response to climate change matches the warming pattern. Global Change Biology, 12, 1969–1976.

Metzger MJ, Bunce RGH, Jongman RHG, Sayre R, Trabucco A, Zomer R (2013) A high-resolution bioclimate map of the world: a unifying framework for global biodiversity research and monitoring (M Sykes, Ed.). Global Ecology and Biogeography, 22, 630–638.

Midgley GF, Hannah L, Millar D, Rutherford MC, Powrie LW (2002) Assessing the vulnerability of species richness to anthropogenic climate change in a biodiversity hotspot. Global Ecology and Biogeography, 11, 445–451.

Mora C, Tittensor DP, Adl S, Simpson AGB, Worm B (2011) How many species are there on Earth and in the ocean? (GM Mace, Ed.). PLoS biology, 9, e1001127.

Myers N, Mittermeier R a, Mittermeier CG, Da Fonseca G a, Kent J (2000) Biodiversity hotspots for conservation priorities. Nature, 403, 853–8.

Nakicenovic N, Alcamo J, David G, et al. (2000) IPCC Special Report on Emissions Scenarios. Cambridge, UK and New York, NY.

Olson DM, Dinerstein E (1998) The Global 200 : A Representation Approach to Conserving the Earth’s Most Biologically Valuable Ecoregions. Issues in International Conservation, 12, 502–515.

Olson DM, Dinerstein E (2002) The Global 200: Priority Ecoregions for Global Conservation. Annals of the Missouri Botanical Garden, 89, 199.

Parmesan C (2006) Ecological and Evolutionary Responses to Recent Climate Change. Annual Review of Ecology, Evolution, and Systematics, 37, 637–669.

Pearson RG, Dawson TP (2003) Predicting the impacts of climate change on the distribution of species: are bioclimate envelope models useful? Global Ecology and Biogeography, 12, 361–371.

Perrings C, Duraiappah A, Larigauderie A, Mooney H (2011) Ecology. The biodiversity and ecosystem services science-policy interface. Science (New York, N.Y.), 331, 1139–40.

Phillips SJ, Dudík M (2008) Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography, 31, 161–175.

Porter WP, Gates DM (1969) Thermodynamic Equilibria of Animals with Environment. Ecological Monographs, 39, 227–244.

Prentice I, Cramer W, Harrison S, Leemans R, Monserud R, Solomon A (1992) A global biome model based on plant physiology and dominance, soil properties and climate. Journal Of Biogeography, 19, 117 – 134.

Sitch S, Huntingford C, Gedney N, et al. (2008) Evaluation of the terrestrial carbon cycle, future plant geography and climate-carbon cycle feedbacks using five Dynamic Global Vegetation Models (DGVMs). Global Change Biology, 14, 2015–2039.

Sitch S, Smith B, Prentice IC, et al. (2003) Evaluation of ecosystem dynamics, plant geography and terrestrial carbon cycling in the LPJ dynamic global vegetation model. Global Change Biology, 9, 161–185.

Taylor CM, Gounou A, Guichard F, Harris PP, Ellis RJ, Couvreux F, De Kauwe M (2011) Frequency of Sahelian storm initiation enhanced over mesoscale soil-moisture patterns. Nature Geoscience, 4, 430–433.

Taylor CM, De Jeu R a M, Guichard F, Harris PP, Dorigo W a (2012) Afternoon rain more likely over drier soils. Nature, 489, 423–6.

Taylor CM, Parker DJ, Harris PP (2007) An observational case study of mesoscale atmospheric circulations induced by soil moisture. Geophysical Research Letters, 34, L15801.

Thomas CD, Cameron A, Green RE, et al. (2004) Extinction risk from climate change. Nature, 427, 145–8.

Thuiller W (2004) Patterns and uncertainties of species’ range shifts under climate change. Global Change Biology, 10, 2020–2027.

Thuiller W, Araújo MB, Pearson RG, Whittaker RJ, Brotons L, Lavorel S (2004) Biodiversity conservation: Uncertainty in predictions of extinction risk. Nature, 430, 2004.

Velarde SJ, Malhi Y, Moran D, Wright J, Hussain S (2005) Valuing the impacts of climate change on protected areas in Africa. Ecological Economics, 53, 21–33.

Veloz SD, Williams JW, Blois JL, He F, Otto-Bliesner B, Liu Z (2012) No-analog climates and shifting realized niches during the late quaternary: implications for 21st-century predictions by species distribution models. Global Change Biology, 18, 1698–1713.

Vuuren DP, Edmonds J, Kainuma M, et al. (2011) The representative concentration pathways: an overview. Climatic Change, 109, 5–31.

Wiens JA, Bachelet D (2010) Matching the multiple scales of conservation with the multiple scales of climate change. Conservation Biology, 24, 51–62.

Wiens JA, Stralberg D, Jongsomjit D, Howell CA, Snyder MA (2009) Niches, models, and climate change: assessing the assumptions and uncertainties. Proceedings of the National Academy of Sciences of the United States of America, 106, 19729–19736.

Wilson EO, Peter FM (1988) Biodiversity (EO Wilson and FM Peter, Eds.). Washington D.C., National Academy Press.

Annex 1

Please see the document entitled “HARTLEY_upgardereport_Annex1.docx”

Annex 2

Please see the document entitled “HARTLEY_upgardereport_Annex2.docx”