1
Bachelor Thesis Earth Sciences
Past sea-level change since the LGM and its effect on the geography of
the Ryukyu islands and the insular area of Kyushu Japan
E.L.P van Es
Student number: 10810846
Supervisor: Dr. K.F. Rijsdijk
IBED - UVA
2
Abstract
This study is conducted to determine the quantitative effect of sea-level rise since the last glacial maximum (21kyr BP till present) on the geography of the Ryukyu islands and the islands of Kyushu of Japan. A paleo-reconstruction study is performed in ArcGIS, to calculate the area loss and distance increase of the Ryukyu islands since the LGM. Also, the connectivity of the islands of Kyushu are modelled. The results show that there has been a great loss of area, with a total of >70%. Also the relative distance increased profoundly. Ranging between 19% till >280%. The reconstruction of the connectivity of the islands to Kyushu showed that the longest attachment of one of the islands has been approximately 60k years. Diversity in species and similarity in species to Kyushu may be explained by this connection. Also the area loss and distance increase of the Ryukyu islands due to the sea-level rise will have altered the biota. Causing higher extinction rates, lower migration rates and higher speciation. However, further research has to be carried out to see if the quantifications of this research are comprehensive with ecological outcomes.
3
Content
1.Introduction...4-7 - Theory - Research Area - Research Questions 2. Methods...8-9 3. Results...10-13 - Area Loss - Distance Increase - Connectivity 4. Discussion...14-16 - Area loss - Distance increase - Connectivity - Future Research - Improvements 5. Conclusion...17 6. Acknowledgements...18 7. References...19-20 8. Appendices...21-264
Introduction
Theory
Biodiversity of islands has long been studied and particularly oceanic islands are known to have high species richness (Kier et al., 2009). In 1967, MacArthur and Wilson introduced the Equilibrium Theory of Island Biogeography (ETIB). This theory explains that the island species richness is caused by a balance, or equilibrium, between immigration and extinction. Immigration and extinction rates are affected by respectively distance to the mainland and island size. Island nearer to the mainland have higher immigration rates than distant ones and smaller islands have more extinctions, than larger islands (figure 1;MacArthur & Wilson, 1967).
Although MacArthur & Wilson made it clear how species diversity patterns are influence by the area and distance on islands they did not include the highly dynamic nature of the oceanic islands. The islands are subject to geological-climatic processes and the ecological EITB did not include these vulnerable natures of islands, but sees islands as static entities (Richman, Case, & Schwaner, 1988; Fernández‐Palacios et al., 2011). Whittaker, Triantis & Ladle (2008) postulated a framework, the General Dynamic Model (GDM), that integrates the concepts of MacArthur & Wilson (1967), but includes the important geological processes that underlie the evolutionary and ecological processes. However, the GDM only explains island dynamics and species evolution for very long timescales, it does not include the shorter time spans effecting species richness and speciation. Fernández-Palacios et al. (2016) formulated the Glacial Sensitive Model (GSM). The GSM includes the complexity of sea-level curves linked to glacial cycles as shorter timescale process and its effect on the geography of oceanic islands and their biota.
Since the origin of our planet sea-levels have been oscillating. So the sea-level follows a recurring curve where glacials alternate with interglacials. The periodicity of this curve is about 100,000 years alternated by shorter periodicities of 41,000 years (Ruddiman, 2003). The most widely accepted theory that explains this is Milankovitch’s hypothesis, which devotes the cause of these periodicities to the earth’s orbit around the sun. This means that the amount of solar radiation is of influence of earth’s climate and thus on the existence of ice ages (Raymo & Huybers, 2008). Milankovitch’s hypothesis states that there are three cycles the earth passes around the sun, the eccentricity, axial obliquity and precession. The eccentricity is the change of the orbit of the earth around the sun, from a more circular to ecliptic orbit. With the ecliptic orbit leading to more extreme seasons. The periodicity of this
Figure 1: Graph of the Equilibrium Theory . Left part shows distance, right part shows area (MacArthur & Wilson, 1967).
5 cycle is 100,000 years (Hays, 197;Ruddiman, 2003) and this is thought to be the main constantly occurring reason of ice ages. The axial obliquity is the tilt of the earth with respect to the orbital plain. It varies between 22.1° an 24.5°, with a bigger angle causing more severe seasons and happens with a periodicity of 41,000 years, ice ages that originate from this periodicity do not always exist, only when certain circumstance are met (Raymo & Huybers, 2008). Sea-levels coincide with these periodicities of climate change and thus follows the same glacial and interglacial patterns.
The most recent cyclical period started approximately 120,000 years ago and lasted for circa 90,000 years (Cutler et al., 2003; Figure 2). Apart from some sea-level rise during small interglacial events (figure 2b), the mean sea-level dropped to -120 meters. This lowest sea-level point coincides with the moment where the ice sheets were at their largest extent and is called the Last Glacial Maximum (LGM; ca 21 kyr BP, Cutler et al., 2003; Camoin et al., 2003). Since this LGM there has been a rapid rise in sea-level (Cutler et al., 2003, point D till F). The average raise has been over 120 meter, converted, this is circa 6 meters per 1000 years (Simaiakis et al., 2017).
However, these are all mean values or eusostatic sea-level values. But sea-levels did not rise equally over the globe. This differentiation is due to isostatic sea-level changes (Mörner, 1976). Eustatic sea-level change are all the processes that influence the total volume of the world’s oceans (Rovere, Stocchi & Vacchi, 2016). The earth’s surface is not perfectly round, but has a relief. This is called the geoid (Mörner, 1976). Due to this unevenly shaped relief, changes in water volume, will not distribute equally over the globe, but eusostatic water levels changes have influence all around the globe. The changes in water volume are caused by the melting or accumulation of continental ice sheets, redistribution between different hydrological reservoirs, variations in ocean water density and by the change in the basin due to sea-floor spreading or sedimentation. The eustatic sea-level change has global influence (Mörner, 1976). Besides the global sea-levels shifts due to volume changes, there is also local change in sea-level due to isostatic influences (Mörner, 1976). The sea-level changes, without a shift in volume. Therefore it will only have local influence. Isostatic change arises due to changes in the earth’s crust (Rovere, 2016). For example: When ice sheets grow on land masses they push the earth’s crust down. When they start to melt, the pressure is released and the earth’s crust bounces back. It will return higher than the old crust level and will cause an decrease in relative sea-level (Rovere, 2016). Also local tectonic uplift or subsidence will cause local sea-level changes (Simaiakis et al., 2017).
Although the sea-levels did not rise equally around the globe, they still rose1 (Cutler et al., 2003). This had a profound influence on - in particular - oceanic islands and their biota (Fernández‐Palacios et al., 2011). The long period (ca 90 kyr) of the glacial period, gave the islands time to grow and give room to species for migration and speciation. However, the rapid sea-level rise (10 kyr) that followed had a profound impact on the biogeography of the islands. The islands lost enormous amounts of their areas and grew further apart from the mainland. Causing higher extinction rates and lower migration rates. This lowers the total species richness (Fernández‐Palacios et al., 2011, 2016; Rijsdijk et al., 2014). Heaney (2000) explains that fragmentation of islands also leads to lower species richness, however to higher
1
Except for ice-proximal locations. See: Khan et al., 2015
Figure 2: Reconstruction of the average sea-level from 120 kyr up till now (Rijsdijk et al., 2014).
6 speciation. The higher isolation of the islands, causes less gene flow, so species diversification is favoured. This speciation results in higher endemism (Heaney, 2000). What the final biota diversity will be has to be examined, however the new equilibrium, after sea-level rise, has to be reached. This state is called relaxation, where an “extinction debt” is being paid, which means that the biota on the islands are not yet fully adapted to the new support of the islands (Triantis et al. 2010).
So, theory makes it clear that profound changes of oceanic islands geography due to sea-level rise effected the biodiversities on those islands (Weigelt et al., 2016). Therefore, in this research the profound changes that arise due to sea-level change will be examined. So, it will be quantified what effect sea-level rise had on the area loss, distance increase and connectivity change of an insular area of Japan.
Research Area Japanese Islands
Oceanic islands are considerably affected by different forces like plate tectonics, island ontogeny, glacial cycles and volcanic activity (Fernández‐Palacios et al., 2011). Therefore they are highly dynamic, but their constantly changing nature makes them hard, however interesting to examine.In this research it is chosen to study the Ryukyu islands and the islands of Kyushu of Japan.
Japan exists of over 7000 islands. The Japanese islands lie at the junction of four tectonic plates. The Pacific and the Philippine Sea oceanic plates and the North American (Okhotsk), the Eurasian (Amurian) continental plates (Taira, 2001). The area therefore is very subject to subductions, collisions, earthquakes, volcanic activities and deep trenches.
Although, there are over 7000 islands, four major islands make up the greatest part of the arcsystem: Hokkaido, Honshu, Shykoku and Kyushu (Taira, 2001). This research, however, cannot include all islands, therefore a specific researched area is chosen. Fernández‐Palacios et al. (2011) stated that big islands and their biota are relatively less influenced by sea-level rise than smaller islands, due to the relative small loss in area. To see a significant difference in the effect of sea-level rise on small oceanic islands this study therefore will focus on the island chain of Ryukyu in the South of Japan. Current biogeographic research is being conducted in this region (Kurita et al., 2018), which makes them interesting as well. Besides this the connectivity of one of the largest islands of
Japan, namely Kyushu, and the islands around it will be researched. The specific research area is shown in figure 4 and a closer image of the islands with names can be found in the appendix 1.2-1.3.
Figure 3: Satellite photo of the researched area. In the top right corner is the insular area of Kyushu. Below that the belt of the Ryukyu islands till Taiwan (Google Earth, 2018).
7 Aim
The aim of this study is to quantify the effect of sea-level rise on the geography of the insular areas of Ryukyu and Kyushu of Japan since the last glacial maximum. This will be done in order to contribute to the Glacial Sensitive Model and to further local, as well as global, biogeographical studies on colonization and migration of species due to sea-level changes. A ArcGIS based paleo-geography reconstruction of the insular area will provide the magnitudes and rates of the area decrease and distance increase.
The study will be divided in three sub-questions:
- How much did the insular area of Ryukyu decrease since the last glacial maximum? - How much did the distance to the mainland of the Ryukyu islands increase since the
last glacial maximum?
- How did the connectivity of the insular area of Kyushu change since the last glacial maximum?
8
Methods
In order to quantify the geographic changes due to sea-level change in the aforementioned insular area of Japan a detailed paleo-geographic research is performed using ArcGIS 10.4.1. With the use of reconstructions, polygons and geometrics, calculations were made to determine the relative area loss and relative distance increase of the Ryukyu Islands and to examine the interconnectivity of the Kyushu islands.
First, the required data had to be collected. As explained in the introduction the sea-level has not changed equally around the globe. In order to make accurate calculations on the insular area, Glacial Isostatic Adjustment (GIA) files are used. These files are adjusted according to the local isostatic values and thus contain the relative sea-level in Japan (Khan et al., 2015). The needed GIA files were provided by E. Koene. The files consisted of 22 TIFF raster files of Japan from 21,000 years ago till present. Meaning each file had a time step of 1000 years. The 22 files had a land-sea distinction. So, every file had a binary distribution with value 0 for sea and value 1 for land. This land and sea distribution is based on the Generalised Sea Level Equation (Kendall, Mitrovica & Milne, 2005). For further information or help contact E. Koene (Koene, 2017).
The files were WGS1984 projected and have a grid size of approximately 1 km2. For accuracy and comparison of eustasy and isostasy the GIA files were compared to a Digital Elevation Model (DEM) which contains the topography and the bathymetry of the insular area of Japan (Becker et al., 2009; GEBCO, 2014). Both the GIA files and DEM file had to be projected to the World Mercator coordinate system, which was the UTM zone 52N for the Kyushu and Ryukyu insular area (ESRI, 2018). The GIA files however failed to project properly. They subsequently had to be copied and an attribute table had to be added before they could be used for further reconstructions.
To calculate the area loss the 21ka map was used to delineate all the Ryukyu islands. This delineation was saved as a new layer. Subsequently, all the islands were separated by
lines (figure 5). Each island selection was then saved as a layer. Then, using a clipping model, every islands selection was clipped in combination with each 22 maps. Every clip than
Figure 4: WorkFlow model of the research.
Figure 5: Island selection of the Research Area. Each selection is used to clip 22 maps and calculate areas.
9 contained an attribute table with pixels for the value 0 and pixels for the value 1. As the pixels with value 1 contained land, those were used to calculate the total area for every island for every 1000 years. As said before the spatial resolution was approximately 1 km2, however to perform the pixel calculations the real grid size of 0.994 by 0.994 was used to get more accurate areas. For some true islands that got fragmented the area change per new existed island was also calculated. Hereafter the rates of change and total area change could be calculated in Excel.
To calculate the relative distance increase polygons were made of the Ryukyu islands and the nearest mainland, which were Taiwan, China and South-Korea. An example is showed in appendix 1.1. This is only done for 21k years ago and present day. Every island polygon was used in the Near Analysis of ArcMap. The attribute table then showed the nearest Euclidean distance from the coastline of the islands to the coastline of the nearest mainland. This Euclidean distance is a straight line as the crow flies. Next, the total distance changes for the islands were calculated in Excel and graphed.
To interpret the connectivity of the islands around Kyushu all the GIA maps were placed on top of each other. Every time one of the big islands separated from Kyushu it was noted in Excel. Hereafter, in combination with known sea-level curves it was calculated how long the islands had been connected to Kyushu. Above that it was calculated how the distance between Kyushu and the mainland changed.
So to conclude, ArcGIS reconstructions and calculations of the relative change in island area, distance and connectivity will be made. Hereafter maps and graphs are made to give a clear overview. It has to be noted that in this research that the effect on the geography of islands due to tectonic uplift and volcanic eruptions has not been taken into account.
10
Results
The results will be showed in the following order: area loss, distance increase and connectivity. Only the most relevant graphs and maps will be shown in this section. All other tables, graphs and maps can be found in the appendices. To create a general overview of the change in the research area figure 6 is presented. The grey parts are 21000 years ago, white land is present day. This is a graph from the GIA files. Comparison between the DEM and the GIA files gave only minor differences, see appendix 1.4. But for this research only the GIA files are used.
Area Loss
Since the end of the LGM the total area loss of the Ryukyu islands has been >70%. The total area loss per island ranged between >40 – 100%. However it was different for every island, most island area loss happened between 19k – 6k years ago. As literature says the highest increase in sea-level is around 16-15ka. This is also visible in the rates at which the islands lost their area. Appendix 2.2 shows the rates of relative area loss in percentages for the islands. As can be seen here as well as in figure 7, some islands completely vanished. The red islands are the ones which endured the greatest area loss, those are: Tarama, Kume, Tonaki and Zumami/Tokashiki.
11
12 Distance Increase
Since the end of the LGM the total relative distance change has increased with >165%. Figure 8 shows which islands had the largest increase in relative distance. The Southern Ryukyu islands, which are more close to Taiwan experienced less influence of distance change. The Northern Ryukyu islands in contrast had a relative distance increase of around 200%. The fragmented Okinawa grew most in distance. Table 1 shows the exact values.
Islands Closest distance
21ka (km) Closest distance 00ka (km) Relative increase (%) Closest Mainland Yonaguni 81 106 30,86 Taiwan Nakanougan 137 174 27,01 Taiwan Iriomotejima 134 181 35,07 Taiwan Ishigaki 134 217 61,94 Taiwan Hateruma 164 196 19,51 Taiwan Tarama 128 271 111,72 Taiwan Miyakojima 161 316 96,27 Taiwan Okinawa 146 560 283,56 Taiwan Tonaki 146 533 265,07 Taiwan Kume 146 494 238,36 Taiwan Zamami/Tokahshiki 146 539 269,18 Taiwan Aguni 162 549 238,89 Taiwan Izena/Iheya 187 627 235,29 China Yoron 231 669 189,61 China Okinoerabu 225 668 196,89 China Tokunoshima 231 688 197,84 China
Amami 226 675 198,67 South Korea
Kikai 284 699 146,13 South Korea
0 50 100 150 200 250 300
Relative distance increase %
Figure 8: Relative distance increase per island. From the Yonaguni till Okinawa are the Southern islands. From Okinawa to Kikai are the Northern Ryukyu islands.
13 Connectivity
Figure 9a-d shows the time when the different islands separated from Kyushu. The islands around Kyushu have been connected to Kyushu till 17ka, 15ka, 14ka, and 12ka. The longest connection has been approximately 60k years. Also the distance between Kyushu and South-Korea has been calculated and the closest distance being a mere 10.6 km.
Figure 9: Paleo-reconstruction of the times the islands got separated from Kyushu. a) First separation of the Tsushima island. b) Separations of Yakushima/Tanegashima. c) Separation of the Koshiki island and of the two islands from 9b. d) Seperation of the last two islands Fukue island and Iki.
14
Discussion
The results from this study provide a quantitative geographic reconstruction of the Ryukyu islands and the islands of Kyushu. The results show that the change in sea-level had a tremendous effect on the insular area.
Area loss
The Ryukyu islands have experienced a total area loss of over 70%, whereby multiple islands got fragmented or some completely disappeared. The average area loss per island is 72,5%, however if the fragmentation of Okinawa and Iriomotejima/Ishigaki is taken into account the average loss per islands is >75%.
The rates of area loss over the last 22ka years per island has been plotted (Appendix 2.2). The highest rates of area loss occur around 16ka and this rapid change lasted approximately till 8ka. This appearance of area loss of the Ryukyu islands is consistent with literature about the timing of the majority of sea-level rise (Lambeck et al., 2002) The rates also show that the Southern Ryukyu islands had a somewhat higher degree of change, they lost their area faster, than the Northern Ryukyu islands. This can be explained by the relative sea-level rise, due to the bathymetry of the earth’s surface around those islands, or the island morphometry, which depends mostly on the age of an island (Whittaker, 2007). Islands that are older are more prone to change than younger islands with steep shores (Fernández-Palacios et al., 2016). The ontogenetic stage, or ‘age’, may explain why the Southern Ryukyu islands changed faster than the Northern ones.
So, the results show that the Southern Ryukyu islands have endured the highest area loss, but the fragmented Okinawa is second in line. The red islands which experienced the greatest area loss (>95%) are Tarama, Kume, Tonaki and Zamami/Tokashiki. This enormous change has had an large effect on the biodiversity of islands (Fernández-Palacios et al., 2011; Weigelt et al., 2016). The islands that suffered most area reductions will have suffered the highest number of species extinctions. However, due to the fast change in area loss, species have not yet equilibrated to new situation were a ‘lowered equilibrium’ has to be reached. They are in still in the extinction debt. This also means that there is a disproportional amount of endemic species on the islands that endured rapid area loss (Simaiakis et al., 2017).
For the fragmented islands of Kume, Tonaki and Zamami/Tokashiki there will be more speciation (Heaney, 2000). Therefore on those islands more native non-endemic species will experienced extinction, because these species could not adapt quickly enough to the new situation.
Distance increase
The Ryukyu islands have experienced an average distance increase of 165%. But the distance increase differed greatly between the islands, ranging from approximately 30 - 280%. The mainland at 21ka included all East-Asia, over the past 21ka years this is completely withdrawn, resulting in three nearby mainlands: Taiwan, South-Korea and China. In this study it was chosen to include Taiwan, although it is actually an island, but to exclude Kyushu as mainland. This is done because Taiwan has been connected to the mainland and shares the same species as the other mainlands. Kyushu however, as will be shown in the next paragraph has never been physically connected to the mainland. Therefore Taiwan, South-Korea and China were used in the distance calculations.
The results show that the Southern Ryukyu islands have experienced relatively small distance increase, whereby some Nothern islands have endured an enormous distance increase. This can be explained by the fact that the shoreline of Taiwan, which was the closest mainland for all Southern islands, is still at the approximately the same place as the shoreline
15 from the mainland of 21ka years ago was. So the only relative distance increase for those islands arises from the fact that their own shorelines retracted. For the Northern Ryukyu islands the closest mainland varied between Taiwan, China and South-Korea. Every islands experienced a great distance increase, but the largest distance increases were calculated for Okinawa, Tonaki and Zamami/Tokashiki. This is due to the fact that these are fragmented islands. In 21ka the island Kume belonged to Okinawa. The closest distance was calculated from Kume. However, as it separated from Okinawa, Tonaki and Zamami/Tokashiki the newly emerged islands and there shores were used to calculate the distance. Now the distance did not only increase due to retracting mainland, but also due to the fact that new islands formed, with shorelines that lie further backwards.
The distance increase also had its influence on the species richness. The immigration and colonisation of islands becomes less common if distance increases and this has its effect on gene-flow of the islands (Siamaikis et al., 2017). Therefore the islands with the largest distance increase will have suffered the highest species decrease.
The combination of the results from area loss and distance increase shows that the islands with the highest area-distance-change are Kume, Tonaki and Zamami/Tokashiki. The biota on these islands will therefore probably have suffered the most. Caused by the highest extinctions rates, distribution change and changed genetic environment (Siamiakis et al., 2017). However because they are fragmented islands there will be higher speciation on those islands, with more niche endemic species (Rijsdijk et al., 2014)
Tarama, which lost most of its area too and increased relatively much distance (100%) compared to the other Southern Ryukyu islands, will have significant different species pattern as well.
Connectivity
The results of the reconstruction of the islands around Kyushu shows that they have been connected to Kyushu and started separating from 18k to 12k years ago. Tsushima, which is situated between Kyushu and South-Korea, separated the first. Between 18k and 17k years ago the mean sea-level rose from -116 m to -109 m (Cutler et al., 2003). Tsushima would have been connected to Kyushu for approximately 6000 years (Lambeck et al., 2002). However, the Fukue Islands and Iki only got separated between 13k and 12k years ago. The mean sea-level than was respectively between -76 and -65 m. Taken figure 1 and Lambeck et al. (2002) into account the will probably have been connected to Kyushu for around 60000 years, this is 10 times longer than Tsushima. Islands that were rapidly separated will probably have more extinction rates of native continental species of Kyushu, than islands that got longer to isolate, because those species had more time to adapt (Simaiakis et al., 2017). Above that the species that are found on the Fukue islands and Iki will probably be more the same to those found on Kyushu (Rijsdijk et al., 2014).
On top of the connection of the islands around Kyushu a distance calculation was made between Kyushu and Korea. At the LGM the distance between Kyushu and South-Korea was only 10.6 km. It may have been possible that humans crossed from East-Asia to Japan at that time (Manabe et al., 2008). This may have its influence on the endemic and native species on the separate islands.
In the introduction it has been explained how the sea-level curves oscillate. Above findings concerning the connectivity of the islands around Kyushu may not only tell something about how these islands were connected past 100k years, but also can enhance our understanding of longer time spans. Thus, creating these overviews for islands for how long they have been connected can enhance the knowledge and understanding of diversity in species.
16 Future research
Although the results show a profound effect on the Ryukyu islands and the islands of Kyushu, research still has to be done on the species richness and endemics of those islands. Does it hold true that there are more extinctions on the islands with the highest area-distance-change, then on other islands? Also, the Ryukyu islands have always been apart and quite far from the mainland. What influence does this have on the species richness and endemics of those islands? Future ecological research should provide the answer to the change in species on the islands.
In this study a paleo-reconstruction has been made. However, past influences presence and the cyclical nature of sea-levels and climate can tell us something about the future. It is very interesting to have further research in the future impact of sea-level rise on the biodiversity of islands (Mimura, 2013).
Also geological research in the morphology of the islands should be made. What is the bathymetry and the ontogenitic stage of the island. So, there can be an exact explanation of why some islands changed faster than others.
Improvements study
Although the study gave profound results, there is certainly room for improvement. In the method it is explained that the areas of the islands were calculated using the pixels with value 1. This was done, because with only one selection all the areas for every 1000 years could be calculated. Thereafter the rates per 1000 years could be interpretated. However, when the polygons were created for the distance calculations it became clear that the pixels and the polygons did not completely overlap. Therefore, the area calculations will be less precise and will probably be bigger than there equivalent polygon calculations. Although, time-consuming, it would be better to create polygons and use these for area calculations instead of pixels.
Another shortcoming of using these selections instead of polygons was that it was harder to make selections of fragmented islands. The lines which were drawn between the islands, became too thick to divide islands which were near to each other. Therefore most fragmented islands were calculated as one island. Only for the islands of Okinawa and Iriomotejima and Ishigaki, there is a distinction made in fragmented islands, but this was only due to the fact that the islands were quite far away from each other. If polygons are used every small island can be selected and the influence of fragmentation on islands and there biota can be research as well.
Comparisons of the GIA files and the DEM showed that there was minor difference between both. However, the research area is at quite a low altitude, therefore it will not have experienced influence of compression and decompression from ice sheets. Not only ice sheets could influence the relative sea-level, so these must also be included in the isostatic changes, before it can improve this study.
As made clear in the methods, this study does not incorporate the influence of uplift and new sedimentation due to volcanic eruptions. To improve this study, it is of importance to take these changes into account as well, because there is tectonic uplift (Yokoyama et al., 2016) and it may influence the outcome of mainly the calculations on area loss (Simaiakis et al., 2017).
17
Conclusion
There has been a lot of research on the highly dynamic nature of oceanic islands and their biota. Multiple theories and frameworks have been postulated, but there is not a comprehensive theory that could include time spans on shorter timescales (Whittaker et al., 2008; Fernández-Palcaios et al., 2016). The glacial sensitive model tries to combine the equilibrium theory with the influence of sea-level change. The sea-level follows a cyclical curve with glacials and interglacials. The last glacial had its maximum 21k years ago (Cutler et al., 2003). Then the a rapid sea-level rise started and in 20k years the sea-level increase with 120 meters. This had a profound effect on all sorts of landmasses, but especially on islands.
In this research the effect of this sea-level rise on the Ryukyu islands and on the insular area of Kyushu has been examined. The area loss, distance increase and connectivity of the research area has been quantified and reconstructed.Indeed, the results showed an enormous change. Over 70% of total islands area has been flooded by water. Especially Tarama and the fragmented islands Kume, Tonaki and Zamami/Tokashiki, severed area loss, with over 95%. The distance of the islands to the mainland also increased significantly. With an average relative distance increase of 165%. The highest relative increase was 283% for Okinawa, followed by Tonaki and Zamami/Tokashiki. The sea-level rise also had an influence on the connectivity of the islands around Kyushu. The connectivity of the islands to Kyushu ranged from only 6000 years to over 60,000 years.
These geographic changes of the islands will have had its influence on the present biota. Examples are higher extinctions and lower migration rates. Higher speciation and diversification of endemic species. The connectivity gives knowledge about where to find the same species. But future ecological researched should point out if indeed this geographic changes altered the islands biota.
However, although future researched should be carried out, it still may be concluded that the sea-level rise since the last glacial maximum had a profound effect on geography of the Ryukyu islands and the connectivity of the Islands of Kyushu.
18
Acknowledgements
First of all, I would like to thank Dr. K.F. Rijsdijk for being the supervisor of my Bachelors Thesis and for giving support whilst writing it. Next, I would like to thank Erik Koene for providing the needed GIA files. Furthermore, I would like to thank Sietze Norder for giving me Excel data of the mean sea-levels. Lastly, I would like to thank Thijs de Boer for granting me access to the GIS-Studio and his help with problems with ArcGIS.
19
References
Camoin, G., Montaggioni, L. & Braithwaite, C. (2004) Late glacial to post glacial sea levels in the Western Indian Ocean. Marine Geology, 206, 119–146.
Cutler, K.B., Edwards, R.L., Taylor, F.W., Cheng, H., Adkins, J., Gallup, C.D., Cutler, P.M., Burr, G.S. and Bloom, A.L., (2003). Rapid sea-level fall and deep-ocean temperature change since the last interglacial period. Earth and Planetary Science Letters, 206(3-4), pp.253-271.
ESRI. (2018). Universal Transverse Mercator Zones. Esri Geodatabase. URL: https://arcgis.com/.
Fernández‐Palacios, J. M., de Nascimento, L., Otto, R., Delgado, J. D., García‐del‐Rey, E., Arévalo, J. R., & Whittaker, R. J. (2011). A reconstruction of Palaeo‐Macaronesia, with particular reference to the long‐term biogeography of the Atlantic island laurel forests. Journal of Biogeography, 38(2), 226-246. Fernández‐Palacios, J.M., Rijsdijk, K.F., Norder, S.J., Otto, R., Nascimento, L., Fernández‐Lugo, S., Tjørve, E. and Whittaker, R.J., (2016). Towards a glacial‐sensitive model of island biogeography. Global Ecology and Biogeography, 25(7), pp.817-830.
GEBCO. (2014). GEBCO_2014 Grid. General Bathymetric Chart of the Oceans. URL: https://www.gebco.net/.
Google Earth. (2018). Satellite picture of the research area. Google
Hays, J. D., Imbrie, J., & Shackleton, N. J. (1976, December). Variations in the Earth's orbit: pacemaker of the ice ages. Washington, DC: American Association for the Advancement of Science. MacArthur, Robert H.; Edward O. Wilson (1967). The Theory of Island Biogeography. Princeton, N.J.: Princeton University Press. pp. 203 pp. ISBN 978-0-691-08836-5.
Manabe, Y., Kitagawa, Y., Oyamada, J., Igawa, K., Kato, K., Kikuchi, N., Maruo, H., Kobayashi, S. and Rokutanda, A., (2008). Population history of the northern and central Nansei Islands (Ryukyu island arc) based on dental morphological variations: gene flow from North Kyushu to Nansei Islands. Anthropological Science, 116(1), pp.49-65.
Miller, K.G., Kominz, M.A., Browning, J.V., Wright, J.D., Mountain, G.S., Katz, M.E., Sugarman, P.J., Cramer, B.S., Christie-Blick, N. and Pekar, S.F., (2005). The Phanerozoic record of global sea-level change. science, 310(5752), pp.1293-1298.
Mimura, N. (2013). Sea-level rise caused by climate change and its implications for society. Proceedings of the Japan Academy, Series B, 89(7), 281-301.
Mörner, N. A. (1976). Eustasy and geoid changes. The Journal of Geology, 84(2), 123-151.
Kendall, R. A., Mitrovica, J. X., & Milne, G. A. (2005). On post-glacial sea level–II. Numerical formulation and comparative results on spherically symmetric models. Geophysical Journal International, 161(3), 679-706.
Khan, N.S., Ashe, E., Shaw, T.A., Vacchi, M., Walker, J., Peltier, W.R., Kopp, R.E. and Horton, B.P., (2015). Holocene relative sea-level changes from near-, intermediate-, and far-field locations. Current Climate Change Reports, 1(4), pp.247-262.
20 Kier, G., Kreft, H., Lee, T.M., Jetz, W., Ibisch, P.L., Nowicki, C., Mutke, J. and Barthlott, W., (2009). A global assessment of endemism and species richness across island and mainland regions. Proceedings of the National Academy of Sciences, 106(23), pp.9322-9327.
Koene, E. (2017). Paleo sea level modelling. PHD-student Zürich.
Kurita, T., Honda, M., & Toda, M. (2018). Species delimitation and biogeography of the Ryukyu ground geckos, Goniurosaurus kuroiwae ssp.(Squamata: Eublepharidae), by use of mitochondrial and nuclear DNA analyses. Journal of Zoological Systematics and Evolutionary Research, 56(2), 209-222. Lambeck, K., Esat, T.M. & Potter, E.‐K. (2002) Links between climate and sea levels for the past three million years. Nature, 419, 199–206.
Raymo, M. E., & Huybers, P. (2008). Unlocking the mysteries of the ice ages. Nature, 451(7176), 284.
Richman, A.D., Case, T.J. & Schwaner, T.D.(1988) Natural and unnatural extinction rates of reptiles on islands. The American Naturalist, 131, 611–630.
Rijsdijk, K. F., Hengl, T., Norder, S. J., Otto, R., Emerson, B. C., Ávila, S. P., López, H., Loon Van, E., Tjorve, E. & Fernández‐Palacios, J. M. (2014). Quantifying surface‐area changes of volcanic islands driven by Pleistocene sea‐level cycles: biogeographical implications for the Macaronesian archipelagos. Journal of Biogeography, 41(7), 1242-1254.
Rovere, A., Stocchi, P., & Vacchi, M. (2016). Eustatic and relative sea level changes. Current Climate Change Reports, 2(4), 221-231.
Ruddiman, W. F. (2003). Orbital insolation, ice volume, and greenhouse gases. Quaternary Science Reviews, 22(15-17), 1597-1629.
Simaiakis, S. M., Rijsdijk, K. F., Koene, E. F., Norder, S. J., Van Boxel, J. H., Stocchi, P., ... & Tjørve, K. M. (2017). Geographic changes in the Aegean Sea since the Last Glacial Maximum: Postulating biogeographic effects of sea-level rise on islands. Palaeogeography, Palaeoclimatology, Palaeoecology, 471, 108-119.
Taira, A. (2001). Tectonic evolution of the Japanese island arc system. Annual Review of Earth and Planetary Sciences, 29(1), 109-134.
Triantis, K.A., Borges, P.A., Ladle, R.J., Hortal, J., Cardoso, P., Gaspar, C., Dinis, F., Mendonca, E., Silveira, L.M., Gabriel, R. and Melo, C., (2010). Extinction debt on oceanic islands. Ecography, 33(2), pp.285-294.
Weigelt, P., Steinbauer, M. J., Cabral, J. S., & Kreft, H. (2016). Late Quaternary climate change shapes island biodiversity. Nature, 532(7597), 99.
Whittaker, R. J., Ladle, R. J., Araújo, M. B., María Fernández‐Palacios, J., Domingo Delgado, J., & Ramón Arévalo, J. (2007). The island immaturity–speciation pulse model of island evolution: an alternative to the “diversity begets diversity” model. Ecography, 30(3), 321-327.
Whittaker, R.J., Triantis, K. & Ladle, R.J.(2008) A general dynamic theory of oceanic island biogeography. Journal of Biogeography, 35, 977–994.
Yokoyama, Y., Maeda, Y., Okuno, J. I., Miyairi, Y., & Kosuge, T. (2016). Holocene Antarctic melting and lithospheric uplift history of the southern Okinawa trough inferred from mid-to late-Holocene sea level in Iriomote Island, Ryukyu, Japan. Quaternary International, 397, 342-348.
21
Appendices
1. METHODS AND ISLAND NAMES
1.1 Example of Near Analysis between the three mainlands (orange), the mainland in 21ka
22
1.2 Ryukyu Islands with names. Islands on the left are referred to as Southern Ryukyu islands,
23
24 2. RESULTS AREA, DISTANCE & CONNECTIVITY
25 0 20 40 60 80 100 120 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Northern Ryukyu Islands
Okinawa Izena/Ih eya Aguni Yoron Okinoer abu 0 20 40 60 80 100 120 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Southern Ryukyu Islands
Yonaguni Nakanou gan Iriomotej ima/Ishig aki Tarama Miyakoji ma Unknow n 2
26
2.3 Paleo-reconstruction of distance change of Kyushu to South-Korea & table values of
connectivity
Connectivity Kyushu
Distance increase 21_ka (km) 00_ka (km)
Kysushu & South Korea 10,6 165,3
Seperation from Kyushu Time bp (k years) Connected (k years)
Yakushima 16-15 10 Tanegashima 16-15 10 Shimo-Koshiki 15-14 13 Fukue 13-12 > 60 Iki 13-12 > 60 Tsushima 18-17 6
