Princess Gilbert 803115990 UP206a- Final Write Up A Spatial Analysis of Factors Impacting Chameleon Diversity in Madagascar Introduction The Central Problem “The central problem of speciation is understanding the origin of those isolating barriers that actually prevent or potentially prevent gene flow in sympatry. This problem is conceptually simple, but its solution involves two formidable tasks. First, one must determine which reproductive barriers were involved in the initial reduction of gene flow between populations. Second, one must understand which evolutionary forces produced these barriers…” (Coyne and Orr 2004). If the above is true for the study of speciation which involves two or more species, such a methodology can be adapted and applied to the study of diversification at the intraspecific level. This dissertation proposal seeks to do just that by: (1) identifying the initial reduction of gene flow within one species; (2) determining the barriers which are involved in this reduction; and (3) elucidating how selection acts upon the genome of that species and partially reduces gene flow. Why Madagascar? Madagascar presents an ideal natural setting for the study of evolutionary processes. Isolated from any other land mass for roughly 88 million years, the flora and fauna of Madagascar have evolved in complete isolation. Such insular isolation allows one to compare a simpler microcosm with the seemingly infinite complexity of continental and oceanic biology (MacArthur and Wilson 1967). Madagascar is exceedingly rich in endemic species. Ninety-five percent of the reptile species, 99% of the amphibian species, and 100% of the primate species are endemic (Yoder and Nowak 2006). This increases the ease with which one can study diversification. Madagascar is also diverse in habitats and climate (see figure 1). Madagascar contains 6 unique biomes which evolved under differing paleogeographic regimes. Following its o separation from Gondwana, India-Madagascar moved into the subtropical arid belt 30 below the equator. This resulted in the formation of the arid spiny bush and dry deciduous forest biomes 88 Mya (Wells 2003). 66-58 Mya, during the late Paleocene/early Eocene, the combination of Madagascar’s movement northward into the equatorial tropical belt and the onset of southeasterly trade winds from the India Ocean, created the eastern evergreen rainforest via orographic precipitation. The northwestern Sambirano rainforest formed most recently (8Mya) following the initiation of the Indian monsoons. The formation of montane ericoid thicket and central highlands are yet to be resolved. Why Chameleons? Within the unique island of Madagascar, chameleons present a wonderful taxa for the study of diversification. Three genera within the Old World family Chamaeleonidae are endemic to Madagascar: Brookesia, Calumma and Furcifer (Klaver and Bohme 1986). The Furcifer genus are inhabitants of arid regions in the west, however, two species (lateralis and oustaleti) are found in western evergreen rainforests (Glaw and Vences 1994). The characteristically slow movement of chameleons also makes for easier sample collection. Furcifer oustaleti is an extremely widespread species. It specializes in dry wooded habitats and, with the exception of lateralis, is the only chameleon adapted to the dry environments of the south as well as the highlands and humid eastern evergreen rainforest. It is the largest known chameleon (males up to 685 mm TL)(Glaw and Vences 1994). This makes Furcifer oustaleti an ideal species to study phylogeography. According to their distribution, F. oustaleti should be found in all 6 major biomes allowing for cross-habitat comparisons and sample replicates Integration with UP206a I am a PhD student in the Ecology and Evolutionary Biology Department and Dr. Estrada has graciously allowed me to enroll in this urban planning course. Because the nature of my dissertation deals with evolutionary processes in Madagascar my midterm and final project were on this geographical location. I created maps of Madagascar at multiple levels illustrating its landscape ( i.e. forest cover, major rivers), administrative provinces, ecoregions (6), a human index grid, human footprint grid, “Last of the Wild” vector data, and deforestation/fires data which affected resea rch areas of interest for my study species. For the midterm project I sought to introduce Madagascar & its major provinces, locate provinces in which my species occurred, illustrate areas of expected low gene flow and I sought to show how fires are an increasing threat to chameleon habitats and pristine areas. For the final project I wanted to illustrate Madagascar’s global position using satellite images, give a deeper explanation of my sampling regime, give a description of my study sites, habitats and biological ecoregions, illustrate the impact of crop lands on chameleon habitats and show a deeper exploration of daily occurring fire data with Spatial Clustering Analysis. My current hypothesis is that the genetics of the individuals from my populations will indicate a phylogeographic pattern. The most prevalent hypothesis in the literature is that rivers have played a large role in the prevention of gene flow between populations historically. For the “original data” requirement of the final project, I incorporated the actual GPS locations and place names where my samples were collected by staff members of the Henry Doorly Zoo ( Omaha, Nebraska). Layouts including this data will be identified in the following description of final project. Layout Descriptions Layout 1 As stated in my introductory section, I wanted to illustrate Madagascar’s global position using satellite images I acquired from http://landsat.usgs.gov/gallery_view.php?category=nocategory&thesort=mainTitle In the layout above I geocoded the Madagascar file so that it would have the correct geographical position spatially in the world and correctly align with the “Continents” shape file I acquired from UCLA’s Map Share website. I turned on major city names for labeling which were included in the attribute table and changed the font colors and sizes so that they would be more visible on this high resolution map. I also included a key, for which I altered the fonts and text colors to match. Layout 2 Layout 3 was my favorite layout of the project because it illustrated the biological story behind project. As shown above, the sites in which my samples were collected are represented by the small “reptile” icon. Rivers are shown in light grey lines and bodies of water are in blue. I used symbology to turn each ecoregion into a different color shade of green. This emphasizes the fact that my study species does in fact occur in many types of habitats found in Madagascar. This is exciting because one would expect these varying populations to have varying genetic structures due to their differing environments. This map is also exciting because it shows that my samples were collected from all parts of the island. This will allow me to look for indications of genetic difference due simply to isolation by distance and such a finding would be meaningful when juxtaposed to competing theories which offer explanations of island diversity. Again, all shape files were geocoded so that size and distances could be measured correctly. Layout 3 Layout 3 is my most complex layout because it included a wealth of information with regards to sampling. The first satellite image used in Layout 1 appears here again, but this time I added my sample collection sites. I also included the major rivers and I used the “Select by Attribute” feature to highlight rivers which were in within a 5 mile distance of my collection sites. Those rivers are highlighted in orange. As shown in the inset, at least two of my collection sites overlap with a river ( Tsimanpetsotsa and Masoarive). Future sampling would need to insure populations on both sided of the rivers were sampled to compare genetic histories at the fine scale level. Overall, for the rivers highlighted, both sides of the river seem to be sampled which really allows me look for indications of unique genetic history ( I’m very excited about this finding). For the bottom maps, I wanted to emphasize sample sizes. As the legend indicates, two image sizes were used for collection sites; those with less that two samples ( or individuals) and those with more than 2. The main thing I wanted to show was that most of my sites have been under sampled. This simply illustrates the need for more sampling at these sites so statistical conclusions can be made about those data coming these sites. This is something I hope to personally accomplish in the 2009. Layout 4 Layout 6 illustrates how crops and agriculture might be impacting the habitats of my chameleons. I also layered this map with a forested area shape file I found on MapShare and I included rivers and lakes for reference. As emphasized in the inset maps, multiple sites are in close proximity to crop lands. This means that the pristine habitats in which these populations might evolved is being lost and this can lead to a number of genetic consequences for the species who live in these places. This layout really shows the need for conservation of pristine lands and habitats. It’s difficult to say what the genetic impact of such activity might be on my chameleon study. But the existence of an anthropogenic impact at all can be viewed as a negative. Layout 7 Layout 7 is in the same theme as layout 6 in that it illustrates the impact of human activity on my study as well as Madagascar as a whole. In this layout I picked the heaviest day of recorded fire incidences ( December 1, 400 fires) and selected locations based on their brightness indices ( information that was provided in the attribute table). I made the sizes of the dots reflective of the brightness of the fire. Colleagues in my laboratory believe that brightness is an indication of the size of the fire picked up by the satellite. I wanted to know if city proximity had any emphasis on the size of the fire, but the largest fires appeared to occur far away from major cites; in rural areas. Sadly, an abundance of fires are occurring in forested areas. This most likely illustrates the prevalence of slash and burn agriculture. For more information on this fire database please see: http://images.google.com/imgres?imgurl=http://www.maildogmanager.com/pics/hotspots /35img2.jpg&imgrefurl=http://maildogmanager.com/page.html%3Fp%3D000001XDDtj KFP98UZkgchujPw4uhcMNdgVqw%3D&usg=__zFkVVZH0xLiaDIH49uKjBoOi_jk= &h=133&w=200&sz=10&hl=en&start=213&sig2=W8IJoQv- VPG9jSugCzwo2g&um=1&tbnid=TbfIsVEK_dEo3M:&tbnh=69&tbnw=104&ei=qgE3S fffDYXaNdnarbYP&prev=/images%3Fq%3DMadagascar%2Bshapefiles%26start%3D19 8%26ndsp%3D18%26um%3D1%26hl%3Den%26client%3Dfirefox- a%26rls%3Dorg.mozilla:en-US:official%26sa%3DN . Layout 8 Because I was able to acquire fire data from everyday in the month of November I decided to see if there were any statistical patterns that could be concluded. To do so, I conducted a multi-distance spatial cluster analysis. This ArcGIS utility tool looks for clustering of fire incidences. It assumes that all points are random and tells you if the incidences are not random and by how much. Here we see that the y do cluster because the observed k is much greater that the upper confidence interval. If the obs. k falls between lower and upper confidence intervals the data is random. A special thanks goes Dr. Henri Thomassen, a post-doc in Dr. Tom Smith’s lab who helped me with this endeavor. Conclusions Human Impact on Chameleon Genetics Future work requires that I investigate the time scale of human impacts versus that of environmental variables on my dissertation study. It is assumed that genetic differences which might arise from human hab itat alterations are not detected by mitochondrial analysis. That level or resolution of disturbance is at too short a time frame to tremendously impact living or extant species. But whether this truly the case remains to be proven. Dispersal Ability Genetic dispersal distances are critical to establishing any real or meaningful hypothesis about diversifying processes. The strongest determin ing factor which will affect gene flow is my chameleon’s ( any other taxa’s) ability to disperse and subsequently mate. “Low-dispersing” taxa are expected to be most impacted by rives for this reason. But dispersal ability does play a tremendous role and I would have liked to have incorporated that into this analysis. Conservation Conservation appears to be the most necessary step. The biodiversity of Madagascar is unlike that of any other place in the world. I was drawn to evolutionary biology in Madagascar because it presents the ideal location to study the results of strong diversifying processes. Selection pressures in combination with separation have created unique populations which have, overtime, diverged and evolved into different species uniquely adapted for their environments. As a young scientist who would like to trace this process in reverse; from present to past, conserving the stability of these processes is paramount and critical to my understanding of the underlying forces at work. At an even more grim level, none of this work or knowledge, is valuable if the areas these species live in and have evolved in are lost. Loss of habitat unravels the stabilizing processes which have created and maintained the beautiful diversity which exists today.