Transcript Molecular Evolution

Chapter 11
Author: Lee Hannah
FIGURE 11.1 Schematic of an SDM. Species distribution modeling begins with selection of a study
area (left). The study area is usually selected to be large enough to include the complete ranges of
species of interest to ensure that data sampling the entire climate space the species can tolerate are
included. Climate variables and other factors constraining species distribution (shaded layers on right)
are then correlated with known occurrences of the species of interest (layer with points). This statistical
relationship can be projected geographically to simulate the species ’ range (bottom shaded area).
Repeating this process using GCM-generated future climate variables allows simulation of range shifts
in response to climate change. Copyright 1998, Massachusetts Institute of Technology, by permission
of MIT Press.
FIGURE 11.2 Global and Regional Vegetation Simulation of a DGVM.
The global distribution of PFTs (top) can be simulated in a coarse-scale
DGVM. The same DGVM run at finer resolution can simulate PFT distribution
with many local features resolved (bottom left). Driving the DGVM with
projected future climates from a GCM provides simulation of change in PFT
distribution due to climate change at either global or regional (bottom right)
scales. From Ronald P. Neilson, USDA Forest Service.
FIGURE 11.3 Gap Model Output. This gap model of forest
composition in Switzerland under climate change shows an early
peak in oak abundance, giving way to a mixed fir – beech forest with
little oak. Copyright 1998, Massachusetts Institute of Technology, by
permission of MIT Press.
FIGURE 11.4 DGVM Intercomparison. Outputs of six different first-generation
DGVMs (top six panels) compared to the composite of all six (bottom left) and
PFT distribution classified from satellite imagery (bottom right). From Cramer,
W., et al. 2001. Global response of terrestrial ecosystem structure and function
to CO2 and climate change: Results from six dynamic global vegetation models.
Global Change Biology 7, 357 – 373.
FIGURE 11.5 Climate and CO 2 DGVM Simulations. DGVM global
change simulations for high emissions (A1) and high climate
sensitivity GCM (Hadley) (A) and low emissions (B2) and low
climate sensitivity GCM (ECHAM) (B). Source: IPCC.
FIGURE 11.6 Generation of a Future Climatology for Species Distribution Modeling.
Because SDMs often require climatologies with horizontal resolutions much finer than
those offered by GCMs, techniques are needed for generating downscaled future
climatologies from GCM outputs. One approach commonly used applies the difference
between GCM simulations of the present and the future to a current fine-scale
climatology. This is done because GCM fidelity to current climate may be imperfect. Use
of a historical fine-scale climatology ensures reasonable reproduction of major climatic
features. The GCM difference (future-present) simulates future warming. Courtesy of
Karoleen Decatro, Ocean o’Graphics.
FIGURE 11.7 SDM Result Correlation to Climate Variables. The cattergrams
indicate relationships between species loss (%) and anomalies of moisture
availability and growing-degree days in Europe. The colors correspond to the
climate change scenarios indicated in the legend. From Thuiller, W., et al.
2005. Copyright National Academy of Sciences.
FIGURE 11.8 Example of SDM Output. SDM output for a protea (pictured) from
the Cape Floristic Region of South Africa. Current modeled range is shown in
red, and future modeled range is shown in blue. Known occurrence points for
the species are indicated by black circles. Figure courtesy Guy Midgley.
FIGURE 11.9 Backwards and Forwards Modeling of Eastern Mole ( Scalopus
aquaticus ). (A) SDM created from known Pleistocene occurrences predicts
present distribution. (B) SDM created from known current distribution predicts
known fossil occurrences. From Martinez-Meyer, E., et al. 2004. Ecological
niches as stable distributional constraints on mammal species, with implications
for Pleistocene extinctions and climate change projections for biodiversity. Global
Ecology and Biogeography 13, 305 – 314.
FIGURE 11.10 Model of Suitable Climate for Glassy-winged Sharpshooter in
the United States. The potential spread of insect pests such as the glassywinged sharpshooter can be predicted using SDMs. From Venette, R. C. and
Cohen, S. D. 2006. Potential climatic suitability for establishment of
Phytophthora Ramorum within the contiguous United States. Forest Ecology
and Management 231, 18 – 26.
FIGURE 11.11 SDM Modeled Changes in European Plant Composition. Spatial sensitivity of
plant diversity in Europe ranked by biogeographic regions, based on results from
multispecies SDM. Mean percentage of current species richness (a) and species loss (b)
and turnover (c) under the A1-HadCM3 scenario. The results are aggregated by
environmental zones. From Thuiller, W., et al. 2005. Copyright National Academy of
Sciences.
FIGURE 11.12 Habitat Suitability for Australian Rain Forests. Suitability for
Australian rain forests is shown for three past climates and the present. The
modeling tool used is Bioclim, an early SDM software. Like other SDMs, Bioclim
can be used to simulate suitability for biomes or vegetation types as well as
species. End panels indicate genetic similarity of forest fragments (right) and
areas that are stable in all modeled time slices (left). From Hugall, A., et al.
Reconciling paleodistribution models and comparative phylogeography in the wet
tropics rainforest land snail Gnarosophia Bellendenkerensis (Brazier 1875).
Proceedings of the National Academy of Sciences of the United States of
America 99, 6112 – 6117.
FIGURE 11.13 SDM Time-slice Analysis. SDM simulation of present and
future range for the pygmy spotted skunk of western Mexico. This model was
constructed in 10-year time slices to allow pathways of contiguous habitat
(present to future) to be identifi ed. Reproduced with permission from the
Ecological Society of America.
FIGURE 11.14 Lake Habitat Suitability Profiles under Double-CO 2 Scenarios.
With warming, lake habitat suitability is very close with depth and over time. Simulated
habitat suitability is shown for double-CO 2 scenarios for lakes in Duluth, Minnesota, and
Austin, Texas. In Duluth, uninhabitable surface water extends deeper and lasts longer in the
climate change scenario. In Austin, a summer window of habitability in the mid-depths
closes, making the entire lake uninhabitable by late summer. From Stefan, H. G., et al.
2001. Simulated fish habitat changes in North American lakes in response to projected
climate warming. Transactions of the American Fisheries Society 130, 459 – 477.
FIGURE 11.15 Maximum Monthly Sea Surface Temperatures.
(A) Observed, (B) 2000 – 2009 projected, (C) 2020 – 2029 projected, (D)
2040 – 2049 projected, and (E) 2060 – 2069 projected. Warmer temperatures
cause bleaching that threatens persistence of coral reefs. From Guinotte, J.
M., et al. 2003. With kind permission from Springer Business Media.
FIGURE 11.16 Aragonite Saturation State of Seawater. Red/yellow, low-marginal;
green, adequate/optimal. (A) Preindustrial (1870), (B) 2000 – 2009 projected,
(C) 2020 – 2029 projected, (D) 2040 – 2049 projected, and (E) 2060 – 2069 projected.
Low saturation states unsuitable for coral reefs collapse in toward the equator as the
century progresses. From Guinotte, J. M., et al. 2003. With kind permission from
Springer Business Media.
FIGURE 11.17 Areas in Which Temperature and Aragonite Saturation State
Combine to Stress Corals. (A) Observed, (B) 2000 – 2009 projected, (C)
2020 – 2029 projected, (D) 2040 – 2049 projected, and (E) 2060 – 2069
projected. From Guinotte, J. M., et al. 2003. With kind permission from
Springer Business Media.
FIGURE 11.18 Output of a Marine Ecosystem Model. Simulated relative percentage change
in phytoplankton production is shown for the region surrounding Australia, based on a
nutrient – phytoplankton – zooplankton model. Production generally increases with
warmer water temperatures, although note the exception in a region off southeastern
Australia. The change is calculated as the percentage difference between the 2000 – 2004
mean and the 2050 mean. From Brown, C. J., et al. 2010. Effects of climate-driven primary
production change on marine food webs: Implications for fisheries and conservation. Global
Change Biology 16, 1194 – 1212.
FIGURE 11.19 EcoSim Food Web Model Results. Based on the production changes
shown in Figure 11.18 , the EcoSim food web model simulates change in abundance of
species of conservation interest. Following increases in production, biomass abundance
increases for most species. However, abundance of some species may decline due to
trophic effects (changed food web relationships). From Brown, C. J., et al. 2010. Effects
of climate-driven primary production change on marine food webs: Implications for
fisheries and conservation. Global Change Biology 16, 1194 – 1212.