Pacific Science, vol. 68, no. 1 July 16, 2013 (Early view) Climate change and weed impacts on small island ecosystems: Lantana camara L. (Magnoliopsida: Verbenaceae) distribution in Fiji By: Subhashni Taylor* and Lalit Kumar Abstract Oceanic island ecosystems are particularly vulnerable to invasion by exotic species and this vulnerability may be intensified by the effects of climate change, particularly if favourable climate conditions allow exotic invaders to spread to new areas. Effective management of such species requires knowledge of their potential distribution under current and future climate. This research examines the value of species distribution modelling in invasive species management in small island ecosystems using the specific case of Lantana camara L. invasion in Fiji. A niche model of L. camara was developed using CLIMEX species distribution modelling (SDM) software. Several sources of data were used to develop the model including phenological observations and geographic distribution records. The model was used to estimate its potential distribution under historical climate. The CSIRO-Mk3.0 Global Climate Model (GCM) was used to explore the impacts of climate change on its potential future distribution. The model was run with the A1B and A2 scenarios for 2030 and 2070. Large parts of both the major islands of Fiji, Viti Levu and Vanua Levu, were shown to have high climatic suitability for L. camara. However, under future climate scenarios, the climatic suitability for L. camara in Fiji was projected to contract. The results illustrate that SDMs can play a useful role in the formulation of cost-effective invasive species management strategies and the resulting species distribution maps have broad applicability throughout the many islands of the South Pacific region. *Corresponding Author E-mail: [email protected]
28
Embed
Climate change and weed impacts on small island - Pacific Science
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
island ecosystems: Lantana camara L. (Magnoliopsida: Verbenaceae) distribution in
Fiji
By: Subhashni Taylor* and Lalit Kumar Abstract Oceanic island ecosystems are particularly vulnerable to invasion by exotic species and this vulnerability may be intensified by the effects of climate change, particularly if favourable climate conditions allow exotic invaders to spread to new areas. Effective management of such species requires knowledge of their potential distribution under current and future climate. This research examines the value of species distribution modelling in invasive species management in small island ecosystems using the specific case of Lantana camara L. invasion in Fiji. A niche model of L. camara was developed using CLIMEX species distribution modelling (SDM) software. Several sources of data were used to develop the model including phenological observations and geographic distribution records. The model was used to estimate its potential distribution under historical climate. The CSIRO-Mk3.0 Global Climate Model (GCM) was used to explore the impacts of climate change on its potential future distribution. The model was run with the A1B and A2 scenarios for 2030 and 2070. Large parts of both the major islands of Fiji, Viti Levu and Vanua Levu, were shown to have high climatic suitability for L. camara. However, under future climate scenarios, the climatic suitability for L. camara in Fiji was projected to contract. The results illustrate that SDMs can play a useful role in the formulation of cost-effective invasive species management strategies and the resulting species distribution maps have broad applicability throughout the many islands of the South Pacific region. *Corresponding Author E-mail: [email protected]
2
Introduction
The Pacific Islands region has been designated as one of the 25 biodiversity "hot spots" for
conservation priority in the world, and the flora of the Pacific Islands has been estimated to
comprise over 3300 endemic plant species that are unique to the region (Myers et al. 2000). The
combined threats of climate change and invasive species puts them at risk of extinction (Meyer
2000, Myers et al. 2000). Invasive species have many negative impacts on the native flora and
communities of islands and the added impacts of climate change can further exacerbate such
threats, particularly if favourable climate conditions allow invasives to expand their distribution.
Therefore, it is important to protect and maintain this rich but fragile natural heritage from the
impacts of invasive species.
A review of invasive species in the Pacific region (Sherley et al. 2000) has outlined a
number of issues that are barriers to effective management of such species in the region. Some
problems that are highlighted in the report include: (1) shortage and inaccessibility of scientific
information on basic biology, including distribution, for risk assessment and effective
management; (2) inadequate quarantine and risk assessment facilities; (3) lack of awareness of
the impacts of invasive species on biodiversity; and (4) threats to biodiversity are not yet well
understood.
Biosecurity agencies require information on the potential distribution and relative
abundance of invasive species under current and future climate scenarios if they are to carry out
risk assessments and formulate effective long term management strategies (Kriticos et al. 2011a).
Species distribution models (SDMs), bioclimatic models, and ecological niche models (ENMs)
are valuable tools that can be used in such instances (Guisan and Zimmerman 2000). A species’
preferred climate is inferred from its distribution data, commonly termed the ‘environmental
envelope approach’ (Barry and Elith 2006). The major assumption behind such models is that
3
climate is the primary factor defining the potential range of plants and other poikilotherms
(Woodward 1987). A range of software has been developed which can be used to model species’
current and future distributions (Peterson et al. 2011). CLIMEX (Hearne Scientific Software
2007, Sutherst et al. 2007) is one such software that has been frequently used in invasive species
distribution modelling (Poutsma et al. 2008, Sutherst and Bourne 2009, Chejara et al. 2010,
Kriticos et al. 2011a). A range of information types, such as experimental observations of a
species’ growth response to temperature and soil moisture, current distribution and seasonal
phenology data, can be used within CLIMEX to model potential distributions of species. In a
comparison of CLIMEX with two correlative modelling methods (MaxEnt and Boosted
Regression Trees), Webber et al. (2011) found that CLIMEX was better placed to project a
species’ distribution in a novel climate such as a new continent, or under a future climate
scenario. Furthermore, a review of the various climate-based packages intended for potential
species distribution assessment found that ‘CLIMEX was the most suitable climate modelling
package for undertaking Weed Risk Assessments because it can support model-fitting to a global
plant distribution, includes a climate change scenario mechanism, and provides an insight into
the plant’s ecological response to climate’ (Kriticos and Randall 2001).
In this paper, we examine the utility of bioclimatic models in invasive species
management in small island ecosystems. Limited research has been carried out in this respect,
particularly in the context of the islands of the South Pacific (however, see Christenhusz and
Toivonen 2008, Rödder and Lötters 2010). We ask the following question: Can species
distribution modelling be used to deal with some of the issues that are seen as barriers to
effective management of invasive species in the South Pacific, specifically the issues related to
lack of information on invasive species distribution?
4
We answer this question by developing a model of the climate responses of the highly
noxious weed, Lantana camara L. based on its native and invasive distribution using the
CLIMEX software package. We then use this model to examine its potential current distribution
and future distribution under climate change on the island of Fiji in the South Pacific.
Materials and Methods
Study Area
The Fijian Islands archipelago is located in the Pacific Ocean between longitudes 174º E and
178º W and latitudes 12º to 22º S. It comprises approximately 330 islands of which one third are
inhabited. The archipelago covers about 1.3 million km2 of the South Pacific Ocean with the
total land area being 18,333 km2. Viti Levu and Vanua Levu are the two main islands, making up
15,985 km2 of the land area (Figure 1). The total population is just over 800,000 with most
people living on the two main islands (Fiji Islands Bureau of Statistics 2011).
Figure 1 near here
Lantana camara L.
Lantana camara L. (hereafter called Lantana) has been ranked among one of the world’s 100
worst invasive alien species (Lowe et al. 2000). Its native range is Central and northern South
America and the Caribbean (Day et al. 2003). However, it is now a major weed in many tropical
and subtropical countries with its global distribution encompassing approximately 60 countries
or island groups between 35º N and 35º S (Day et al. 2003). Lantana was introduced to the
Pacific islands after European contact and since then has rapidly naturalized and become a
noxious weed in this region (Figure 1). It has been reported on many islands such as Hawaii,
Cook Islands, Fiji, Samoa, Vanuatu, New Caledonia and Solomon Islands (Thaman 1974,
Waterhouse 1997, Day et al. 2003, Global Invasive Species Database 2012) and is considered
5
among the most significant invasive taxa found on the majority of the islands in the region
(Meyer 2000). Lantana forms dense stands that exclude native seedlings by inhibiting their
germination, seedling growth and survival through competition (Sharma et al. 2005).
Furthermore, a single plant can produce up to 12 000 seeds per year which are dispersed over
long distances by birds and mammals (Stock et al. 2009). The ability to spread through layering
also makes it a successful invader (Day et al. 2003, Stock et al. 2009). These characteristics,
combined with a favourable climate, an abundance of open habitats and a relative absence of
natural competition, have allowed Lantana to become naturalized in many of the Pacific island
groups (Thaman 1974). It can potentially cause a reduction in biodiversity and increased
susceptibility to fire intrusion (Berry et al. 2011). Other potential impacts include decline in soil
fertility and an alteration of ecosystem processes (Day et al. 2003). Lantana has been classified
as a “dominant invader” and one of the most important threats to biodiversity in many Pacific
island nations due to its widespread occurrence and its ability to form dense stands thereby
severely impacting native biota (Meyer 2000).
Distribution of Lantana in Fiji
The distribution of Lantana in Fiji was ascertained from a variety of sources, including Smith
(1991), Day and Neser (2000), A. Macanawai (Principal Research Officer, Plant Protection
Section, Department of Agriculture, Koronivia Research Station, Nausori, Fiji, personal
communication) and A. Tye (Former Invasive Species Adviser, Secretariat of the Pacific
Regional Environment Programme). In Fiji, Lantana has been reported along road sides in most
provinces of Viti Levu and in some coconut plantations on Vanua Levu and Taveuni (A.
Macanawai, personal communication). It has also been reported as a garden escapee and being
widely distributed in Fiji in cultivated areas, particularly on the drier sides of the large islands
(A. Tye, personal communication) as well as in the cooler and wetter eastern and mountainous
6
regions (Day and Neser 2000). On Viti Levu, Lantana was reported in Lautoka, Nadarivatu,
Yaqara, Navua, Colo-i-Suva and Lami while in Vanua Levu it was found in Labasa and
Savusavu as well as on Taveuni Island (Smith 1991). Smith’s (1991) observations were made
over 20 years ago and therefore will need to be interpreted with caution although the personal
communication data is more recent.
CLIMEX Software
A model of the potential distribution of Lantana under current and future climate scenarios was
developed using CLIMEX software (Hearne Scientific Software 2007, Sutherst et al. 2007). An
eco-physiological growth model forms the basis of this software where it is assumed that a
population experiences a favourable season with positive growth and an unfavourable season
with negative population growth. Parameters that describe a species’ response to climate can be
inferred from its geographic range and phenological observations (Sutherst et al. 2007). The user
can then apply these parameters to novel climates to project the potential range in new regions or
climate scenarios (Kriticos et al. 2011a). An annual growth index (GIA) describes the potential
for population growth during favourable climate conditions. The GIA is determined from the
temperature index (TI) and moisture index (MI) which represent the species’ temperature and
moisture requirements for growth. The probability that the population can survive unfavourable
conditions is described by four stress indices (cold, wet, hot and dry) and up to four interaction
stresses (hot–dry, hot–wet, cold–dry and cold–wet). The growth and stress indices are calculated
weekly and combined into an overall annual index of climatic suitability, the Ecoclimatic index
(EI) which is theoretically scaled from 0 to 100. An EI value of zero indicates that the species
will not be able to survive at that location, 1–10 indicate marginal habitats, 10–20 can support
substantial populations while EI values >20 are highly favourable (Sutherst et al. 2007). For a
detailed description of parameters, see Sutherst and Maywald (1985). In this study, parameters
7
describing Lantana’s climatic preferences were fitted using the methodology described in
Sutherst and Maywald (1985), Chejara et al (2010) and Kriticos et al (2011a).
Meteorological Data and Climate Change Scenarios
The modelling component of this study employed the CliMond 10´ gridded climate data
(Kriticos et al. 2011b). Five variables, average minimum monthly temperature (Tmin), average
maximum monthly temperature (Tmax), average monthly precipitation (Ptotal) and relative
humidity at 09:00 h (RH09:00) and 15:00 h (RH15:00), were used to represent historical climate
(averaging period 1950–2000). The same five variables were used to depict potential future
climate in 2030 and 2070, based on the CSIRO-Mk3.0 Global Climate Model (GCM) (Gordon et
al. 2002) with the A1B and A2 SRES scenarios (IPCC 2000). The A1B scenario describes a
world with a balanced use of fossil and non-fossil resources while A2 describes a varied world
with high population growth but slow economic development and technological change. The B
family of scenarios were not assessed here because of the findings of Rahmstorf et al (2007) that
recent global temperature increases were much higher than the hottest IPCC scenarios. The
CSIRO-Mk3.0 GCM was selected for this study because it performed well compared to other
GCMs at representing basic aspects of the observed climate at a regional scale (Hennessy and
Colman 2007, Kriticos et al. 2011b).
Fitting CLIMEX Parameters
Information on the global distribution of Lantana was downloaded from the Global Biodiversity
Information Facility (GBIF). The GBIF is a database of natural history collections around the
world for various species which is available for download. A total of 4126 records were
downloaded of which only 1740 records were used in parameter fitting because many records did
not have geolocations or were repetitions and were subsequently discarded. Further data on
Lantana’s exotic distribution in South Africa (SAPIA 2006) and Asia (Biswas 1934, Jafri 1974,
8
Thakur 1992, Chen and Gilbert 1994, Press et al. 2000) were used for fitting stress parameters.
Both native and exotic distribution data were used for model parameterization to ensure that the
complete range of environmental conditions in which lantana may occur was covered. Growth
parameters were fitted using seasonal phenology data from the southern states of Brazil (Winder,
1980, 1982). Although Winder’s study was restricted to Lantana tiliaefolia and L. glutinosa,
these data were used in parameter fitting based on the ecological similarity of these two species
to the weedy taxa of Lantana. Each parameter was iteratively adjusted until a reasonable match
was seen between the potential and known distribution of Lantana in these areas. The intention
was to get maximum EI values near known vigorous populations and to minimize EI values
outside the recorded distribution of Lantana. The parameters were checked to ensure that they
were biologically reasonable (Table 1). For a detailed explanation of the parameter-fitting
procedure and model validation, refer to Taylor et al. (2012) and Taylor and Kumar (2012). The
fitted parameters were then used to project lantana’s potential distribution in Fiji under historical
climate (averaging period 1950–2000) and climate change scenarios.
Results
Historical Climate
Figure 2 shows the projected distribution in the Pacific region for the islands that could be shown
at this scale. These were Vanuatu, New Caledonia and Solomon Islands. The presence data
shown in Figure 1 accords well with the projected distribution on these three islands in Figure 2.
The observed distribution of Lantana in Fiji is shown in Figure 3 while the potential distribution
under historical climate is shown in Figure 4. The current distribution shows a good match with
the modelled distribution. Most of Vanua Levu,Viti Levu and Taveuni are shown to be highly
suitable or suitable for Lantana.
9
Future Climate
Potential distribution under future climate scenarios using the CSIRO-Mk3.0 GCM are shown in
Figures 5 and 6. A contraction in suitable climate areas was seen for this climate change model,
particularly throughout the island of Vanua Levu but also on the eastern half of Viti Levu
(Figures 5 and 6) with this trend even clearer in the 2070 scenario. The two emission scenarios
showed very similar results for both time periods.
The changes in wet stress between current climate and 2070 was also mapped (Figure 7)
as this was the only stress that appeared to be impacted by climate change. Increasing levels of
wet stress is seen on the eastern side of Viti Levu and throughout Vanua Levu.
Table 1 near here
Figures 2, 3, 4, 5, 6 and 7 near here
Discussion
The utility of species distribution modelling was examined as a tool for the purposes of
informing the formulation of long term management strategies for invasive species on small
island ecosystems. This was done using the specific case of Lantana invasion in Fiji. The results
show that under current climate, large parts of Fiji are conducive to Lantana growth and spread
(Figure 4) and this accords well with the observed distribution (Figure 3). The likelihood of
further spread of this highly noxious weed from infested to uninfested areas is possible given the
rapid increase in travel and transportation. Furthermore, habitat disturbance caused by activities
such as logging, land clearing for agriculture and introduction of large mammals such as cattle
and sheep provide further suitable habitats for Lantana invasion (Thaman 1974, Day et al. 2003).
In particular, containment efforts around the three southern provinces on Viti Levu where
10
Lantana currently is not known to occur (Figure 3) and where climatic suitability is shown to be
marginal (Figure 4) may be effective in preventing spread to these uninvaded regions.
An important aspect of invasive species management is to educate the citizens and to
raise public awareness of the impacts of such species, not only on agriculture but also on native
biodiversity. The maps of potential Lantana distribution produced in this study can be a useful
tool in public awareness campaigns. Furthermore, maps of potential distribution resulting from
such modeling exercises may also be valuable for other aggressive invaders that have been
identified in many South Pacific countries such as African tulip tree (Spathodea campanulata),