Environmental Biology - Chapter Two PDF

40 PHYSICAL ENVIRONMENT Virtual view of the physical form of southern Guam. This computer-generated 2.5D perspective of Cocos Lagoon and southern Guam was created using ArcGIS’s 3D Analyst Extension by draping a 2004 IKONOS satellite image mosaic over a 2007 LIDAR-d...

40 PHYSICAL ENVIRONMENT Virtual view of the physical form of southern Guam. This computer-generated 2.5D perspective of Cocos Lagoon and southern Guam was created using ArcGIS’s 3D Analyst Extension by draping a 2004 IKONOS satellite image mosaic over a 2007 LIDAR-derived digital elevation model. (Graphic by Bureau of Statistics and Plans; IKONOS imagery copyright Space Imaging LLC, used with permission.) eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 41 THE PHYSICAL ENVIRONMENT The pattern of islands that Darwin observed and used to explain atoll formation (Sec. 1.2) explains the geology of many tropical Pacific islands, but some have more complicated histories, mostly from having been raised up again after coral growth put a cap of limestone on top. This applies to all the southern Mariana Islands from Guam to Saipan, and to the Rock Islands of Palau. Still other islands consist of crustal rocks; these include the main island group of Yap and the main islands of Fiji. The land plant and animal communities of the islands are influenced by the topography of the land, the composition of the rock, and the geological history of the island. The physical environment of island organisms includes the climate (long-term conditions) and weather (day-to-day changes in atmospheric conditions). Among the realities of the physical conditions on many islands in Oceania are the changes wrought by ENSO (El Niño) and typhoons. These physical conditions are the subject of this chapter. eBook for classroom use only. © 2014 Bess Press Inc.  42 PHYSICAL ENVIRONMENT 2.1 Geology and geography Oceanic islands include a diverse array of landforms. From volcanically active peaks projecting up more than 900 m (nearly 3,000 ft) to low-lying limestone atolls that are barely above sea level, there is tremendous variation in size, elevation, age, and geological composition. Nevertheless, all of the different islands of Oceania have similar origins that can be traced back to processes in the Earth’s crust below the ocean’s surface. 2.1.1 Island formation is driven by the movement of the mantle and crustal plates If one was to look at the model of the Earth that was sliced in half, it would look similar to that of a hard-boiled egg with three main layers: the core, the mantle, and the crust (Fig. 2-1). The innermost layer is the core. It is approximately 3,500 km thick and is divided into two sub layers, inner core and outer core. The inner core is solid and composed mostly of iron and nickel surrounded by a liquid outer core. The mantle is approximately 2,900 km thick and it, too, has an inner fluid part and a solid outer layer, directly under the crust. The core temperature of the Earth is approximately 5,700 K (roughly 9,800 °F), only slightly lower than the surface of the sun. The mantle varies in viscosity, but is generally extremely hot and fluid near the Earth’s core and more viscous and elastic closer to Earth’s surface. It Figure 2-1. Cut-away diagram of is generally believed that the mantle near the Earth’s core rises and expands as it becomes Earth showing the mantle and heated and then descends after it cools. The temperature-driven rising and sinking of the core regions. mantle forms convection currents similar to those seen in water and air. The solid outer layer of the planet is Earth’s crust, but unlike an eggshell, it is fragmented into 14 major plates (Fig. 2-2a). This general map, like all big pictures, conceals a great deal of detail (Fig. 2-2b), which can be essential for understanding local geology. (Remember that maps are models—Sec 1.4.3.) These plates rest on top of the mantle and are buoyed along in part by the movements of the mantle below and by more complex forces related to the thickness and density of the plate itself. As a result, these plates are in constant motion, moving against, away from, or sliding past one another in a process called plate tectonics. These plates vary in density depending on whether the crust is oceanic or continental in origin. Oceanic crust is actually denser than continental crust and contains a greater average concentration of heavier chemical elements such as iron than continental crust. Oceanic crust is also thinner than continental crust and typically much younger (Table 2-1). New oceanic crust forms at divergent boundaries underwater where two plates are moving away from each other along the mid-ocean ridges in a process called seafloor spreading (Fig. 2-3; shown as red lines in Fig. 2-2b). As the two plates spread apart, fresh eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 43 (a) Figure 2-2. Tectonic plates. (a) Map of the world showing the major tectonic plates. Diverging boundaries (mid-ocean ridges) and convergent boundaries (subduction zones) indicated by arrow directions. (b) Detail of the plate boundaries in the Western Pacific, also showing the directions and speed of travel of each fragment. ([a] USGS, public domain; arrows added. [b] Modified from a map by Eric Gaba on Wikipedia, derived from data on a wall map by Peter Bird, used with permission of Eric Gaba.) (b) eBook for classroom use only. © 2014 Bess Press Inc.  44 PHYSICAL ENVIRONMENT Table 2-1. General characteristics of oceanic and continental crust. magma emerges from the mantle and fills in the space. The new crust Oceanic Crust Continental Crust that forms in these spreading centers then cools as it slowly moves away Crust Thickness [Plate from the axis of spread, along with 7 km [70 km] 30–50 km [125 km] the rest of the plate it is attached to. Lithosphere Thickness ] Thus as the plate migrates away from Density of Rock 3.0 g/cm3 2.7 g/cm3 the spreading center, new crust is continually added to its trailing edge. Probable Composition Basalt Granite Meanwhile, some parts of the plate Age (oldest) ~190 million years 3.964 billion years form convergent boundaries with other plates. At these boundaries (indicated on Figure 2-2a by arrows pointing toward the boundary and on Figure 2-2b by purple lines), one of the plates may plunge down towards the mantle in a process known as subduction (Fig. 2.2b blue lines with triangles; Fig. 2-3). The other plate may override the subducting plate, experiencing uplift in the process. Subduction zones are characterized by the presence of a trench, which is a submarine depression on the sea floor where the two plates meet. Trenches may be thousands of meters deep (Fig. 9-16); the deepest point on Earth (10,900 m, 35,760 ft) is in the Mariana Trench southwest of Guam (Sec. 9.4). Thus while huge areas of the seabed are relatively featureless abyssal plains, they also feature dramatic ridges thousands of miles long, numerous extremely deep trenches, and many island arcs comprising volcanic mountains that rise 4,000–6,000 m before they even break the water’s surface. The Big Island of Hawai‘i is taller from seabed to peak than Mt. Everest is from sea level to peak. The intense pressure and friction between two plates at a subduction zone is great enough to melt crust, especially when seawater in the rocks lowers the melting point at the interface. This friction also produces many deep earthquakes, indicated by the asterisks in Figure 2-3. The melted rock heats and expands, working its way through cracks and spaces in the overriding crust as magma until it emerges underwater (left side of Fig. 2-3) or comes up through the continental crust (right side of Fig. 2-3). The molten rock cools upon contact with the seawater, eventually solidifying. As more mantle seeps through the crust and builds up in the overriding plate, it eventually forms a volcano. As more eruptions build the cone, the volcano may eventually protrude above the ocean’s surface, at which point it is considered an island. Subduction zones tend to form island chains that run parallel to the trenches on the overriding plate. The Mariana Islands are an excellent example of an island arc formed through subduction volcanism. Hotspot volcanism is another type of volcanic activity associated with Pacific island formation (Fig. 2-3). In contrast to subduction zones, hotspots are not necessarily associated with trench systems. Rather, hotspots are focused areas of volcanic activity that are believed to be the result of superheated mantle plumes from deep within the earth. The magma melts through the overlying plate. As the magma emerges from the hotspot, it forms an underwater volcano and may eventually build an emergent island. These magma eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 45 plumes can be relatively stationary, even as the plate moves over the hotspot. As a result, Figure 2-3. Block diagram island chains can also form in association with hotspots, with older islands being carried of Earth showing the major away from the hotspot and newer islands continually forming. The Hawaiian Island chain, features of plate tectonics. including the Emperor Seamounts, is the result of the Pacific Plate moving over a very long- In the middle, the mid-ocean lived stationary hotspot that is already forming a new mountain southeast of Hawai‘i Island ridges are formed by magma while volcanoes on Hawai‘i are still active. floating up from the core. At the right and left, the oceanic crust is subducted under the 2.1.2 Land formations change over geological time continental crust of another oceanic plate. Friction causes Earth has a very long history recorded in its rocks (the oldest known rocks are 4.0 billion earthquakes (asterisks) and as years old). The continents have not always been in the same places on the planet, and the the crust melts, magma rises to coastlines have varied a great deal. Continents, like the islands we just discussed, also ride the surface forming volcanoes on on crustal plates, and their movements have altered their climates and ecosystems over land or as island arcs. (Modified hundreds of millions of years. We will return to ecosystem changes in Chapter 10, but here from graphic by Jose Vigil, USGS; we must introduce the geological time scale and mention geological changes in sea level. public domain.) Geologic time is a classification system (Sec. 4.3.1) in which time from the beginning of the world is divided first into eons and then into eras, but these are so big that biologists rarely have cause to speak of them. Commonly used levels are periods and epochs (Fig. 2-4). For example, the more recent epochs, during which the Mariana Island chain has formed, are part of the Cenozoic Era—that is, since the end of the dinosaurs (and much other life; see Sec. 10.4.5), at 65.5 Ma (Mega-annum, or million years ago). Cenozoic means “new/ recent animals” and signifies that in this period birds and mammals replaced dinosaurs as the dominant land animals. There are two periods in the Cenozoic, the Tertiary and the Quaternary, and within these a series of epochs whose names all end in –ocene (from the same Greek root as Ceno-, recent). These epochs start with the Paleocene (“early recent”) eBook for classroom use only. © 2014 Bess Press Inc.  46 PHYSICAL ENVIRONMENT Figure 2-4. Geologic time and end in the current epoch, the Holocene (“wholly/really scale. This chart shows only the recent”) (Fig. 2-4). The Quaternary Period, characterized Phanerozoic Eon, which began by a series of ice ages and the appearance of anatomically at 542 Ma (million years ago), modern humans, is divided in turn into two periods, with the earliest rocks containing Pleistocene and Holocene. The volcano that is now Guam recognizable animal fossils in the goes back at least 44 million years (that is the age of the Cambrian Period (Phanerozoic oldest known rocks), in the mid-Eocene (“dawn of the new”) means “evident animals”). Before Epoch. Just as biologists continue to debate and refine that (not shown), time extends classification of organisms, geologists continue to debate back through the Proterozic to and refine the boundaries of the time-scale categories; times before life. The informal for instance, in 2009 the International Union of Geological name Precambrian covers all the Sciences redefined the base of the Quaternary and the base time before the Cambrian. Notice of the Holocene. These new boundaries are shown on the that time is not shown to scale, U.S. Geological Survey chart in Figure 2-4.1 but becomes progressively more compressed toward older ages. The span of geologic time is very hard to grasp. One well- (Reprinted from USGS 2010, worn comparison is that if the age of the Earth were public domain.) represented by a 24-hour clock, human beings would have appeared only about 1.5 minutes before midnight. Another way to visualize it is to mark out a line on the edge of a football field, in which each meter represents one million years. At this scale, the first settlement of the Pacific islands would be a mere 3.5 millimeters from the close side (about the width of this capital M); the origin of modern man would be at about the 10 centimeter (4 in) mark, and the end of the dinosaurs would be 65 m (211 ft) away at the other end of the field. Geologists are able to work out relative ages of rocks by looking at similar strata (layers) at different places, and can determine chronological ages of the rocks by studying traces of radioisotopes in the rocks that decay (change) in known ways over long periods of time. Both volcanic rock (also called igneous rock) and sedimentary rocks are deposited in layers, the former as a result of different lava flows or other eruption events, the latter the result of different erosion processes taking place, with particles of varying sizes and crystal compositions. Limestone from reefs is also sedimentary rock, since the actively growing parts are continually broken off and the remains of the corals and other calcareous animals and plants then accumulate as sediments on the reef flat and in deeper water below the reef (Fig. 9-1). However, the analysis of rock layers is complicated by the process of erosion: When rocks are exposed to rain and wind, layers are worn away, and eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 47 Figure 2-5. Sea level changes since the mid-Pleistocene. (a) Longer term changes showing sea levels well above present level as well as below. (b) Changes since the peak of the last ice age. The Holocene Epoch is now defined as starting 11,700 ± 99 years before 2000 CE. eBook for classroom use only. © 2014 Bess Press Inc. 48 PHYSICAL ENVIRONMENT some of the geologic history disappears. If new rocks are later laid down on top, there is a discontinuity between the older, eroded layers and the new layers. Sea levels change over geologic time as a result of several factors, including the relative heights of land masses as the plates jostle about. The most familiar changes in sea level are those associated with ice ages that have taken place during the past 130,000 years (Fig. 2-5a). Changing sea levels changed the shapes of islands, sometimes radically (Fig. 6-13). These lowered the sea level about 100 m (300 ft) three times,2 and sea level stabilized at its present level only about 6,000 years ago (Fig. 2-5b). There is evidence that sea levels earlier in the Pleistocene Epoch were as much as 100 m higher than today, which would have left Guam largely underwater if there were no rise of the crustal plate. 2.1.3 The southern Mariana Islands and some other Pacific Islands have complex geological history In Chapter 1, we described the formation of high islands in Micronesia, where volcanoes rose above the ocean surface to become islands before eroding and sinking to become atolls (Sec. 1.2). However, the volcano that became Guam apparently did not rise above the waves, but was an undersea mountain while volcanic activity lasted, and was only uplifted after corals had built on top of it. Guam’s geological history began approximately 20° farther south of its current location and possibly further west, given its origins along the Palau-Kyushu Ridge, which runs up the middle of the Philippine Plate. Guam’s geological history is tied intimately to the movements of the large Pacific Plate and the smaller Philippine Plate on its western flank. Prior to Guam’s creation, the Pacific Plate was sliding past the Philippine Plate in a northerly direction (a transform plate boundary: Figs 2-2a, b). However, in the mid-Eocene Epoch, sometime before 43 Ma, as the Indian subcontinent began to collide with Eurasia, the Pacific Plate began moving north-northwest, colliding with and subducting under the smaller Philippine Plate. As the Pacific plate slid under the Philippine Plate and plunged into the mantle, friction between the two plates generated heat that created molten rock (magma), below Earth’s crust. Under the tremendous heat and pressures between the two plates, the magma began to expand and rise up toward the overlying Philippine Plate. Eventually, the molten rock seeped through cracks and fissures along the eastern edge of the Philippine Plate, emerging as lava on the sea floor deep below the surface of the Pacific Ocean. The lava from this initial episode slowly formed the underwater volcano that became the basement volcanic bedrock of Guam. The rocks from this event belong to the Facpi formation and are the oldest surface volcanics on Guam, but there are older rocks out of reach below them, at the base of the volcano. Much of the volcanism associated with the Facpi formation was submarine lava flows in the form of pillow basalt, the remnants of which are still visible in the southern village of Umatac. eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 49 The Facpi formation formed over a period of about 3–4 million years from 43 to 39 Ma. This was followed by a transition to a more volatile type of volcanism associated with spectacular eruptions casting off large amounts of ash and other volcanic debris, still underwater. These pyroclastic eruptions lasted from 32 to 29 Ma and formed the second major phase of volcanic buildup, called the Alutom formation. These volcanic rocks included substantial deposits of sandstone, mudstone, and mixtures of larger rocks embedded within ash and dust. The end of the development of the Alutom formation coincided with a long, quiescent period from 29 to 20 Ma. Interestingly, it was during this period that Guam split from its location along the Palau-Kyushu Ridge and migrated farther east in the direction of the Pacific Plate. By about 20 Ma, however, Guam began the third and final episode of major volcanic construction along what is now known as the West Mariana Ridge, approximately 120 miles west of Guam’s present location (the divergent boundary labeled in Figure 2-2b). This final volcanic episode resulted in the Umatac formation, which forms the bulk of the volcanics exposed in southern Guam. The Umatac formation comprises three subunits, or members (that is, the Schroeder, Bolanos, and Dandan members; Figs 2-6a, 7-1b) and was largely the result of volatile volcanic eruptions. It is currently believed that the Umatac formation was deposited between 20 Ma and Figure 2-6. Volcanics vs. about 5 Ma. Volcanism then ceased and Guam began rifting away from the West Mariana limestone. (a) Rocks tell a story; Ridge to its present location (a process called back-arc spreading). Throughout the buildup students examine a lava flow of the Facpi and Alutom formations, Guam was still an underwater volcano. The remainder from the Dandan member of the of Guam’s geological history is one of carbonate (limestone) buildup and tectonic uplift Umatac formation. (b) Detail of over the last 5 million years. Barrigada limestone (dark green area on Fig. 2-7) showing the By the end of the major volcanic episodes, the top of the volcano was near the ocean fine-grained, smooth texture surface and there was a period of sedimentary limestone deposition. Then, when back- (contrast with Fig. 2-8b). (Quarry arc spreading started, it separated the huge volcanic mountain into two, and there was shown in Fig. 12-18.) another brief round of volcanism, so that volcanic rocks lie over the earliest limestone. (a) (b) eBook for classroom use only. © 2014 Bess Press Inc.  50 PHYSICAL ENVIRONMENT Figure 2-7. Geologic map of Guam. Showing the predominantly limestone areas (shades of blue, and green in the north and southeast) and volcanic areas (red, purple, and brown in the south, plus two outcrops in the north). Also shown are alluvium (deposits along river valleys) and beach deposits. (Map by John Jocson using USGS data in the public domain.) eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 51 (a) (b) Eventually, volcanism moved north into the other islands (many still active), and Guam Figure 2-8. Mariana Limestone. acquired several layers of framework (coral) limestone. Finally, starting around 1–2 Ma, (a) The cliff at Two Lovers Point, Guam has been lifted up so the ancient corals are now 100-200 m above sea level. As a Guam, and the isolated rock result of this long history, Guam can best be described as a volcanic island for the southern at Tanguissan Beach are fossil half of Guam and a raised limestone island for the northern half (Fig. 2-7). coral reef facies; the hills in the background (Nimitz Hill) are To a geologist, the rocks of an island tell a story (Fig. 2-6a). One way in which geologists part of the volcanic Alutom interpret evidence in the rocks is by using the principle that the events that shape the Formation. (b) Detail of a sand- Earth today are the same as those that shaped it in the past. Thus, studies of the rocks of smoothed rock outcrop on the volcanic eruptions that people have witnessed can be used to understand ancient volcanic beach near Lost Pond, showing rocks. The sequence of layers records the order in which the various deposits were laid recognizable coral skeletons in down. Maps of the seafloor are also vital, and there are many other kinds of evidence that the matrix (most of the limestone geologists use to understand the Earth and make detailed models such as the map in in northern Guam is sharply Figure 2-2b. eroded into karst). As described above, three volcanic formations dominate southern Guam as we see it today: the Facpi, Alutom, and Umatac formations (Fig. 2-7). The Facpi formation consists of alternating strata of pillow lava and breccia (cemented rock fragments) that are found on the southwestern coastal areas of the island from Facpi Point to the village of Merizo. The Alutom volcanic formation overlays part of the Facpi formation dating to the late Eocene– early Oligocene in age. The Alutom formation (brown on Fig. 2-7) occupies the south central portion of the island and underlies the limestone bedrock of northern Guam, where it protrudes through the limestone to form Mt. Santa Rosa and Mataguac Hill. This volcanic formation consists of layers of breccia, sedimentary conglomerates (blocks or stones of different sizes) and sandstone, calcareous shale made of the remains of foraminifera (Sec. 9.1.4) and pyroclastic rocks (volcanic fragments resulting from an explosive eruption) and lava. The Umatac formation is the youngest of Guam’s volcanic formations, dating to the late Oligocene to mid-Miocene Epoch. The Umatac formation consists of several rock types, resulting from sequences of volcanic activity and sedimentation, consisting of lava eBook for classroom use only. © 2014 Bess Press Inc. 52 PHYSICAL ENVIRONMENT flows, sedimentary sandstones, breccias, conglom­erates, and limestone layers. Scattered throughout the southern hills are small remnants of limestone formations that once covered the southern island and have since eroded away leaving only small outcrops. Northern Guam can be described as a gently sloping limestone plateau. Although several limestone formations have been described in northern Guam (Fig. 2-7), this half of the island is dominated by two: the Barrigada Limestone (green on the map) and the Marianas Limestone (various shades of blue on the map). The Barrigada Limestone is described as a fine-grained detrital deep-water limestone made up of foraminifera fossils that settled on the ocean floor during the Miocene-Pliocene Epoch. The fine-grained character can be seen in Figure 2-6b. The Barrigada Limestone is exposed on the surface of Northern Guam in a ring-shaped formation bordered by the younger Mariana Limestone in the middle and along the outer borders of the plateau. The Mariana Limestone overlays much of northern Guam (Fig. 2-8a) and the southeastern coastline of southern Guam from Yona to Talofofo village. The Mariana Limestone is Pliocene-Pleistocene in age (approximately 1–4 million years old), a shallow-water limestone deposited in a coastal reef environment as the island was slowly uplifted due to plate tectonics; fossil corals are evident in the rock (Fig. 2-8b), in contrast to the smooth texture of Barrigada Limestone. The different facies (a facies is a body of rock with specified characteristics) that represent the Mariana Limestone reflect the different reef habitats that can be found on modern reefs today. These include a Reef Facies, Fore-Reef Facies, Detrital Facies, and a Mollusk Facies. There is also a distinct type of Mariana Limestone called the Hagatna Argillaceous Member that borders south-central Guam. This member contains fine-grained clay particles that washed off southern Guam and deposited into the matrix as the limestone formed. 2.1.4 Atoll islands are formed by coral Atoll islands (Fig. 2-9) are built by waves on barrier reefs. This process starts even before the volcanic island has completely sunk, and is why Guam has a coral island on the barrier reef around Cocos Lagoon. Coral sand and rubble pile up in some places, but erode away in others. The geography of each island is a dynamic balance between erosion and deposition. Storms help form and destroy such islands. Bruce Richmond of the U.S. Geological Survey studied how atoll islets form, writing, Atoll islets are some of the youngest and most dynamic of earth landforms. They begin as a product of catastrophic storms and remain precariously balanced between the destructive forces of waves, currents, bioerosion, and chemical dissolution and the constructive effort of reef-produced calcium carbonate sediment.3 He classified islets into four types, described how they form and persist, and considered how they might respond to sea level rise. Briefly, the islet types are the following, as shown in Figure 2-9. Type I islets are sandy and occur along the leeward atoll rims near reef passages. They grow and shrink but remain about the same size and in the same location. However, they are very susceptible to large storms because the sand is usually only lightly cemented together. The coral rubble that builds the other islet types becomes cemented eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 53 Figure 2-9. Atoll islet types, not to scale. Type I is a small, sandy islet on the sheltered side of a lagoon, showing the waves and currents that keep adjusting the size and shape of the islet (based on Tepuka islet, Funafuti). Type II islets are largest, most stable (cemented), and have the best soil and freshwater. Type III islets are mid-sized but tend to be reshaped dramatically during storms. (From B.M. Richmond 1993, reprinted with permission of UOG Marine Laboratory) eBook for classroom use only. © 2014 Bess Press Inc.  54 PHYSICAL ENVIRONMENT together and stabilized (Fig. 2-9). Type II islets (Figs 1-2d, 2-9b) are the ones people most often live on, because they are the largest, most stable, and have the best-developed soil. Over the long term, type II islets tend to keep getting larger, although like all types, these islets can get bigger or smaller depending on typhoon waves. Type II islets are often hook-shaped and form on sharp bends on the atoll rim. Examples include Djarrit and Laura islets on Majuro Atoll and many other permanently inhabited islets in Micronesia. Type III islets (Fig. 2-9c) are long and narrow, often curved. These and Type IV islets (such as Cocos I., Guam; not shown in Fig. 2-9) have complex combinations of characters and are not stable. They tend to grow dramatically during catastrophic storms and gradually erode away until another storm builds them up again. They often have a series of beach ridges Figure 2-10. Climate types in the tracing their history. tropical Pacific, as defined by the Köppen-Geiger classification, according to Thomas 1963. The broad-scale classification is based on temperature and rainfall. Mountains cause additional local 2.2 Climate: Introduction rainfall and rain shadows (e.g., in The second substantial part of the physical environment is climate. Climate is the long- Hawai‘i). The classification codes term pattern in factors such as temperature, rainfall, day length, and their seasonal changes. are as follows: Abiotic factors also include the weather, the constant variations in conditions such as temperature, rain, and wind, together with significant phenomena such as typhoons and El A = Tropical (hot) and wet Niño/Southern Oscillation (ENSO). enough for tall trees Af = rain all year Aw = dry season in winter of the respective hemisphere B = Dry (hot or cold) BS = steppe-like climate (wet enough for grassland) BSh = hot BSk = cold BW = desert BWh = hot BWk = cold (C [warm temperate] and D [snow climates] are not shown on this map.) eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 55 2.2.1 There is geographic variation across the tropical Pacific Rainfall and temperature vary from island to island, and sometimes even across one island. Temperature on tropical islands de­pends mainly on altitude; it is cooler at the tops of mountains. If the mountain is high enough, the top may be covered in snow, like Mauna Kea, Hawai‘i, even though the latitude is tropical. Rainfall patterns are complex on both small and large scales. On a larger scale, some areas of the Pacific are wet and others are dry (Fig. 2-10). The rainfall pattern moves east and west periodically as a result of El Niño, as described in Section 2.3. On a local scale, mountains greatly affect rainfall. On islands with prevailing winds from one direction, the windward side will be wetter, especially at middle altitudes, and the leeward (sheltered) side will be drier (Fig. 3-22). This is because clouds pick up moisture from the warm ocean water. The warmer the air, the more moisture it can carry. When the winds meet the mountains, they must go over or around them. As the air rises, it cools and cannot hold as much water, so that water falls as rain; this is called orographic rainfall. By the time the air gets to the other side of the mountain, it has lost most of its water, so this area, called the rain shadow, gets little rain. If the mountain is very high, the rain may be gone even before the air gets to the top. Thus mountains may differ Figure 2-11. Rain and rivers on Guam. Map showing the in rainfall from top to bottom and from east to west. These distribution of rivers, the variation in rainfall (numbers are inches variations cause differences in vegetation zones (Sec. 3.3.5). per year), and the location of the places mentioned in the text. Even on small, relatively flat islands such as Guam, rainfall is (After Tracey et al. 1964.) not evenly distributed (Fig. 2-11). Pacific island environments are shaped by two sets of factors: 1) the geological nature of the island (high/low; volcanic/carbonate), and 2) the geographic location and It is not only what an island is but where therefore rainfall patterns. These are the driving forces of island ecosystems and ecosystem dynamics; in other words, it is that determines its ecology—and its it is not only what an island is but where it is that determines vulnerability to climate change. its ecology—and its vulnerability to climate change. 2.2.2 There are seasonal and long-term variations in weather Traditional cultures, which relied on the environment for their livelihood, observed the winds, trees, and animals to know when to plant and when to fish for each species (Fig. 2-12).5 People also wanted to know what the weather would be like over the next growing season, or through the next dry season. eBook for classroom use only. © 2014 Bess Press Inc.  56 PHYSICAL ENVIRONMENT The weather is constantly changing. Rain often comes as brief local showers; temperatures change from day to night and from season to season, even in the tropics. Tropical storms are more frequent in some months than others. Some years are wetter, or drier, or stormier. Until recently, most of these fluctuations made little sense. Weather forecasters could look at an existing weather system and predict the local, short-term changes, but long-term Fig. 2-12. The Palauan’s perceived model of cyclic time. Ngermetenglel’s clock and calendar showing the land and sea seasons, with some planting and fishing times. (Reprinted from Klee 1976 with permission of Micronesica.) eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 57 forecasts were hardly “scientific.” In order to predict a long-term pattern, we need to know what patterns exist and what early signs there are to tell us where we are in the pattern. Seasons are the best-known patterns: summer/winter; monsoons; wet season/dry season; breadfruit season; and so forth. In areas where there are big differences between seasons, everybody knows what to expect, more or less, in each season. Even in places where the climate changes are not very strong, seasons exist, and plants and animals respond to them. Beyond these regular variations, extreme weather events challenge plants and animals. Severe storms, floods, and droughts are potentially catastrophic. Droughts have had serious effects on island people. While starvation (described in the legend and the historical accounts below) can now be prevented by shipping food aid, water is more of a problem. Moreover, food aid has only recently become possible, whether for Micro­nesian islands via Federal Emergency Measures Administration (FEMA) (Sec. 11.5), or from the global community to semi-desert countries in Africa. In the past, people were unlikely to know of famines elsewhere or have any means to respond. In the future, if world populations continue to grow, the willingness of countries to send food aid may diminish. There are legends about the origin of breadfruit trees in many Pacific Island cultures, and some of these refer to famines.6 Guam’s legend, reproduced below, specific­ally describes the conse­quences of a severe famine. Some versions of the story mention drought and/ or typhoon as the cause of the famine. There are also actual accounts of famines, such as the one that took place on Kapingamarangi Atoll (see below). Famines were com­mon on Pacific Islands not only because of the weather, but also as a result of destruction of crops and food stores during wars. GUAM’S LEGEND OF THE BREADFRUIT While Guam was still very young, the island was divided into they discovered that the famine was island wide. The children, two kingdoms. One man ruled the southern kingdom and his who were very tired and hungry, were put to bed for the evening brother ruled the northern kingdom. A great famine struck the while the men talked late into the night. island. With food nearly exhausted, each ruler decided to go Arising later than usual the next morning, the men learned to the opposite end of the island to search for food. The ruler that all the children had died in their sleep. In sorrow the two from the south packed his bag with a small portion of food rulers buried their children. After the ceremony, they placed and started towards his brother’s village. He also had his three latte stones at the heads of their children’s graves. In sheer ex­ children and the three strongest men of the village with him. haustion, they collapsed. When the men awoke, they saw a Unknow­ingly and at the same time, his brother, the older ruler, strange tree growing at the head of each grave where the latte packed his bag with a few scraps of food. Accompanied by his stone had been. They ate the fruit of the tree and called it lemai, three children and several strong men, he headed south. or breadfruit. From that day on, there has never been a famine In the middle of the island, the two brothers met. Neither group on the island because of the miracle of the breadfruit tree. 7 of people had eaten for two days. After greeting each other, eBook for classroom use only. © 2014 Bess Press Inc. 58 PHYSICAL ENVIRONMENT The drought on Kapingamarangi led to a large fraction of the Kapinga resettling on Pohnpei as environ­mental refugees. This is the story as told by M. D. Lieber in Exiles and Migrants in Oceania: In 1916, a drought began which was to last for two years and which would culminate in the deaths of over ninety people. As the soil dried up and staple food plants became unproduc­tive, the threat of chaos and panic grew on the atoll. Theft of food became common, sometimes resulting in violence. As food resources dwindled, the inevitability of famine became apparent to all. No Kapinga could remember a drought of these proportions, and people were unprepared for the situation. A Japanese teacher and govern­ment local affairs officer named Huria was living on Kapingamarangi at the time. Having estab­ lished a position of authority on the atoll, he was under pressure by the Kapinga to do something. Working through the atoll chief and a council of men appointed by the chief through his urging, Huria was able to institute a rationing program to conserve drinking coconuts. This was done by controlling movement from the major residential islets to coconut stands on the outer islets. Men of the council, called “masters,” administered punish­ments for violating these regulations. Huria also attempted to limit population growth by placing a ban on premarital sexual relations and by prohibiting many marriages, again by decree of the chief. The chief, by 1917, was a man appointed by Huria to replace the old chief, who was on his deathbed. None of the emergency measures was able to stave off the starvation and death which finally resulted from the prolonged drought, nor were people and plants the only casualties. The ancient religion and its priest­hood collapsed as years of debunking by outsiders, a growing skepticism of some Kapinga, and the obvious inability of the priesthood to alleviate the drought and famine demoralized the population. When a mission­ary from the neighboring atoll of Nukuoro appeared on a visiting ship in 1917, and later returned with gifts of food and offers of salvation from future disasters, the population was converted wholesale to Christianity.8 As growing populations, modern intensive agriculture, and climate change push the limits of where and when to grow, we come again to the importance of understanding the environment. With world food production closely tied to existing climate patterns, we now need to know about global climate patterns. We need to be able to predict and prepare for unusual years. The knowledge of El Niño is a still-unfolding story of how scientists are learning about a major climate pattern. What was once thought to be a local phenomenon affecting fisheries in Perú is now known to involve large weather changes in the Pacific region that extend even as far as East Africa and the east coast of the Americas.9 2.3 El Niño/Southern Oscillation (ENSO) The coast of Perú is a desert. In most years there is hardly any rain at all, but in some years the summers are very wet. Then the deserts turn green; there is more food and more water. The sea is warm and strange creatures appear in it. People refer to these years as times of abundance. Since Christmas in Perú occurs at the beginning of their summer, people associate the good years with the Christ Child, hence the name “El Niño.” But as commercial eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 59 fisheries de­veloped, providing sardines and anchovies to the world market, Peruvian fishers came to know El Niño as years of disaster. The warm water drove the anchovies and sardines out of their waters. Oceano­graphers tried to find out why the Figure 2-13. Climate system changes during El Niño. Diagrams of the Pacific Ocean warm waters sometimes occurred. showing the changes in winds, currents, rain, thermocline, and sea level. (a) The normal and La Niña tradewinds and upwelling of cold water off the coast of South A century ago, halfway round the world, America. Southeast winds along the coast of Perú push warm water out across a British meteorologist named Sir Gilbert the equatorial Pacific, making the thermocline deeper in the western Pacific and Walker was trying to discover the cause of shallower in the eastern Pacific than during El Niño. This water is replaced by cold variations in Asian monsoons from year to upwelling along the Peruvian coast. Rains form over the warm water around the year. Several times over the past decades, western Pacific islands. (b) During El Niño, the easterly tradewinds weaken or even the monsoons had not arrived. The reverse; upwelling weakens and fails to pull up cold, nutrient rich water; and the rains resulting droughts caused major famine in move east. (After Wallace and Vogel 1994.) eBook for classroom use only. © 2014 Bess Press Inc.  60 PHYSICAL ENVIRONMENT Figure 2-14. Effects of El Niño on global rainfall. Stippled areas (inside solid lines) are drier; areas inside dashed lines are wetter. The duration of the drier or wetter weather is related to the El Niño year as defined by meteorologists. For instance, Nov. (0)–May (+) means that the altered weather lasts from November of the official El Niño year until May of the following year. (From Lander 1994, adapted from Ropelewski and Halpert 1987.) India, but attempts to find causes of the drought had failed. Walker, as a mathematician and director-general of observatories in India, realized the need for examining massive amounts of data to look for correlations (rather than causes), and used human computers to crunch the numbers. In 1923, he discovered the Southern Oscillation. Atmospheric pressure is normally high over the southern Pacific Ocean and low over the Indian Ocean. Walker observed that when pressure went up over the Indian Ocean, it went down over the southern Pacific. Winds always blow from high-pressure areas to low-pressure areas. This difference in pressure between the air over the Pacific and Indian Oceans is what causes the tradewinds to blow across the Pacific (Fig. 2-13). Walker saw that when pressure went down over the Pacific Ocean and up over the Indian Ocean, there was less of a difference, and the easterly winds did not blow as hard. He hypothesized that the Southern Oscillation was the cause of the periodic monsoon failures.10 Later, other meteorologists noticed that there was a pattern of rainfall on the desert islands of the central Pacific Ocean, such as Christmas Island (Kiritimati Island in Kiribati), on the equator between Hawai‘i and Tahiti. When the easterly winds are strong (in what we still refer to as “normal” years), there is little or no rain. When the easterlies weaken, the islands experience heavy rains. The pattern began to emerge: droughts in India, Indonesia, Australia; deserts flourishing on Christmas Island and in Perú. But how did the oceans fit into this? Why did the anchovies disappear when the deserts bloomed? The connection between El Niño and the Southern Oscillation was made in the 1960s, after scientists had collected data on tropical ocean temperatures. University of California professor Jacob Bjerknes linked changes in water temperature with Walker’s Southern Oscillation. Because El Niño and the Southern Oscillation are part of the same phenomenon, it is now officially known as El Niño/Southern Oscillation, or ENSO. During “normal” years, the indices fluctuate near zero (Fig. 2-15). Extreme variations are identified as El Niños (warm, wet years), and “La Niña” (unusually cool, dry years). These characterizations as wet or dry of course reflect the view from South America (Fig. 2-14). In the western Pacific Islands, eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 61 Figure 2-15. Southern Oscillation Index 1950–2005. The negative phase of the SOI represents below-normal air pressure at Tahiti and above- normal air pressure at Darwin. Strongly negative values correlate with El Niño events; the strongest ones of the past 30 years are labeled. Positive values correspond to La Niña events. (From NOAA, public domain.) Australia, and India, it is the El Niño years that are dry, because we are on the opposite end of the see-saw. However, the severe El Niño of 1982–83 surprised scientists because it did not fit the pattern of events during the previous three decades. Typhoons, droughts, fires, and floods from that El Niño caused an estimated $8 billion in damages (in dollars of those days).11 The diagram in Figure 2-13 explains what happens during El Niño. When the tradewinds are strong (“normal” and La Niña years), they push water across the tropical Pacific, making sea level about 2 ft higher in the western Pacific than the eastern Pacific. As surface water is blown away from South America, water is pulled up from the ocean depths to replace it. This current of water, called upwelling, is cool and also rich in nutrients. The nutrients support growth of microscopic plants (phytoplankton), which in turn support a food chain leading to anchovies and sardines. During El Niño, the tradewinds weaken in the eastern Pacific. West of the Inter­national Date Line, the winds may even reverse. Warm water is not pushed as far across the ocean, and the upwelling is thereby reduced. More importantly, warm water becomes deeper as the thermocline oscillates along with the sea level (Fig. 2-13). With weaker winds to drive upwelling and a deeper layer of warm water, upwelling along the South American coast does not reach down into the cold, nutrient-rich water. Without deep upwelling, there are no nutrients at the surface, therefore no phyto­plankton and no fish. During such times, the sea level on western Pacific islands drops because the winds are not piling up water on their shores. This exposes the reefs and kills many marine organisms.12 Changes in the wind patterns also cause the area where typhoons form to move east.13 During the El Niño of 1982–83, Hawai‘i and Tahiti both experienced major typhoons, which was unusual. The 1992 typhoon season was memorable in Guam because five passed over or near the 30 mile island; Florida and Hawai‘i also suffered major hurricanes that year. This, too, was an El Niño year. Clearly, there are important reasons for being able to predict when El Niño is starting. eBook for classroom use only. © 2014 Bess Press Inc.  62 PHYSICAL ENVIRONMENT SIDEBAR 2.1: Starving marine iguanas and missing jellyfish: using natural experiments to understand ENSO impacts In this panel, we describe two very different biological effects of ENSO in weeds had been replaced by brown seaweeds. These are not broken down island archipelagoes at opposite sides of the Pacific: one the result of El by the symbiotic bacteria. But worse, they are also loaded with chemi- Niño conditions, the other the result of La Niña. cals (tannins) that precipitate proteins, including digestive enzymes, and further reduce the animals’ ability to extract nutrients from their food. During the 1997–98 El Niño, up to 90% of the iguanas died (Fig. 1), but The Galápagos Islands straddle the Equator 975 km (600 mi) off the coast different islands had varying mortalities, and smaller animals survived of South America; although they are geographically tropical, the ocean is better than large animals. cool and there are no corals there. The Galápagos are right in the middle of the warm water that results when the tradewinds weaken and the Michael Romero and Martin Wikelski wanted to find out why some igua- cold upwelling stops. When we visited in the middle of 1998, one of the nas survived better than others. They found from their studies during the longest and strongest El Niños on record was just ending. Pulses of cold, 1997–98 event and the following 1999 La Niña32 that levels of the hor- upwelled water were just beginning to be recorded on data loggers that mone corticosterone were higher during famine than during feast times. the Charles Darwin Foundation Research Station had placed at 10 m (30 At the population level (pooled data for all individuals, but separate data ft) depth around the islands. The rocky shores, famed for the abundance for different islands), there was a negative correlation between high cor- of marine iguanas and sea lions, had become reeking graveyards. For the ticosterone and low survival. Individual animals with the highest hor- sea lions, the problem was the same as for the fishermen who named El mone levels often had wasted muscles. But correlation is not necessarily Niño: no upwelling nutrients, so no phyto­plankton productivity—lead- cause and effect, and they wanted to test the hypotheses concerning the ing up the food chain to no fish. role of corticosterone in survival. This could be done by following indi- vidual animals through famine, but they could not create famine; they For the marine iguanas, the problem was quite different: They were starv- needed a natural experiment, that is, where nature imposed the famine ing to death with full bellies! They eat seaweeds in the intertidal zone on individuals they were already following. Fortunately for their work, and shallow subtidal, and have a bacterial flora in their guts to help them ENSO is a recurrent event. digest green and red seaweeds that usually predominate there. With the warm water—32 °C compared to 18–23 °C—the green and red sea- This natural experiment of El Niño-induced starvation provides several major advantages for using marine iguanas as a model for examining the role of corticosterone in surviving natural stressors. First, reliance on one food source subjects marine iguanas to a fairly regular cycle of unintentional fasting, and potential starvation, every few years. Second, these iguanas have few natural predators, leaving the El Niño-induced starvation as the major stressor for adult iguanas. Third, even though they can live several decades, marine iguanas are highly sedentary and rarely move from a several hundred metre area of coastline. Strong site fidelity allows us to monitor individual animals and to estimate annual survival rates. Fourth, owing to their unfamiliarity with humans, marine iguanas are easy to capture repeatedly.33 Corticosterone is a “stress hormone” that had been studied by the bio- medical community in human situations, but was just beginning to be seen by ecologists as useful in monitoring the health of wild animal pop- ulations. Romero reviewed the various mechanisms by which corticoste- Figure 1. Marine iguanas that died of starvation during the rone might work.34 Subsequently, in 2002 he and Wikelski gathered their 1997–98 El Niño in Galápagos. baseline data—corticosterone levels in healthy, individ­ually tagged ma- eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 63 rine iguanas—and waited for the next El Niño. Because there appeared to be size differences and potentially sex differences, they selected only large males, the most likely to suffer from famine. The moderate ENSO of 2003 (Fig. 2-15) provided just enough of a famine for their test. (Luckily it was a moderate event, so that only 23% of the tagged animals died.) They found that the animals that survived were better able to turn off the stress hormone, which may have survival benefits in the short term, but is associated with body protein breakdown in the final stages of star- vation. Starving animals first consume their carbohydrate, then their fat stores, and finally their protein (muscle tissue). Animals with high cor- ticosterone entered the final phase sooner and were more likely to die; those that could turn off the initial stress response and lower their hor- mone levels survived. Wikelski even found evidence that marine iguanas Figure 2. Life cycle of Mastigias jellyfish showing the alternate can reversibly shrink their body size in response to famine.35 (scyphistoma) stage. (Reprinted from Patris et al. 2012, design by Mandy Etpison, © Coral Reef Research Foundation.) The islands of Palau are tropical both on land and in the ocean, and be- sides the diverse coral reef life have some very interesting lakes with jel- ing field observations on the numbers of medusae and the temperature, lyfish (described in detail in Sec. 6.1.5). One lake in particular, popularly salinity, and oxygen profiles in the lake (see Fig. 6-15). And finally, they called Jellyfish Lake,36 became an important tourist destination in the also made observations in comparable lakes nearby. Out of all these tests mid-1980s, both for divers between dives and as a destination in its own and observations, the only hypothesis that was supported was that the right for snorkelers going to see the swarms of harmless golden jellyfish. disappearance of medusae in Jellyfish Lake was the result of high tem- The population of jellyfish appeared to be stable at 1.6–1.8 million (mea- peratures. These in turn came about because very wet conditions of the sured in 1979 and 1996; see Sec. 5.3.5). But from September 1998 to May strong La Niña that followed the El Niño created a freshwater layer at the 1999, it crashed to zero. No one knew why, no one knew if the population surface, and that impeded mixing of the lake water.38 would recover, and all manner of causes were proposed as explanations: The disappearance was attributed variously to the 1997–98 El Niño, to stagnation Beyond its interest in showing that tourism had not damaged the re- of the lake because conduits that normally allowed tidal flux were somehow source, their study showed that tight coupling of an ENSO signal and bio- blocked, or to tourists who may have stolen or eaten the jellyfish, poisoned the logical and physical conditions in Jellyfish Lake could be useful in moni- medusae with sunscreen, or polluted the lake with urine.37 toring and understanding ENSO. This signal is obscured in the nearby lagoons where water conditions are quite different. Moreover, the lake bottom has very fine undisturbed sediments going back to the origin of But the big El Niño of 1997–98 was over, and none of the explanations the lake some 5,000 years ago (Fig. 6-13). It may be possible to detect seemed likely. Scientists at the Coral Reef Research Foundation, who had ENSO signals from the past and so have a longer perspective on the fre- been working on the jellyfish and the lakes for several years, undertook quency and intensity of ENSO events.39 to figure out what happened. First they discovered that while the familiar medusa stage in the jellyfish life cycle had vanished, the species existed in the tiny benthic scyphistomae stage (Fig. 2). They tested several hy- potheses by running experiments on both the medusae and the scyphis- tomae, and they looked for effects on both the animals themselves and the relationship with symbiotic zooxanthellae. They also continued mak- eBook for classroom use only. © 2014 Bess Press Inc. 64 PHYSICAL ENVIRONMENT If the winds cause the ocean to change, what causes the winds to change? Now scientists see the interaction between ocean and atmosphere as the cause of the ENSO. Warm water warms the air above it, and more evaporation causes clouds to form. These changes in the air affect the winds. There is a kind of dialogue between water and air. The atmosphere responds within days or weeks to ocean temperature changes, whereas it takes the ocean many months to respond to atmospheric changes. This difference in time frames keeps the oscillation going: It is this “memory” of the ocean that makes the Southern Oscillation continual. The oceanic conditions at a certain time are not simply determined by the winds at that time, but also depend on winds at earlier times. During El Niño, for example, the ocean has a “memory” of winds that prevailed during La Niña, and it is the delayed responses to those winds that brings about the termination of El Niño and introduces the next La Niña. 14 The oscillations have a major impact on rainfall in the western Pacific Islands. Islands that are normally wet, such as Guam, Palau, and Pohnpei, become dry. Islands that are normally dry, such as the Marshall Islands and Kiribati, become wet (Fig. 2-14). Drought as severe as the one that occurred in 1982–83 might occur in western Micro­nesia every 125 years; in the eastern Carolines, perhaps every 250 years.15 Western Micronesian islands (Palau, Yap, Chuuk, and the Mariana Islands) normally have dry seasons, which are more easily worsened into drought than the less seasonal climates of the eastern Carolines (Pohnpei and Kosrae). The climate and sea level changes can have a big impact on small islands that have a limited water supply and use reefs as important fishery resources. In the past, such changes would have had a bigger impact than at present, because now we can import food if crops or reef fisheries fail. But as island populations grow and push the limits of water supply, droughts could once again become threatening. Is it now possible to predict an ENSO event? The delayed response in ocean conditions comes some time after the pressure change, so the pressure change seemed like an early warning. Peruvian scientists began issuing predictions each November for conditions in the following year, and farmers used these predictions to choose the best mix of wet- climate crops (e.g., rice) and dry-climate crops (e.g., cotton). This has given scientists some confidence in their developing model of the ENSO, but there are still problems with the model. El Niño conditions started in 1990 as predicted, but did not follow the apparent pattern of recent years. Scientists expected El Niño conditions to last a year or so and then swing back to “normal,” but they lasted into 1995 with some variation (Fig. 2-15). None of the computer programs at present can model these changes. These swings are triggered by still-unknown fluctuations in the ocean and atmosphere. Research is continuing to seek better predictors. For example, multivariate indices incorporate data from other sources, such as sea surface temperature anomalies. Each ENSO event is different in timing and the way it develops. Strong La Niña conditions (such as occurred during much of the first decade of the 21st century) can also be disastrous in some parts of the world. Meanwhile, meteorologists are trying to find correlations between local weather and the evolving ENSO models. For example, Mark Lander at the University of Guam has found eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 65 that rainfall in Guam and Micronesia correlates with the direction of the mean Southern Oscillation Index El Niño/Southern Oscillation is a vast weather during January to June.16 If the mean is rising, the pattern involving the ocean and the atmosphere. climate is dry (1988, 1992, 1995); if it is falling, the climate is wet (1994). (1993 was an in-between year, in During El Niño years, easterly tradewinds weaken which Guam was very dry but Micronesia had normal and change the depth of the thermo­cline. These rainfall.) changes cause changes in sea level and rainfall. 2.4 Typhoons (hurricanes) 2.4.1 Typhoons are giant whirlwinds The sounds of a typhoon tearing at your home create a long-lasting memory. The eyewall is an awe-inspiring horizontal waterfall of rain going 150 miles per hour across the land. Trees bend and break. Waves whipped up by the storm break over the reefs and flood lowland areas. Waves may fill houses with sand or sweep them off their foundations. As the typhoon moves away and the winds die down, people come out to see the damage. When the scare is over, the cleanup begins. Branches, leaves, coconuts, power lines, and construction debris litter the ground. The destruction of property and human lives can be dramatic. Tropical cyclones are spiral storms that form in the Pacific Ocean, in the Atlantic Ocean, and in the Indian Ocean. They affect many parts of Micronesia, the Philippines, Polynesia, and the Caribbean islands (Fig. 2-16); the Western Pacific receives the largest percentage of the global total and the great majority of the category 5 storms or supertyphoons. Typhoons, Figure 2-16. Tracks and intensities of all tropical storms. Graphic has intensities layered so that category 5 (supertyphoon) is on top, hence the big red patch in the western Pacific where many strong typhoons occur. Note the absence of storm tracks very close to the Equator. Location of Guam indicated by asterisk. Saffir-Simpson Hurricane Intensity Scale: TD = tropical depression; TS = tropical storm; 1–5 = hurricane / typhoon categories from weakest to strongest. (Graphic © globalwarmingart.org) eBook for classroom use only. © 2014 Bess Press Inc.  66 PHYSICAL ENVIRONMENT also called hurricanes, are tropical cyclones with maximum sustained winds of 120 kilometers per hour (74 miles per hour) or more. They are called hurricanes east of the International Date Line, typhoons in the western Pacific (north of the equator). Typhoons and hurricanes are the same, but tropical cyclones also include tropical depressions and tropical storms (winds less than 120 kph). We will refer to them in general as typhoons, except when naming specific storms or referring specifically to Caribbean hurricanes. Typhoons begin as areas of weakly circulating clouds called tropical depressions. They get their energy from warm ocean waters. They spin counter­clockwise in the northern hemi­ sphere and clockwise in the southern hemisphere. They form a few degrees north and south of the equator, especially where the water is warmest, drawing their energy from the heat in the water. Because warm water is the “engine” that powers typhoons, typhoon intensities appear to be getting stronger with global warming. Typhoons can intensify or weaken quickly, sometimes catching island residents unprepared despite warnings. Of course, for most of human history on tropical islands, there was no warning system; people relied on magic as defense against typhoons.18 Many factors influence how a typhoon moves and changes, and many of these factors are not yet well understood. Meteorologists must study all the surrounding weather to make their predictions of what each storm is likely to do. Staff of hurricane centers conduct scientific research using the data from each storm to add to their knowledge. Some of these centers are the Joint Typhoon Warning Center (Hawai‘i) and the National Hurricane Center (Florida).19 Table 2-2. Areas of typhoon/hurricane development and the change in numbers of Typhoons are giant whirlwinds, category 4 and 5 storms during two 15-year periods.* (See also Fig. 2-16.) like torna­does but bigger and lasting for days rather than Period minutes. The wind speeds in the center of a typhoon 1975–1989 1990–2004 are comparable to small tornadoes (about 160 kph Basin Number Percentage Number Percentage or 100 mph). Some Atlantic/ Caribbean hurricanes gen­erate East Pacific Ocean 36 25 49 35 tornadoes when they move onto the mainland, but this West Pacific Ocean 85 25 116 41 apparently does not happen North Atlantic 16 20 25 25 over islands. The fastest winds in the typhoon are under the Southwestern Pacific 10 12 22 28 “wall” cloud, near the eye. Wind speeds decrease quickly away North Indian 1 8 7 25 from the center, so that a storm passing 150 km or even 75 km South Indian 23 18 50 34 away may be hardly felt. As the * Reprinted from Webster et al. 2005 with permission of Science.40 eye passes, the winds suddenly die down to near zero, then just eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 67 as quickly increase back to maximum speed from a different direction on the other side of the eye. Meteorologists differentiate intensity from the strength of a storm. They measure the maximum winds near the center as the intensity of the storm. They use strength to indicate how far away the winds are still strong, or the size of the storm. Hurricanes moving over shallow water of coastal mainland areas can create extreme tides called storm surges. Around oceanic islands, however, this does not happen because of the deep water near shore. Instead, rapidly moving typhoons can build huge waves if they are traveling fast enough to stay on top of the waves. Waves like these produce very erratic water levels on islands instead of big sheets of floodwater. Each storm has a unique combination of conditions, as Stuart Pimm and colleagues explain about Caribbean hurricanes: [T]here is no typical hurricane.... Andrew was relatively dry, fast-moving, of small diameter, caused little saltwater incursion, and was not followed by fires. Other Florida hurricanes have been wetter, slower-moving, have damaged vegetation on a wider path, and inun­dated larger areas with saltwater. The fires that followed hurricane Gilbert in Yucatan killed more trees than did the storm itself. Two rain gauges that survived the storm recorded approximately 5 cm of rain, making Andrew a relatively dry hurricane. In this region [south Florida], less severe storms have dropped more than 12 cm of rain, and rainfall in particularly wet storms can be more than a meter.20 Typhoons are a significant part of the physical environment in the Pacific and Carib­bean islands. Damaging winds occur frequently, at least when seen on the time scale of trees and ecosystems. In the Caribbean, hurricanes strike individual islands every 15 to 20 years. Guam had 52 tropical cyclones pass within 70 miles in the half century from 1945 to 1995 (this includes the tracks of all storms that became typhoons at some stage; however, most were not typhoon strength at their closest point to Guam). Guam’s record is three eye passages (when the eye passes over some part of the island) in less than three months between August and November 1992 (plus two more within 70 miles).21 There is a strong correlation between the speed of the wind and the frequency with which a location will experience that speed. For instance, a supertyphoon (winds of 150 mph) will hit Guam about once every 60 years. A supertyphoon passing to the north or south may not cause supertyphoon winds on the island. Minimal-strength typhoon winds (75 mph) will happen every 4 to 5 years on average. These statistics may change if ENSO patterns change because of climate change. Typhoons bring changes to ecosystems; later we will describe impacts on forests (Secs. 7.3.8, 7.3.9) and reefs (Sec. 9.3.6). eBook for classroom use only. © 2014 Bess Press Inc. 68 PHYSICAL ENVIRONMENT 2.4.2 Typhoons are common in Western Pacific islands Sometimes Pacific Islands become virtually abandoned as a result of a typhoon, as happened to the Southwest Islands of Palau in 1905.22 Typhoons are particularly damaging to low-lying coastal areas and atolls because of the high sea levels or waves they can cause. People who settled the atolls in the Western Caroline Islands soon found that typhoons would be part of life, as Catherine Lutz explains: Typhoons, as much as any other single environmental factor on Ifaluk,... helped to produce the contemporary social organization and emotional config­uration of the island, as we will see. A severe typhoon can result in the near total devastation of the food supply; winds of up to 140 miles per hour can fell the island’s massive breadfruit trees as well as coconut trees and buildings. Rising seas can cover the atoll, washing over the taro gardens and killing the plants.* It has been estimated that six to ten years may be required after such a typhoon inun­dation before the island vegetation returns to full flower. Waves from a typhoon also smash the living coral communi­ties on the reef, thereby cutting back severely on the number of fish they can support. *A typhoon’s inundation of the taro gardens can have other effects, as Marshall notes: “Wave damage can also eradicate the intricate boundaries marking individually owned taro plots, an action equivalent to loss of land registration files in the United States” 23 The stark list below gives official testimony to Typhoon Owen’s romp across the Caroline Islands from November to December 1990. Only a few weeks earlier, Typhoon Mike struck Palau, devastating Kayangel Atoll and damaging Babeldaob and Koror. A few weeks after Owen, in late December, Typhoon Russ made Guam a federal disaster area. POHNPEI: 2 killed when a live power line fell and struck them. CHUUK STATE: declared a U.S. federal disaster area, 1000 people left home­less, major power failures. Hall Islands: extensive crop damage, nearly all homes destroyed, all food crops destroyed. Namonuito Atoll: extensive crop damage, nearly all homes des­troyed, all food crops destroyed. Pulap Atoll: extensive crop damage, 99 percent of homes destroyed. YAP STATE: declared a U.S. federal disaster area. Satawal: reported winds in excess of 100 mph, 95 percent food crop des­troyed, 90 percent homes damaged, all power lost. Lamotrek Atoll: reported winds in excess of 100 mph, 85 percent homes destroyed, 95 percent food crop destroyed, all power lost. Elato Atoll: 99 percent dwellings destroyed, 90 percent food crops destroyed. eBook for classroom use only. © 2014 Bess Press Inc.  PHYSICAL ENVIRONMENT 69 Ifalik Atoll: dwellings—no report, 95 percent food crops destroyed, 20 percent land eroded. Woleai Atoll: 85 percent dwellings, 90 percent food crops destroyed. Faraulep Atoll: 20 percent dwellings, 100 percent canoes, 100 percent food crops destroyed, 20–30 per­ cent land eroded. (Automatic wea­ ther obser­ vation site on Farau­lep was lost during passage of Owen. The shore was completely eroded away leaving the tower on its side in 10 feet of water and 20 yards off the beach. Site is now abandoned. ) Ulithi Atoll: 30 percent dwellings and government buildings, 100 percent food crops destroyed.24 To most people, the damage is a disaster. However, to a biologist whose study site is in the path of a typhoon, the damage may be literally a windfall. There is no way to duplicate the effects of a typhoon. Scientists can only learn about the effects of major disturbances by waiting until one happens in an area they have already studied (Sec. 7.3.8 and as noted for ENSO in Sidebar 2.1). As with any other damage assessment, a forest survey is not much use unless “before” data are available for com­parison. Descriptions of undisturbed areas are the baseline data against which to measure the changes. When Hurricane Andrew cut across southern Florida in August 1992 it struck three national parks, including a marine park.25 In 1988 to 1989 three other major hurricanes had hit areas with biological study sites: Hurricane Gilbert (September 1988) swept along the south coast of Jamaica; Hurricane Joan (October 1988) hit Belize; and Hurricane Hugo (Sep­tember 1989) hit the U.S. Virgin Islands, Puerto Rico, and South Carolina.26 From studies before and after these recent storms, scientists are beginning to learn some of the impacts on forests, freshwater ecosystems, and reefs; we will describe these after we describe the ecosystems in subsequent chapters. Scientists are also getting a better idea about what they still need to discover. They will continue to monitor their study areas to watch how the ecosystems recover (or change) over months and years. Scientists speculate that these ecosystems will recover because hurri­canes are a natural part of the environment. However, they are uncertain both about the details of the recovery and about how the human impacts on these ecosystems will alter or impede natural recovery. In their summary report on the immediate effects of Hurricane Andrew, Pimm and survey team leaders pointed to the need for long-term follow up: Only continued monitoring will allow modeling and then testing of speculations about the long-term consequences of ecosystem frag­mentation, alien plants, the full range of ex­ternal threats, and natural episodic events. 25 eBook for classroom use only. © 2014 Bess Press Inc. 70 PHYSICAL ENVIRONMENT 2.4.3 Typhoons affect atolls particularly strongly Typhoons affect some atoll areas, such as the Marshall Islands, outer islands in the Feder­ated States of Micronesia, and the Tuamotus of