analysis of biogeographic impacts
Keywords: insular biogeography, Late Pleistocene, last glacial maximum, sea-level rise,
species richness, endemism, Philippine archipelago
Bachelor thesis
Thomas Budie
ID: 10660607
03-07-17
Amsterdam
Supervisor: Dr Kenneth Rijsdijk
Co-supervisor: Dr Bas van Geel
Abstract
Since 1967, studies of island biogeography have been dominated by ‘The Theory of Island Biogeography’ from the ecologists McArthur and Wilson. They stated that the amount of insular species is determined by an equilibrium between immigration and extinction rates, which are influenced by insular isolation and size. While this theory is of great importance, it is too static to fully capture the dynamic character of islands. For instance, speciation has also proven to be a major drive behind species diversity and endemism. Moreover, past geographical changes such as sea-level rise should be integrated as well as this results in area loss, distance increase, fragmentation and disappearance of land bridges: all aspects that influence immigration and extinction. A GIS-based reconstruction of the Philippine archipelago has been created in order to discover the consequences of sea-level rise after the Last Glacial Maximum (LGM). It has turned out that the region has been affected greatly, as 40% of total insular area was lost (with a majority of islands losing >75%) and the mean relative distance increase almost reached 4000%. These features have been combined to form a qualitative index of Area-Distance-Change that shows the most affected islands. High ADC islands are expected to show high extinction rates, low immigration rates, high extinction debts and proportionally lower species richness (and vice versa for low indices). Documentation of mammalian species diversity/endemism in the Philippines has been collected and reviewed to verify these expected consequences of post-LGM sea-level rise. However, there are still many anomalies which highlights the importance of integrating more insular features throughout space and time in future analyses. By taking the first steps towards the complex reality of island dynamics, this research aims to provide a basis for a comprehensive model of island biogeography, with the Philippines in particular.
Keywords: insular biogeography, Late Pleistocene, last glacial maximum, sea-level rise,
species richness, endemism, Philippine archipelago
Acknowledgements
I would like to thank Dr Kenneth Rijsdijk for all his enthusiasm and support throughout the project. He has been of great help and he really motivated me to put a lot of effort in this thesis. Second, Cyril Hammoud was also very important for this research, as he provided essential steps and aspects of GIS-based paleogeography. Because of him, I could go through the process rather unimpeded. Furthermore, my fellow students who also analysed the biogeography of the Philippines, Ranes Rioza and Jeroen Gerdes, were of great support (morally and practically). At last, my gratitude goes towards Professor Lawrence Heaney for having a skype conference with us and of course for all his scientific research (which is a major fundament of this project).
Table of contents
AbstractAcknowledgements 1. Introduction
a. The dynamic equilibrium theory b. Island dynamism
c. Milankovitch hypothesis
d. Sea-level change and glacial isostatic adjustment 2. Research questions and aims
3. Methods a. Data
b. GIS-based paleo-geographic reconstruction 4. Results a. Paleo-geographic reconstruction b. Area loss c. Distance increase d. Area-Distance-Change Index 5. Discussion a. Area loss b. Distance increase c. Biogeographic comparison
d. Glacial isostatic adjustment and improvements 6. Conclusion
7. Bibliography 8. Appendix
1. Introduction
Since 21 thousands years ago, Earth’s climate rapidly experienced an increase of temperature and accompanied sea-level rise of 120 meters (Lambeck et al. 2014). This Late Pleistocene climatic shift has had a significant effect on the biogeography of islands (Fernández-Palacios et al. 2015). The most apparent consequences of higher sea-levels on insular systems are area loss, increased isolation, dis-connectivity and fragmentation (Rijsdijk et al. 2014). This highly dynamic geographical feature of islands leads to an incredibly complex construction of species richness and degree of endemism. For instance, area loss and increased distance to mainland lead to higher extinction rates, while insular isolation is simultaneously linked to higher speciation rates (Heaney 2000). A GIS-based reconstruction of the Philippine paleo-geography will reveal geological changes in the region, which will be laid out against existing literature of the biogeography. The Philippines are considered as highly appropriate for this study, because aside from being one of the largest archipelagos (+7000 islands), the area contains an extremely high concentration of (endemic) species per land unit (Brown et al. 2013). Especially mammalian species richness and endemism have been studied, hence the focus on this biota (Heaney 1986). A qualitative index, based on relative area loss and relative distance increase, was created in order to use during analysis of the Philippine biogeography.
a. The dynamic equilibrium theory
In 1967, MacArthur and Wilson introduced their famous dynamic equilibrium theory in their book “The theory of island biogeography”. This theory states that species richness on an island is the product of a dynamic equilibrium between two parameters: extinction and immigration (MacArthur & Wilson 1963). These parameters are subsequently linked to island area and isolation. For instance, large islands will have a greater carrying capacity, meaning such an island could harbour more species (low extinction rate). This is also because of less competition between species (Warren et al. 2015). Additionally, a large island will heighten the chance of species to reach the island during active or passive colonization; target-area effect. Isolation on the other hand has a negative effect on the immigration rate, because a far located island is much harder to reach. Also, a nearby island may experience a so-called rescue effect, where immigrants reinforce almost extinct populations (Fernandez-Palacios et al. 2015). In short, large and nearby islands are assumed to possess a higher species richness (figure 1).
Figure 1: Graphic display of the dynamic equilibrium model of island biogeography (Loh et al 2013). T = turnover rate, S = species richness.
b. Island dynamism and speciation
Although MacArthur and Wilson’s theory undoubtedly captures an important part of insular biogeography, it neglects the highly dynamic characteristic of islands throughout space and time. Considering islands as static environments greatly reduces the accuracy of explaining biogeographic patterns, since exactly this dynamism may withhold island ecosystems from reaching the equilibrium (Whittaker et al. 2008). The factors that should be taken into consideration are speciation and geological changes such as sea-level change (Heaney 2000).
In sharp contrast to the dynamic equilibrium theory, Heaney (2000) describes how isolation leads to higher speciation rates, and thus higher species richness and endemism (figure 2). He states that a low immigration rate favours species diversification as low gene flows reduce homogeneity. According to Lomolino (2001), speciation will be most frequent on large, isolated islands (figure 3). Because of the exceptional high degree of endemic species in the Philippines and increased sea-level rise-related isolation, this hypothesis is of great significance in this study.
Furthermore, because Earth’s climate has frequently transitioned between glacial and inter-glacial periods during the Pleistocene, so has the sea-level (and therefore area / isolation). This could result into a species richness equilibrium that gradually moves along with the geological changes of an island. The period between newly set conditions and the eventual species equilibrium is called “relaxation”, where an “extinction debt” is being paid (Triantis et al. 2010). This means that a highly affected island due to sea-level rise may not show a deprived biodiversity yet.
Figure 2: Conceptual model of species diversification due to lower immigration rates (Heaney 2000). At point A, gene flow is low enough to set diversification process in motion. Point B shows moment where threshold is overcome and new species have their sister taxa on the island instead of source area.
Figure 3: Graphic display of tripartite model where speciation rate is incorporated as the third biogeographic process (Fernández-Palacios 2015).
c. Milankovitch hypothesis
The theoretical bottom-line of this research are climatic oscillations. While it is easily observable that the climate varies frequently throughout the years, its gradual fluctuations are much harder to quantify. This is because they can happen over a timespan of decades, centuries and millennia. Fortunately, through geo-ecological evidence these paleo-climates have be reconstructed and analysed (Berger 1980). For the past million years, the climate has been dominated by glacial cycles of 23,000, 41,000 and 100,000 years (Imbrie 1993). Although these reconstructions are broadly backed and hardly spur extensive debate, the mechanism behind the climate fluctuations is prime target for controversy (Broecker & Denton 1990).
The most accepted theory is the Milankovitch hypothesis, which describes the effect of Earth’s movements on its climate. To be clear, the temperature and thus the climate on Earth are bound to the amount of solar energy that reaches Earth’s surface: irradiation (Berger 1980). And precisely this variable alters if Earth’s movements change. The variations happen in Earth’s eccentricity, axial obliquity and precession, which are orbital forced (and thus produce a pattern) (Imbrie et al. 1993). Eccentricity shifts result into more circular or elliptical orbits around the sun, which respectively leads to less or more extreme seasons. Secondly, axial obliquity means that the axial tilt varies over time, with higher degrees of tilt producing more severe seasons. Lastly, Earth’s precession indicates the direction of the axis of rotation, since it wobbles on its axis. This will lead to even more extreme or mild seasons (depending on which hemisphere), since the hemispheres will flip when Earth is facing the sun during perihelion (the moment when Earth is closest to the sun during one orbit) (Hays, Imbrie & Shackleton 1976).
As soon as Earth’s climate transcended into an interglacial era 21 thousand years ago, the major ice sheets covering the Northern and Southern hemisphere started to melt. This has resulted into a sea-level rise of 120 meters (figure 4) (Lambeck & Chappell 2001).
Figure 4: Global eustatic sea-level change during the last 120k years. Periods F, E and D are of significance in this research (Fernández-Palacios et al. 2015).
d. Sea-level change and glacial isostatic adjustment
Because the glacial period endured for a much longer extent of time than the current interglacial, Earth’s sea-level has been significantly lower during the majority of the Late Pleistocene (figure 4) (Fernández-Palacios). This difference has had a profound impact on land bodies and on islands in particular. Two consequences have been mentioned before: area loss and distance increase. Other changes are fragmentation (paleo-islands splitting into smaller islands), formation of paleo-peninsulas and disappearance of land-bridges (table 1).
However, sea-level rise is less straightforward than assumed. Although it is commonly known that the sea-level is unevenly distributed across the Earth because of tidal forces, it also differentiates greatly on a much larger time scale. This is known as eustatic and isostatic sea-level changes, which are respectively global and local effects (Mörner 1976). Eustatic means that the volume of water or the basin of a sea changes. This variable depends on glacial freezing/melting or tectonic movement (sea basin becomes larger or smaller – and thus can hold more or less water). Isostatic sea-level change is, however, much different and complex. It comprises elevation increase or decrease of crust underneath or around water bodies, which leads to relative sea-level change. The mechanisms behind this are compression and decompression. Compression happens during the build up of massive ice sheets that literally push continental land downwards and surrounding (sea) crust upwards (figure 5). In opposite, decompression takes place when ice sheets (partly) disappear and the land slowly rebounds. Moreover, the ice sheet’s gravitational pull will produce local sea-level differences as well (Fleming et al. 1998). Because glacial ice sheets are predominantly present around the poles, their effects are especially significant for sea-levels in high latitude regions (Clark & Mix 2002). Nevertheless, this study has used digital elevation models adjusted for isostatic sea level change in order to improve the accuracy of reconstruction.
Figure 5: Schematic display of compression and decompression during glacial shifts (British Geological Survey 2017). During the glacial period, the ice sheet pushes the crust downwards. It rebound as soon as the ice sheets has disappeared during the interglacial.
2. Research questions and aims
The aim of this research is to quantify the effects of sea-level rise after the Last Glacial Maximum (LGM) in the Philippine archipelago. Through a GIS-based reconstruction of the paleo-geography, data of area loss, isolation and fragmentation are obtained and combined to create a qualitative index of highest affected islands. These changes are essential for explaining species richness and degree of endemism. Furthermore, isostatic adjustment of the digital elevation model will be included in order to examine the effects of (de)compression and glacial gravitational forces in the Philippines. By answering the following research questions, this study aim to provide new insights into the biogeographic complexity of the Philippine archipelago:
“How much did Philippine insular area decrease after the Last Glacial Maximum?”
“How much did Philippine insular distance to mainland increase after the Last Glacial Maximum?” “What are the expected biogeographic consequences of sea-level rise in the Philippines?”
3. Methods
After obtaining regular digital elevation models (DEM) and glacial isostatic adjusted (GIA) raster files of the Philippines, the paleo-geography has been reconstructed through ArcGIS 10.2.2. This allowed Polygons and geometrics to be generated automatically. On the basis of these data, relative area loss and relative distance increase could be gathered and combined to form an Area-Distance-Change index. This index was personally created and indicates which islands have experienced the most severe geographical changes after post-LGM sea-level rise. Subsequently, these results are laid out against an extensive literature study on Philippine mammalian species richness and endemism.
a. Data
Erik Koene has provided 22 .tiff-raster files containing GIA paleo-topography of the Philippines for every 1000 years from 21 thousand years ago until now. They are WGS1984 projected and have a spatial resolution of approximately 1 km. This research has only incorporated present and -21ky raster files (Figure 6 & 7). For an isostasy and eustasy comparison, a digital elevation model from the Marine Geoscience Data System was collected, containing the topography and bathymetry of the Philippines (figure 8) (Ryan et al. 2009). It has a spatial resolution of 238 meters.
Figure 6: Raster file of Philippine paleo-topography – present (Koene 2017).
Figure 8: Digital elevation model of the Philippines (Ryan et al. 2009).
b. GIS-based paleo-geographic reconstruction
The entire research has been accomplished by using ArcGIS 10.2.2. The imported GIA raster files, showing a binominal land-sea distinction, were used to create island polygons. In order to create a land-sea distinction with the DEM, the raster calculator tool was applied which assigns a binominal value to land (>=0) and sea (<0) for present, and land (>=-120) and sea (<-120) for 21ky ago. Subsequently, the ‘raster to polygon’ function generated an excessive amount of polygons, which had to be reduced before analysis. For instance, some polygons had to be merged with one another because corner interaction does not automatically result into a unified polygon. Moreover, islands smaller than 1km2 were removed from the attribute table, because these are smaller than the spatial resolution. In order to calculate the geometrics needed for paleo-reconstruction, both layers (GIA and DEM) first had to be projected on the World Mercator coordinate system.
Fragmented islands that had a common ‘insular ancestor’ were then manually exported as individual layers, while true islands (islands that did not fragment, but solely shrunk) and paleo-peninsulas (previously connected to mainland) were exported altogether into two separate layers. The layers’ attribute tables received extra fields with the polygon’s past and present area, in order to return the relative area loss per island type. Relative distance increase to mainland (Borneo) was calculated by using Euclidean distance from the mainland, where minimal distances were acquired through the zonal
After the relative area loss and relative distance increase per island had been calculated, an Area-Distance-Change (ADC) index was personally created (table 2). Relative distance increase is based on 33% and 66% tercile values, while relative area loss is defined by three natural breaks (Jenks). This has been done because all tercile value thresholds would occur above 95% relative area loss. This led to a non-informative overview, as all large islands would be classified the same.
Table 2: Qualitative typology of islands based on relative area loss and relative distance increase to mainland. Values of relative distance increase are based on 33% and 66% tercile and relative area loss is based on 3 natural breaks (Jenks).
4.
Results
Below, there will initially be an overview of the overall effects of sea-level rise in the Philippine archipelago. It will contain general information of the paleo-geographic reconstruction and a display of mean geographical changes per island type. Thereafter, the geographical changes related to sea-level rise, area loss and distance increase, will be presented.
a. Paleo-geographic reconstruction
After removing the polygons (islands) that did not meet the preconditions for analysis, the archipelago consisted of 425 islands at present time. Since the LGM almost 40% of total insular area has been submerged due to sea-level rise. Although the years between 21ky ago and present have not been reconstructed, several authors claim that the bulk of post-LGM sea-level rise has happened between 16.5 and 8 thousand years ago (figure 9) (Lambeck et al. 2014). Accompanied by this was the fragmentation of paleo-islands into numerous smaller ones: 425 islands were originally just 188 islands. This represents an island increase of 126%.
Figure 9: Sea-level data for the past 35k years including 2σ error estimates (Lambeck et al. 2014).
Island types and their particular vulnerability to sea-level rise are indicated by mean geographical characteristics and changes (table 3). Fragmented islands appear to be most abundant and have the largest surfaces (table 3 & figure 10). However, they have experienced the most intense area loss as well (figure 11). This is rather logical as these islands used to be aggregated to one another and therefore automatically lose a lot of area (land does not solely have to be flooded for area loss). This also explains the fact that fragmented islands have the largest relative distance increase: only a minor sea strait has to be created for an island to turn from near located to far away (figure 12).
Area Distance to mainland Type Present (km2) Loss (%) Present (km) Increase (%)
True island (n = 42) 107 (149) 52 (18) 859 (193) 23 (7) Fragmented island (n = 359) 883 (8138) 98 (9) 607 (336) 3963 (7133) Paleo-peninsula (n = 17)* 41 (102) - 18 (24) -
Table 3. Mean geographical characteristics and changes including standard deviation per island type.
* = Area loss and distance increase are not calculated as Borneo has not been included in the analysis. Furthermore, paleo-peninsulas were previously part of the mainland, so the area loss and distance increase would be disproportionately high.
Figure 10: Mean size and mean distance to mainland for true and fragmented islands.
Figure 11 & 12 (respectively left and right): Mean relative area loss and mean relative distance increase for true and fragmented islands.
0 200 400 600 800 1000 Size (km2) Distance (km) True Island Fragmented island 0 50 100 150 Area Loss (%) True Island Fragmented island 0 1000 2000 3000 4000 5000 Distance increase (%) True Island* Fragmented island
Figure 13: Map of Philippine island typology for 425 islands including names of major island groups.
Furthermore, the glacial isostatic adjusted (GIA) raster files from Erik Koene have been compared with the digital elevation model (DEM) in order to test the significance of isostasy. There were only minor differences at shorelines with some extra pixels here and there. In terms of difference, the GIA amount of land 21ky ago was just 0.23% larger than the regular DEM.
b. Area loss
For visualisation, islands’ relative area loss has been assigned to 5 classes (figure 14). Fragment islands like Mindoro and all true islands have been affected the least. Mindanao and Luzon, which during the LGM were connected to each other and formed a very large island, experienced moderate area loss. Other large island groups such as Palawan, Sulu and the Visayas suffered major area loss due to retreating shorelines and fragmentation.
c. Distance increase
Fragmentation and area loss led to a wide variety among island in terms of distance increase from the mainland (figure 15). Palawan and several Paleo-Luzon islands have experienced the most intense consequences of sea-level rise. Luzon and the Eastern Visayas have been moderately changed, while Mindoro, the Western Visayas, Mindanao, Sulu and the Babuyans have suffered the least. Worth noticing are these Babuyan islands in the north which are very far located, but have been stable throughout the glacial shift nonetheless.
d. Area-Distance-Change Index
Each island has been given an ADC index ranging from 1 to 9. While table 2 shows 9 distinct indices, the islands only met the conditions of 6 indices. The map below provides a visual overview of the least or highest affected islands in the Philippine archipelago after post-LGM sea-level rise (figure 16). In line with the previous paragraphs, the Palawan complex and several Paleo-Luzon islands show the most dramatic change. Other high index-islands are the Eastern Visayas, the Sulu complex, and the Western Visayas. Luzon and Mindanao have changed moderately, while the Babuyans, all true islands and Mindoro received low indices.
5.
Discussion
a. Area loss and fragmentation
As the paleo-geography reconstruction has revealed, the Philippine archipelago has been dramatically altered due to post-LGM sea-level rise. Almost 40% of total insular area was lost, while the majority of islands lost more than 75%. The difference between these numbers can be ascribed to fragmented paleo-islands, which led to more than a doubling of Philippine islands: 188 to 425 (larger than 1 km2).
Fragmented islands with major area loss should show higher extinction rates and lower immigration rates. However, high endemism is expected on these islands as well, because speciation is spurred by adaptive radiation during the creation of new environmental niches (Simaiakis et al. 2017; Weigelt et al. 2016). Such a process could result in many species being more resilient to geological changes, because endemics are probably better adapted to insular dynamism than migratory continental species (Rijsdijk et al. 2014; Simaiakis et al. 2017). Hence, it is predicted that on highly fragmented islands especially non-endemic species experience the highest extinction rates due to area loss. This could reveal regional patterns as fragmented islands that were aggregated 21 thousand years ago are more likely to share species than island groups that were never connected to each other (Fernández-Palacios et al. 2015).
The most affected islands (>75% area loss) are three Paleo-Luzon islands, the Visayas, the Sulu complex and the Palawan complex (figure 14). Because of this and their intense fragmentation, it is expected that these islands show higher extinction rates since the LGM, declined immigration rates (target-area effect), while also higher speciation rates. Although these parameters would lower the overall species equilibrium, the equilibrium does not have be reached yet (which could indicate an extinction debt). Luzon and Mindanao have reduced moderately (50%-75%), which should also have resulted in lower immigration rates / higher extinction rates (although less extreme). Mindoro and the true islands are relatively unaltered (<50%), and so should their relaxation dynamics.
b. Distance increase and fragmentation
According to the dynamic equilibrium theory, islands with a high relative distance increase are expected to have a lowered species equilibrium. This is because these islands will be more difficult to reach during passive or active colonization, while the extinction rate goes up as well due to a diminished rescue-effect. A rescue-effect means that immigrants reinforce extinction-prone populations (Fernández-Palacios et al. 2015). Of course, both factors are rather species-dependent as aviary or wind-dispersed species are affected less (Simaiakis et al. 2017). The reason why relative distance increase is applied instead of an absolute distance increase is because it is assumed that small distance changes on far-located islands will have a minimal effect on the immigration rate of many native species (Simaiakis et al. 2017). For instance, the Babuyan islands are most remote, but they hardly changed after the end of the LGM (figure 15). Hence, species richness will have probably been quite stable since then. The Palawan complex on the other hand once had a very high proximity, but sea-level rise resulted in a gap of approximately 150 km. This obviously had an enormous effect on immigration and extinction.
Although lower species richness is expected on far-located islands, increased isolation (just as fragmentation) stimulates speciation rates. This is because gene flows will decrease and gene pools will be split up during fragmentation (Qian & Ricklefs 2000; Heaney 2000). Moreover, endemics located on islands with frequent isolation-fragmentation cycles, will probably be less vulnerable for distance increase than native species.
Besides the Palawan complex, islands that have experienced relatively major distance increase are Luzon, the Eastern Visayas and the Sulu complex (figure 15). The first two used to be merged with Mindanao to form a gigantic island. However due to extensive land submergence in the Eastern Visayas region, these islands ended up disconnected and became more isolated. The Sulu complex saw a significant imbalance of impact: the western island kept the same proximity while the eastern islands’ distance grew by more than 142%. Mindoro, the Babuyans and the Western Visayas (although the latter suffered strong fragmentation) kept quite the same distance to the mainland.
c. Biogeographic comparison
The Area-Distance-Change index is used to provide a clear understanding of the most affected islands. It is a combination of relative area loss and relative distance increase. Islands with the highest index are expected to have the highest extinction rate and lowest immigration rate. Secondly, their relaxation dynamics will be out of equilibrium and thus should have a large extinction debt. Lastly, species richness is assumed to be proportionally lower on these islands, while the proportion of endemics should be higher. Logically, all if these predictions are the opposite for islands with the lowest ADC-index.
Through the past decades, Philippine species richness has been extensively studied by several biogeographers such as Lawrence Heaney. Their accounts of mammalian diversity and endemism will roughly demonstrate whether post-LGM sea-level rise resulted in the expected biogeographic consequences mentioned earlier.
Palawan
The Palawan complex received the highest ADC-index of 9, however the number of native (and endemic) mammalian species is still extremely high: 58 natives, with 54% of non-flying species endemic (Esselstyn et al. 2004). Fragmentation and increased isolation can probably explain the high degree of mammalian endemism, especially since Palawan used to be extremely close to the mainland. Nonetheless, lower species richness was expected regarding the extreme area loss. This could indicate that a major extinction debt is luring in the background.
Eastern Visayas
This region located between Luzon and Mindanao greatly shrunk in area, while the islands became more isolated as well (ADC 8). Heaney (1986) documented 10 native terrestrial mammals in Samar, 14 in Leyte and 17 in Dinagat (could be an indication that distance matters as Leyte and Dinigat are currently closer to the mainland/Mindanao). Although these numbers are relatively low and thus match the expectations, these islands are not considered as species-poor compared to similar sized islands (Heaney & Rabot 1982).
Western Visayas
With an index of 7, this island group experienced major shifts in area / isolation, while sea-level rise disintegrated the region quite dramatically. The island Negros has 13 non-volant mammalian
endemism was expected, species richness is in line with the expectations. Interestingly, Diesmos et al. (2002) has refuted the assumption that fragmented islands often share species (in this case herpetofauna, as mammalian diversity on other Visayas is poorly known): Negros (4), Panay (4) and Cebu (2) all have species that do not occur on other Western Visayas.
Mindanao
The second largest island in the Philippines has moderately changed (ADC 5) during the glacial shift. With 30 native species, of whom 9 are endemic, Mindanao has a rather high species richness (Heaney 1986). Although it is much lower than Palawan’s (especially if the difference in size is taken into account). The relatively low proportion of endemism (36%) does fall in line with the expectations, as Mindanao hardly suffered extreme fragmentation.
Luzon
The biodiversity of Luzon has been extensively studied, and is relatively up to date in terms of documentation. Despite being the largest Philippine island, Luzon still had some significant area loss and distance increase (ADC 5). Heaney et al. (2016) show a species richness of 56 native mammals, with an endemism of 76%. It is considered as a species-rich island, but in terms of species per area it still runs short of Palawan. The exceptional high proportion of endemism may be the reason why Luzon hardly shows a deprived biodiversity, since the extant endemic species are perhaps better adapted to habitat shrinking and isolation. Moreover, speciation could have replaced immigration as the main contributor to mammalian diversity (Heaney et al. 2016). Speciation on Luzon is believed to be an outcome of lower gene flows, enclosed habitat patches and diverse environmental niches because of elevation gradients (Balete et al. 2011).
Mindoro
By having the lowest ADC-index (ADC 1), Mindoro should show stable relaxation dynamics, a low extinction debt, a proportionally flourishing species diversity and low endemism. However, Heaney (1986) documented just 17 native mammals of which 6 are endemic (43%). This relative low species richness but high proportion of endemism parallels the geological age of Mindoro (Heaney 1986). Mindoro has joined the Philippine archipelago along with the Palawan complex over a timespan of more than 10 million years ago (Concepcion et al. 2012). He states that due to the long-term isolation of Mindoro, colonization has been very rare. Moreover, overall high elevation on the island may result in a poor species equilibrium because of elevation gradients in area and climate, montane isolation and feedback among zonal communities (rescue effect) (Lomolino 2001).
It appears that current documentation of Philippine mammalian biodiversity poorly matches the expected biogeographic patterns. For instance, high ADC islands show no deprived biodiversity while low ADC islands show poor species richness. This could mean numerous things: first, lack of evidence of extinctions does not have to mean that there is a lack of extinctions (fossils could still be hiding underneath the soil / sea level). Second, species prone to geological change have already disappeared a long time ago. And last but not least, area loss and distance increase are not responsible for higher extinction rates.
d. Isostatic adjustment and improvements
Philippines are located at a low latitude (5° – 20°), the influences of decompression, compression and glacial gravitational pull are less significant in this area. Also, mainly continental regions are prone to glacial isostasy as ice sheets grow more extensive on land masses (Peltier 1998). Other possible reasons could be differentiating properties of the mantle underneath the Philippine crust, like high viscosity, and changes in Earth’s centrifugal potential due to its variable rotation (Mitrovica 2001).
To improve this type of study and move towards a general glacial sensitive model of island biogeography, several features have to be included in the analysis (Fernández-Palacios et al. 2015; Valente et al. 2014; Lieberman 2005):
- Shifts in vertical temperature and precipitation gradients during glacial transitions will provide insights of zonal ecosystems, species distribution ranges and summit ecosystems.
- Alternating marine currents and wind patterns could explain passive colonization patterns (dispersal) and marine species distribution.
- Inter-island distance increase, because many species use other islands as stepping stones during colonization instead of traversing large bodies of water.
- Rates of geographical change, because some species are more vulnerable to high rates of change (less time to adapt).
- Intra-island connectivity will highlight the capability of species to distribute themselves on an island, since mountain ranges / rivers could hamper colonization.
- Island ontogeny dealing with sea mount emergence, island building and erosion show long-term area changes, elevation shifts, topographical complexity and habitat diversity.
- Tectonic movement of islands, as this will result in vicariance effects, whereby species are physically split, or geodispersal where barriers for gene flows are removed.
6.
Conclusion
A GIS-based reconstruction of Philippine paleo-geography has provided a clear overview of the consequences of post-LGM sea-level rise in the archipelago. The region experienced major area loss, fragmentation and distance increase from the mainland due to retreating shorelines. Especially fragmented islands have been affected the most (compared to true islands): an average of 98% area loss and a mean distance increase to mainland of 3963%. The island groups that saw the largest proportion of land being submerged or broken off were the Palawan complex, Sulu complex, the Western and Eastern Visayas and several Paleo-Luzon islands (>75%). Luzon and Mindanao have moderately changed (<75%), while Mindoro was left most untouched after rising sea-levels (<50%). Outliers in terms of relative distance increase were the Palawan complex (>473% and up to 21763%) and also several Paleo-Luzon islands (>473%). Luzon, The Eastern Visayas and Sulu changed moderately. Mindanao, Mindoro and the Western Visayas relatively maintained their proximity to Borneo. These geographical changes combined resulted in an ADC-index which ordered the island groups from high to low change: Palawan > Eastern Visayas > Sulu > Western Visayas > Luzon > Mindanao > Babuyans > Mindoro > all true islands.
Expected was that documentation of species richness and endemism on these islands would match the expected biogeographic consequences of sea-level rise. So high ADC islands should show high extinction rates, lower immigration rates, a high extinction debt, a higher degree of endemism and a proportionally lower species richness. However, it turns out that not all (documented) islands meet these expectations. This could be due to unrevealed extinctions, species resilience to geological change or even that the area-distance hypothesis is false. Future insular biogeographic research will hopefully expose the reasons for these (mis)matches by integrating temperature and precipitation shifts, changes in ocean and wind currents, inter-island distance increase, rates of geological change, intra-island connectivity, island ontogeny and tectonic movements.
At last, the paleo-geography of the Philippines was glacial isostatic adjusted, but this did not result into significant improvements of the reconstruction (only 0.23% of extra landmass appeared). This can probably be ascribed to the low-latitude location of the Philippines, mantle properties or Earth’s centrifugal potential.
7.
Bibliography
Balete, D. S., Alviola, P. A., Duya, M. R. M., Duya, M. V., Heaney, L. R., & Rickart, E. A. (2011). Chapter 3: The Mammals of the Mingan Mountains, Luzon: Evidence for a New Center of Mammalian Endemism. Fieldiana Life and Earth Sciences, 75-87.
Berger, A. (1980). The Milankovitch astronomical theory of paleoclimates: a modern review. Vistas
in Astronomy, 24, 103-122.
British Geological Survey. (2017). Sea level and coastal changes. Retrieved June 22, 2017, from http://www.bgs.ac.uk/discoveringGeology/climateChange/general/coastal.html
Broecker, W., & Denton, G. (1990). What drives glacial cycles?. Scientific American, 262(1), 43-50.
Brown, R. M., Siler, C. D., Oliveros, C. H., Esselstyn, J. A., Diesmos, A. C., Hosner, P. A., et al.,
& Welton, L. J. (2013). Evolutionary processes of diversification in a model island archipelago. Annual
Review of Ecology, Evolution, and Systematics, 44, 411-435.
Clark, P. U., & Mix, A. C. (2002). Ice sheets and sea level of the Last Glacial Maximum. Quaternary
Science Reviews, 21(1), 1-7.
Concepcion, R. A. B., Dimalanta, C. B., Yumul, G. P., Faustino-Eslava, D. V., Queaño, K. L., Tamayo, R. A., & Imai, A. (2012). Petrography, geochemistry, and tectonics of a rifted fragment of Mainland Asia: evidence from the Lasala Formation, Mindoro Island, Philippines. International Journal
of Earth Sciences, 101(1), 273-290.
Esselstyn, J. A., Widmann, P., & Heaney, L. R. (2004). The mammals of Palawan island, Philippines. Proceedings of the Biological Society of Washington, 117(3), 271-302.
Diesmos, A. C., Brown, R. M., Alcala, A. C., Sison, R. V., Afuang, L. E., & Gee, G. V. (2002). Philippine amphibians and reptiles: an overview of species diversity, biogeography, and conservation. Philippine biodiversity conservation priorities: a second iteration of the National
Biodiversity Strategy and Action Plan, 26-44.
Fernández-Palacios, J.M. (2015). Island biodiversity in past, present, and future [Powerpoint slides]. Retrieved from https://blackboard.ic.uva.nl/bbcswebdav/pid-5877356-dt-content-rid-8500712_1/courses/2310N001.5512ISLA6Y.S15.1.2015/ISLANDS%205%20Island%20biodiversity
%20in%20past%2C%20present%2C%20and%20future%20-%20Jos%C3%A9-Mar%C3%ADa%20Fern%C3%A1ndez-Palacios.pdf
Fernández-Palacios, J. M., Rijsdijk, K. F., Norder, S. J., Otto, R., Nascimento, L., Fernández-Lugo, S., et al., & Whittaker, R. J. (2015). Towards a glacial-sensitive model of island biogeography. Global Ecology and Biogeography.
Fleming, K., Johnston, P., Zwartz, D., Yokoyama, Y., Lambeck, K., & Chappell, J. (1998). Refining the eustatic sea-level curve since the Last Glacial Maximum using far-and intermediate-field sites. Earth
Hays, J. D., Imbrie, J., & Shackleton, N. J. (1976). Variations in the Earth's orbit: pacemaker of the ice ages. American Association for the Advancement of Science.
Heaney, L. R. (1985). Zoogeographic evidence for middle and late Pleistocene land bridges to the Philippine Islands. Modern Quaternary Research in Southeast Asia, 9, 127-144.
Heaney, L. R. (1986). Biogeography of mammals in SE Asia: estimates of rates of colonization, extinction and speciation. Biological journal of the Linnean Society, 28(1-2), 127-165.
Heaney, L. R. (2000). Dynamic disequilibrium: a long-term, large-scale perspective on the equilibrium model of island biogeography. Global Ecology and Biogeography, 9(1), 59-74.
Heaney, L. R., & Rabor, D. S. (1982). Mammals of Dinagat and Siargao Islands, Philippine Heaney, L. R., Balete, D. S., Duya, M. R. M., Duya, M. V., Jansa, S. A., Steppan, S. J., & Rickart, E. A. (2016). Doubling diversity: a cautionary tale of previously unsuspected mammalian diversity on a tropical oceanic island. Frontiers of Biogeography, 8(2).
Imbrie, J., Berger, A., Boyle, E. A., Clemens, S. C., Duffy, A., Howard, W. R., et al., & Mix, A. C. (1993). On the structure and origin of major glaciation cycles 2. The 100,000-year cycle. Paleoceanography, 8(6), 699-735.
Lambeck, K., & Chappell, J. (2001). Sea level change through the last glacial cycle. Science, 292(5517), 679-686.
Lambeck, K., Rouby, H., Purcell, A., Sun, Y., & Sambridge, M. (2014). Sea level and global ice volumes from the Last Glacial Maximum to the Holocene. Proceedings of the National Academy of
Sciences, 111(43), 15296-15303.
Lieberman, B. S. (2005). Geobiology and paleobiogeography: tracking the coevolution of the Earth and its biota. Palaeogeography, Palaeoclimatology, Palaeoecology, 219(1), 23-33.
Loh, E. H., Murray, K. A., Zambrana-Torrelio, C., Hosseini, P. R., Rostal, M. K., Karesh, W. B., & Daszak, P. (2013). Ecological Approaches to Studying Zoonoses.
Lomolino, M. V. (2001). Elevation gradients of species-density: historical and prospective views. Global Ecology and biogeography, 10(1), 3-13.
Lomolino, M. V., & Weiser, M. D. (2001). Towards a more general species-area relationship: diversity on all islands, great and small. Journal of Biogeography, 28(4), 431-445.
MacArthur, R. H., & Wilson, E. O. (1963). An equilibrium theory of insular zoogeography. Evolution, 373-387.
Mörner, N. A. (1976). Eustasy and geoid changes. The Journal of Geology, 84(2), 123-151.
Mitrovica, J. X., Milne, G. A., & Davis, J. L. (2001). Glacial isostatic adjustment on a rotating earth. Geophysical Journal International, 147(3), 562-578.
Peltier, W. R. (1998). Postglacial variations in the level of the sea: Implications for climate dynamics and solid-earth geophysics. Reviews of Geophysics, 36(4), 603-689.
Qian, H., & Ricklefs, R.E. (2000). Large-scale processes and the Asian bias in temperate plant species diversity. Nature, 407, 180-182.
Rijsdijk, K. F., Hengl, T., Norder, S. J., Otto, R., Emerson, B. C., Ávila, S. P., et al., & 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.
Ryan, W.B.F., S.M. Carbotte, J.O. Coplan, S. O'Hara, A. Melkonian, R. Arko, R.A. Weissel, V. Ferrini, A. Goodwillie, F. Nitsche, J. Bonczkowski, and R. Zemsky (2009), Global Multi-Resolution
Topography synthesis, Geochemestry, Geophysics, Geosystems, 10, Q03014,
doi: 10.1029/2008GC002332
Simaiakis, S. M., Rijsdijk, K. F., Koene, E. F., Norder, S. J., Van Boxel, J. H., Stocchi, P., et al., & 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.
Triantis, K.A., Borges, P.A.V., Ladle, R.J., Hortal, J., Cardoso, P., Gaspar, C., Dinis, F., Mendonça, E., Silveira, L.M.A., Gabriel, R., Melo, C., Santos, A.M.C., Amorim, I.R., Ribeiro, S.P., Serrano, A.R.M., Quartau, J.A., Whittaker, R.J., 2010. Extinction debt on
oceanic islands. Ecography 33 (2), 285–294.
Turner, C. S., Tamblyn, A., Dray, R., Maunder, L., & Raines, P. S. (2003). The biodiversity of the Upper Imbang-Caliban Watershed, North Negros Forest Reserve, Negros Occidental, Philippines. Coral Cay Conservation, London, 80.
Valente, L. M., Etienne, R. S., & Phillimore, A. B. (2014). The effects of island ontogeny on species diversity and phylogeny. Proceedings of the Royal Society of London B: Biological
Sciences, 281(1784), 20133227.
Warren, B. H., Simberloff, D., Ricklefs, R. E., Aguilée, R., Condamine, F. L., Gravel, D., et al., & Conti, E. (2015). Islands as model systems in ecology and evolution: prospects fifty years after MacArthur-Wilson. Ecology Letters, 18(2), 200-217.
Weigelt, P., Steinbauer, M. J., Cabral, J. S., & Kreft, H. (2016). Late Quaternary climate change shapes island biodiversity. Nature, 532(7597), 99-102.
Whittaker, R. J., Triantis, K. A., & Ladle, R. J. (2008). A general dynamic theory of oceanic island biogeography. Journal of Biogeography, 35(6), 977-994.
