ࡱ> Y 0LbjbjWW B== H#]4h0dlAf(Y Y Y TAVAVAVAVAVAVA$BDzAY  @Y Y Y zA  Y :TAY TA Z c\7 TA@PeA v@\Atmosphere-Ocean Interactions The most costly natural disaster the United States has experienced occurred in August 1992 when Hurricane Andrew hit Florida. Hurricanes (Huracan = the Taino god of the wind) and typhoons (Taifung = Chinese for "great wind") are both tropical cyclones; Hurricanes occur in the Atlantic Ocean and typhoons in the Pacific Ocean. To qualify as a hurricane or typhoon the storm must have sustained winds exceeding 64 knots (74 miles per hour). Though the high winds inflict damage when these storms hit land, it is the huge waves; storm surge and associated flooding that cause most of the destruction. These ferocious atmospheric storms cannot survive without the moisture supplied by warm ocean water. They are vivid examples of the coupling between the earth's oceans and atmosphere. Ocean currents and variations in oceanic water temperature have been known for many centuries. Benjamin Franklin was the first person to chart the Gulf Stream. He did this in 1769 based on his observations and those of the whalers off the coast of New England. Benjamin Franklin also made some careful observations of the atmosphere and was no doubt aware that interactions existed between the atmosphere and the ocean. This chapter explores the links between the atmosphere and oceans.  (Figure copied for Garrison, figure 2.21 - not sure where he got it. Oceanography Oceanography is the study of the oceans. An understanding of weather and climate requires knowledge of oceanography because of the interaction between the atmosphere and the oceans. There are three important roles the oceans play in relation to weather and climate: 1) They are a source of atmospheric water vapor and 2) They exchange energy with the atmosphere and 3) They transfer heat poleward. Each year approximately 396,000 cubic kilometers (94,644 cubic miles) of water vapor circulates through the atmosphere through the hydrological cycle. Most of the water that recycles into the atmosphere comes from the oceans (Chapter 1). About 334,000 cubic kilometers (79,826 cubic miles) of water evaporates from the oceans every year. At any given time, the atmosphere contains approximately 15,3000 cubic kilometers (3,580 cubic miles) of water. It takes approximately two weeks for all of the water in the atmosphere to be recycled. The oceans provide the majority of water needed to form precipitation. Exchanges of heat and moisture occur at the interface between the atmosphere and the ocean. Figure 8.1 shows the net energy gains and losses at this interface for winter and summer. These energy exchanges include latent heat, sensible heat, and radiative exchanges. On average, the ocean gains energy during summer and losses energy to the atmosphere in winter. So, on average the oceans warm in the summer and cool during the winter. These energy budget maps (Figure 8.1) are a measure of the average interaction between the atmosphere and ocean. Maximum exchanges of energy occur in the Northern Hemisphere winter to the east of North America and Asia. Warm ocean currents flow northward in these regions (Chapter 3) and supply energy to winter storms that are leaving these continents heading eastward. As we will learn later in this chapter, warm ocean waters also supply the energy that fuels hurricanes. The rate of heat and moisture transfer, as discussed in Chapter 2, depends on the temperature difference between the air and the water and the winds. Warm sea surface temperatures and strong winds are favorable for large heat exchanges between the ocean and atmosphere. So, it is important to understand where the warm waters of the oceans are and to know why they are there. We learned in Chapter 3 that warm currents of the middle and high latitudes border the East Coast of some landmasses. Countries such as Norway have mild winters and small annual temperature ranges. The West Coasts of landmasses that boarder cool waters, such as San Francisco, have low average temperatures and small annual ranges of temperature (Chapter 3). This Chapter will discuss the origins of these cool and warm currents Like the winds, the ocean currents transport heat from the tropics to higher latitudes. Figure 8.2 shows the poleward transport of heat in the Northern Hemisphere by the atmospheres and oceans required to balance the radiation budget, as discussed in Figure 2.20 (reproduced as an inset of Figure 8.2). The poleward transport of heat by the atmosphere is discussed in the previous chapter. At about 30( latitude, the atmosphere and ocean each transport about the same amount of heat. Equatorward of 30( the ocean transfers the majority of heat required to maintain a balance. The atmospheric transport has a broad peak at around 30(N and 60(N. This transport of heat by the atmosphere and ocean is coupled to our current climate. Any changes in the ocean circulation patterns may change the heat transport by the oceans and thus impact climate. Changes in the ocean and its circulation patterns affect the atmosphere and visa versa. Solar energy is the primary source of energy that warms the oceans. Atmospheric conditions determine the amount of solar energy reaching the ocean. Cloud free conditions result in more energy entering the oceans while cloudy conditions reduce the solar energy reaching the surface. Winds also play an important role in determining ocean temperatures. The solar energy that is absorbed at the surface of the water can be mixed to deeper layers by the winds, which stir the upper layer of the ocean. Ocean Temperature Chapter 1 described the atmosphere based on vertical distribution of temperature. In the oceans, we are interested in the vertical distribution of water density. Observations of temperature with depth in the ocean at a given location reveal a vertical distribution as demonstrated in Figure 8.3. Based on measurements of temperature we can classify the ocean into three vertical zones. The characteristic feature of the top 100 meters of the ocean at a given location is a constant temperature. The uniform layer is called the surface zone or mixed layer. Wind driven waves and current mix this layer so that the temperature is relatively uniform. Only about 2% of the world's ocean waters are within the surface zone. The bottom layer, or the deep zone, lies below about 1000 meters and is composed of cold water with temperature between -1C to 3C (30.5-37.5F). The temperature is uniform in this deep layer. The transition zone between the surface and deep layers is the thermocline (therm means heat and clinare, to lean). In the thermocline temperature decreases rapidly with depth down to a depth of about 3000 feet. Thermocline is a zone of the ocean in which the temperature and decrease rapidly with depth. Figure 8.3 is a typical temperature distribution of the ocean with depth. This temperature variation with depth is a function of latitude. Figure 8.4 shows the temperature profile with depth for oceans typical of tropical, midlatitude and polar regions. The deep zone temperatures are similar for all three regions. The tropical waters have the warmest surface zone temperatures and the polar regions are the coldest. This is because the polar regions receives less solar energy. The tropical and midlatitude regions have the strongest thermocline. Sea surface temperature (SST) distributions Interactions between the atmosphere and ocean occur at the surface transfer heat and moisture. The distribution of sea surface temperatures is therefore important in meteorology. Indeed, as we shall see later in this chapter, the sea surface temperature, or SST, determines the distribution of hurricanes. Figure 8.5 shows the SST distribution in the world's oceans, with red regions representing warm water and blue regions cold. There are a few general conclusions we can draw from this figure, in addition to the one already made regarding tropical waters being warmer than the polar waters. These additional observations are: Western coasts in the subtropics and middle latitudes are bordered by cool-water. Western coasts in tropical and subtropical latitudes are bordered by warm-water. Eastern coasts in the mid latitudes are bordered by warm-water currents Eastern coasts in polar regions are bordered by cool-water. In addition to the temperature distribution patterns, it is important to know in which direction these warm and cold waters are moving. We know from energy transport principals that on average cloud waters move equatorward and warm waters poleward. Ocean Currents An ocean current is a massive ordered pattern of water flow. There are two basic types of ocean currents, surface currents and deep water currents. Surface currents occur in the uppermost 10% of the world oceans, 90% of the ocean waters circulate beneath the surface currents in the deep-water currents. Like the atmosphere, the oceans are subjected to forces that determine the direction and intensity of the flow of water. First we will list these forces and then discuss how they affect ocean currents. The primary forces that determine the ocean currents are: Friction due to wind blowing over the water, Gravity, which arises from differences in the height of the ocean surface due to thermal expansion and contraction, Density differences that arise due to differences in water temperature and salinity, Coriolis effect, just as the atmosphere is, and, The shape of the ocean basins. Surface currents The wind and absorption of solar energy drive the ocean surface currents. Air blowing over the oceans transfers some of the energy from the moving air to the water. This causes the water to flow. The surface current is comprised of the water flowing beneath the surface winds and extends to a depth of about 100 meters. As you learned in Chapter 6, the global wind patters are a function of latitude and at the surface can be broken down into three main patterns: the trade winds, the mid-latitude westerlies and the polar easterlies. The trade winds and the midlatitude westerlies have a strong influence on the ocean surface currents because they are steady and flow over large areas of open water. Gyre is a current that circulates around the periphery of an ocean basin. Figure 8.6 shows the surface water currents. Warm water flows to the cold polar regions while cool waters flow equatorward. So, as in air currents in the atmosphere, ocean currents transfer heat poleward. In the tropics, the trade winds push water from one side of an ocean basin to the other, causing the water to pile up on one end of the basin. Since the water is higher on one side, gravity will act to move the water down to lower heights, but it is opposed by the trade winds. As with the winds, the Coriolis effect acts over large distances. In the Northern Hemisphere it acts to the right and to the left in the Southern Hemisphere. This effect causes the to water to flow within the basin in a circular pattern called a gyre. Details of the North Atlantic gyre are shown in Figure 8.7, along with the fundamental surface wind direction. Solar heating also influences surface ocean currents. Tropical regions receive more solar radiation than midlatitude and polar regions. So, just like the atmosphere, on average tropical waters are warmer than polar waters. As water warms it expands. As a result, the sea level near the equator is about 8 centimeters higher than the sea level of mid-latitude oceans. Gravity causes water to flow edownhill, and so the water moves from tropics toward the poles. However, the Earth is rotating faster than the water can flow, so the water tends to flow toward the west, and the Coriolis effect moves it to the right contributing to the formation of the gyres. As we learned in Chapter 6, friction and the Coriolis effect caused the wind to turn to right in the Northern Hemisphere. We see a similar effect in surface ocean currents. The surface water currents shown in Figure 8.7 are flowing at an angle of approximately 45( to the right of the surface winds. This results from a combination of friction and the Coriolis effect. These surface currents pull water below them so that the flow extends down approximately 100 meters (330 feet). The direction of the currents changes with depth. This is illustrated in Figure 8.8 for a Northern Hemisphere current. The direction of the water flow decreases with depth and rotates counterclockwise. This spiraling motion is called the Ekman spiral. Ekman Spiral is an explanation of how wind blowing over the ocean surface can generate water flow below the surface. To explain the Ekman spiral, let's divide the ocean into slabs of water. Starting with the top layer and working down into the ocean. Friction pushes the surface water and the Coriolis force moves the water to the right, and the water moves at a 45( angle to the right of the wind. This shallow surface layer of water exerts a frictional drag force on the layer below. This moves the water below, but at a lower speed and the Coriolis force again bends it to the right. Each layer exerts a frictional drag on the layer below, but the flow decreases with depth as friction slows the water, explaining the decrease in water speed with depth. Each underlying layer of water turns to the right due to the drag exerted by the layer above and the Coriolis effect-explaining the clockwise rotation of the water movement. The warm currents on the western side of the major Ocean basins (east coast of the continents) are called the Western Boundary Currents. The western boundary currents move over large distances. The Gulf Stream is the largest of these currents. Figure 8.9 is a detailed infrared satellite image of the Gulf Stream. Note the looping meanders and eddies. Eddies form when the current loops and cuts-off an eddy (Figure 8.10) from the main flow. Warm eddies that form in the Gulf Stream rotate counterclockwise, while cold eddies rotate clockwise. Similar eddies can form in the upper regions of the atmosphere! Coastal Upwelling Upwelling is a circulation pattern where cold nutrient-rich deep waters move to the surface. Upwelling can be generated along coastlines by the winds. The wind-driven surface water can induce vertical motions of the upper layer of the ocean. This process of inducing vertical motions is called upwelling. Coastal upwelling occurs along coastlines and brings cold, nutrient-rich water to the surface, an important source of nutrients for other marine organisms. Consider a wind blowing parallel to the shoreline. This is a typical situation off the west coast of continents due to the presence of the subtropical highs. Friction causes the surfaces waters to move and the Coriolis effect deflects the motion of the water. The resulting Ekman spiral causes the surfaces waters to move away from the shoreline. As the surface water moves away from the coast it is replaced by water rising from below (Figure 8.11), which is nutrient rich and cold. The coastal upwelling is observed off the west coast of North and South America images of sea surface temperatures (Figure 8.5). Deep Water currents About 90% of the world ocean currents occur in the ocean deep waters. The slow circulation throughout the great depths of the oceans are driven by density differences in water. Cold water is denser than warm water and saline water is denser than fresh water. Because the deep-water currents are driven by density difference it is also called a thermohaline circulation (therme meaning heat and halos salt). An ideal thermohaline circulation is shown if Figure 8.12. As warm tropical surface water flow poleward in the surface layer, the water cools as it transfers heats to the atmosphere. As it cools, the density increases. When the surface waters reaches the polar regions, the water is cold and it sinks. This sinking occurs over a relatively small area near the poles. It moves toward the equator below the surface water and slowly rises to the surface to replace the surface water moving poleward. The upward motions are very slow, approximately 1 centimeter (about 0.5 inch) per day. El Nio What is El Nio El Nio is a warming of the equatorial Pacific Ocean between South America and the Date Line. El Nio is a natural variation of the ocean and atmosphere and is not the result of human activities. El Nio is an excellent example of the interaction between the atmosphere and the ocean and how these interactions effect climate. The waters off the western coast of South America typically experience upwelling. The deep ocean waters are cold and nutrient-rich. Every two to seven years shifts in winds and ocean currents bring warm water from the other side of the Pacific Ocean to displace the nutrient-rich cold water that normally wells up from deep in the ocean. This warming phenomenon typically first appears off the coast of South America around the Christmas season and lasts for several months. The term "El Nio", or "the Christ Child", was used by the fisherman along the coast of Ecuador and Peru, where the economy of the coastal communities depend on the cold nutrient-rich waters. The warm El Nio waters disrupt the marine food chain in the region and the local economy suffers from the loss of fish. We have recently recognized that El Nio is linked to shifts in global weather patterns. Under normal conditions, the SSTs (sea surface temperature) off Peru are cold due to coastal upwelling and waters in the western equatorial Pacific are warm. Sea surface temperatures off South America's west coast range from the 60s to 70sF, while SST are greater than 80F in the central and western Pacific. The steady easterly trade winds push the water towards Australia and Indonesia. As the water flows westward it warms due to absorption of solar energy and heat exchanges with the atmosphere. Over the western Pacific Ocean, the warm water evaporates and is accompanied by precipitation (Figure 8.13a). An El Nio is triggered when the trade winds weaken or reverse direction. This allows the large mass of warm water that has piled-up near Australia and Indonesia to move eastward along the equator until it reaches the coast of South America. During an El Nio event, there are two very distinct changes in the equatorial Pacific Ocean: 1) Cold coastal waters are replaced by warm waters and 2) the height of the ocean surface drops over Indonesia and rises in the Eastern Pacific. Satellites can measure both of these features with an active sensor called an altimeter. Figure 8.14 shows global deviations in the height of the ocean surface as measured by a NASA satellite. The difference in height between the eastern and western tropical Pacific Ocean is more than 60 centimeters (12.5 feet)! It is important that satellites can measure the height of the ocean surface because this is the first thing to change in and El Nio event. Satellite observations can predict El Nio, and improve seasonal weather forecasts! How does El Nio affect global weather patterns? Rain clouds follow the warm water, affecting regional precipitation patterns. With the El Nio, the western Pacific experiences below normal precipitation as the heavy rains move eastward towards South America (Figure 8.13b). The shift in the large tropical rain clouds alters the typical pattern of the jet stream. The tropical convective systems disrupt upper air patterns much like islands modify the winds that blow on them, only on a much larger scale. This shift in the jet stream impacts global weather patterns (Figure 8.15). To determine the impact of El Nio on global climate patterns, meteorologists compare average weather conditions with weather conditions experienced during an El Nio year. There have been seven recent major El Nio events: 1957-58, 1965, 1968-69, 1972-73, 1976-77, 1982-83, 1986-87, 1991-92 and 1994-95. The 1982-83 El Nio was the strongest of this century. Figure 8.16 compares differences in temperature and precipitation between normal and El Nio years for winter and summer conditions. The impacts of El Nio upon climate in temperate latitudes show up most clearly during wintertime. During an El Nio event, a weak polar jet stream forms over eastern Canada s and as a result a large part of North America is warmer than normal. During winter, the southeast United States experiences above normal precipitation. The increased rainfall across the southern United States and in Peru has caused destructive flooding. Drought in the West Pacific has also been associated with devastating fires in Australia and Indonesia. With El Nio conditions, changes in precipitation and temperature patterns have an effect on snowfall in the United States (Figure 8.17). In the Southwest, there is a slight tendency toward cooler winters, and a strong tendency toward wet winters, which makes higher elevation snowpack deeper. In the Pacific Northwest, El Nio winters are warmer and drier than usual, so that at a given elevation there is less precipitation and the freezing level is at a higher altitude. The type of precipitation is more likely to be rain and the accumulation season is shorter. These factors produce a smaller snowpack accumulation by the end of winter in the Pacific Northwest. A significant reduction in total winter snowfall also occurs in the Midwest and New England regions during strong El Nio. Atmospheric scientists have recently realized that the El Nio is associated with a seesaw of atmospheric pressure between the eastern equatorial Pacific and Indo-Australian area. This seesaw in pressure is referred to as the Southern Oscillation. Scientists have been studying this oscillation since the 1890s. The Southern Oscillation is an irregular fluctuation of atmospheric conditions over the Indian and Pacific Oceans. Generally, when pressure is high over the Pacific Ocean, it tends to be low over the eastern Indian Ocean, and vice versa. The Southern Oscillation is monitored by measuring sea-level pressure at Tahiti in the east, and at Darwin, Australia in the west. The difference in the pressure at these locations is called the Southern Oscillation Index (SOI). With a high positive SOI pressure is high in the western Pacific, the trade winds are strong and rainfall in the warm western tropical Pacific is plentiful. The Southern Oscillation is closely linked to the El Nio. High negative values of the SOI represent an El Nio, or "warm event." El Nio and Southern Oscillation often occur together, but also happen separately. When an El Nio and Southern Oscillation occur together, the event is referred to as ENSO. La Nia Is La Nia the opposite of El Nio? La Nia is defined as cooler than normal sea-surface temperatures in the eastern tropical Pacific Ocean. La Nia is the counterpart of El Nio. In a La Nia event, the sea surface temperatures in the equatorial eastern Pacific drop well below normal levels. Intense trade winds move warm surface waters of the Pacific westward, while increase cold water upwelling in the eastern equatorial Pacific. Figure 8.18 compares the sea surface temperature and winds for three Decembers, a normal year, and El Nio year and a La Nia year. La Nia years often, but certainly not always, follow El Nio years. La Nia conditions typically last for 9-12 months but on occasion may persist for as long as two years. Both El Nio and La Nia impact global climate patterns. In many locations, especially in the tropics, La Nia produces the opposite weather departures from El Nio. During La Nia, weather is drier than normal in the Southwest U.S. in late summer through the subsequent winter and in the Southeast in the winter. Drier than normal conditions also typically occur in the Central Plains in the fall. With a well-established La Nia the Pacific Northwest is wetter than normal in the late fall and early winter. As for temperature distributions in the US, on average La Nia winters are warmer than normal in the Southeast and colder than normal in the Northwest (Figure 8.19). Tropical Cyclones: Hurricanes and Typhoons What are they? The Earths atmosphere and oceans can interact in all sorts of ways. Weve already seen what happens when the vast Pacific and the air above it talk to each other over periods of several years: El Nio and La Nia. A different kind of weather feature develops when certain regions of the tropical oceans interact with the atmosphere during the summer and fall of each year. From space, they look like large circular swirls of clouds (Figure 8.20). They tend to be a few hundred miles (several hundred km) in diameter. These swirls are clearly much bigger than an individual thunderstorm, but are also much smaller than the global circulations induced by El Nio. Tropical cyclones are storms driven by atmosphere-ocean interactions on time and space scales shorter than El Nio-Southern Oscillation. In the Americas, the worst of them are called hurricanes. In the western Pacific they are called typhoons. Because of the location of their birth and their pattern of clouds, these swirls are given the generic name tropical cyclones. In the tropical regions of North and Central America, the worst of them are called hurricanes. Residents of the western Pacific call them typhoons. In most other parts of the world, such as the Indian Ocean, they are simply called cyclones. In an average year, about 5 or 6 hurricanes form in the Atlantic and the Gulf of Mexico, 9 in the eastern Pacific off of Mexico, and 16 typhoons in the western Pacific Ocean. Regardless of their name, these storms are the most highly organized and destructive weather patterns on Earth. A hurricane in 1900 hit Galveston, Texas and killed approximately 8000 people. A cyclone ravaged Bangladesh in 1970 and killed 300,000 people. And in 1992, Hurricane Andrew blew through southern Florida and caused about $25 billion in damage (Figure 8.21). Table 8.1 lists the most damaging hurricanes to hit the U.S., including many mentioned in this chapter. To save lives and property, we need to learn more about these remarkable storms. What do they look like? The clear area of lowest pressure at the center of a strong tropical cyclone is called the eye. The circular region of strong thunderstorms immediately surrounding the eye is called the eye wall. The name tropical cyclone, plus photographs of these storms, reveals quite a bit about their makeup. As we learned in Chapter 6, a cyclone is a center of low pressure. Weather satellite photographs (Figure 8.22) indicate that the innermost part of a strong tropical cyclones center is almost entirely clear of clouds. This region is known as the eye. The eye can be as small as 5 miles (8 km) across, or as large as 50 miles (80 km) or more in diameter. Immediately surrounding the eye is a narrow, circular, rotating region of intense thunderstorms called the eye wall. The eye wall winds thrust up as much as a million tons of air every second. Figure 8.23 shows you the cumulonimbus clouds of the eye wall from the perspective of an airplane flying inside the eye itself! Perhaps the most remarkable aspect of the hurricane is its near-perfect circular pattern of the eye and the surrounding cloudiness, which is especially true for the most intense tropical cyclones. Like a 360 slam dunk in basketball, the whirling circles-within-circles of a strong hurricane are evidence of a rare combination of power and coordination. Next, we learn where the power of tropical cyclones comes from, which is intimately related to where they develop. How and where do they form? Tropical cyclones are essentially large weather engines. Any engine needs energy to run. In Chapter 7 we learned that the unequal heating of different parts of the Earth by the sun drives the huge engine of global weather and climate. But the tropics are another matter, since temperature contrasts there are usually small. Warm air, warm waterit sounds like an advertisement for a vacation cruise, not a prelude to stormy weather! The secret energy source of a tropical cyclone is the large latent heat of water, which we first discussed in Chapter 2. Air over the tropical oceans is drier than you might think. This is because the trade winds are partly made up of air that has descended from above, part of the downward branch of the Hadley Cell we studied in the last chapter. Even though the air and water may both be warm and calm, evaporation can take place because the air is not at 100% relative humidity. Silently and invisibly, water changes from liquid to vapor and enters the atmosphere. The energy required to make this change comes from the sun, and it is lying in waitlatentready to be released when the vapor is condensed into liquid again. This happens in rising air in a cloud or thunderstorm. However, this process is not enough by itself to power a tropical cyclone. A tropical cyclone adds fuel to its own fire by drawing surface air toward its low-pressure center (see Chapter 6). The tight pressure gradient nearer the center means that the winds are stronger and stronger as the air approaches the eye. The faster the wind blows, the more evaporation takes place (this is why you blow-dry your hair or hands instead of merely warming them). Increased evaporation means more water vapor in the air and more energy ready to be liberated in the tropical cyclones thunderstorms as water vapor condenses. Evaporation and condensation of water, in short, are the key to understanding the tropical cyclones power. Figure 8.24 summarizes the atmosphere-ocean interactions that fuel it. How strong is this engine? The energy released due to condensation in a single day in an average hurricane is at least 200 times the entire worlds electrical energy production capacity! Part of this energy goes into reducing the central pressure of the storm and strengthening the winds. Figure 8.25 shows the minimum possible central sea-level pressure for a tropical cyclone based on sea surface temperatures in an average September. Notice how low the pressures can go: below 900 millibars in some regions! Tropical cyclones can create the lowest sea-level pressures observed on Earth, the record being 870 mb about the same surface pressure as at Denver, Colorado, the Mile-High City! Figure 8.25 also shows us where tropical cyclones develop. The very lowest pressures, and therefore the strongest possible tropical cyclones, are located in the regions of warmest ocean temperatures. In regions with SSTs below 26.5C (80F), no numbers for pressures are shown because tropical cyclones cannot form there. This is the fundamental reason why they are tropical cyclones. But why only in the tropics? Because, as we discussed in relation to El Nio, evaporation rates increase very quickly the warmer the temperature is. In other words, a few degrees difference in SSTs can be the difference between no storm, a run-of-the-mill tropical cyclone, and a record-setting hurricane. Some low-latitude regions are not conducive to tropical cyclone development. The far eastern boundaries of the tropical oceans are home to cool ocean currents, as we learned earlier in this chapter. The ocean temperatures are below 26.5C, and therefore tropical cyclones do not form there. Tropical cyclones also do not form within about five degrees of latitude of the equator. The sea surface temperatures there are definitely warm enough for tropical cyclones. But tropical cyclones never form near the equator. Why is this so? Because, as we learned in Chapter 6, the Coriolis force is very small near the equator. This means that air tends to flow straight into low-pressure centers there. The low cannot intensify because it fills up before the pressure can drop very much. How are they structured? Now that we know the power source for tropical cyclones, we can explore their remarkably coordinated structure in more detail. The pattern of winds in a tropical cyclone is fascinating, and we can explain it fairly easily. Figure 8.26 is a time series of wind speeds when powerful Hurricane Celia passed directly over a weather station in Texas. The winds of a tropical cyclone are strong, and consistently strong for many hours. The bulls-eye of isobars around the eye of the tropical cyclone ensures that it is a big blow, as we learned in Chapter 6: a strong pressure gradient means a strong wind. The opposite is true in the eye of the tropical cyclone, as Figure 8.26 reveals. As the eye passes over a location it causes a nearly calm halftime in-between the front and back sides of the storm. Why? The air spiraling into a strong tropical cyclone rises in the eye wall, and it is there that the strong pressure gradient ceases. Instead, air sinks gently from above. We learned earlier that sinking air warms adiabatically and is very stable. Therefore, this is why the eye in a strong storm is cloudless (except for some harmless cumulus and cirrus clouds), calm and warm. Wind shear is the change of wind speed and/or direction as you go up in the atmosphere. Tropical cyclone circulations are disrupted by large amounts of wind shear. Sinking air suggests an anticyclone (Chapter 6). Indeed, one of the hallmarks of a strong tropical cyclone is an intense low at the surface and a high-pressure center or ridge near the tropopause. Generally, the stronger the cyclone, the stronger the upper-level ridge or high. This high is partially caused by the cyclone itself. Without it, the cyclone cannot remove air quickly enough from the eye to keep its pressure low. Therefore, the nearby presence of an upper-level low or trough, or even a high-speed jet stream, is enough to weaken or destroy a tropical cyclone. The weakening in these cases is attributed to wind shear, meaning that the winds are changing fast enough as you go up to disrupt the cyclones own high-building process at those levels. Figure 8.27 is a schematic view of how the winds in a hurricane differ in the lower and upper levels in both growing and weakening cases. The rainbands of a tropical cyclone are the spirals of clouds and thunderstorms that curve into the eye wall and supply the cyclone with its fuel source of moisture. The radar image of Hurricane Hugo in Figure 8.28 illustrates two more important features of tropical cyclones. First, notice that the radar-reflecting rain in this hurricane, like most, is not symmetrically distributed like the winds. Instead, the heaviest rain is contained in spiral rainbands that curve into and blend with the eye wall. This characteristic of tropical cyclones, so fundamental that it is incorporated into the weather-map symbol for them, is still not well understood by meteorologists. This is one of many reasons that hurricane hunters still fly planes into these storms and gather weather data such as that shown in Figure 8.27; Box 8.1 explains the work of hurricane hunters in more detail. The second feature can be seen in the wind measurements from aircraft that are overlaid on the radar image in Figure 8.28. The winds are strongest on the north and east sides of the cyclone. Surprisingly, this is not because the pressure gradient is different on one side of the storm versus another. It is because the storms forward motion adds or subtracts from the winds. In the same way, if you throw a baseball out of a moving car, the baseballs velocity is the combination of the velocity of the ball and the velocity of the car. Figure 8.29 shows that the storms forward speed adds to the wind on the right side of the eye, but subtracts from the wind on the left side of the eye. As a result, the difference in wind on the right versus the left side of the eye is twice the cyclones forward speed. We now know quite a bit about the organization of a tropical cyclone. The fact that these storms are given names (Box 8.2) suggests that, like people, they are unique and have lives just like people. And so next we examine the life cycle of a typical tropical cyclone. What are the different stages of their lives? Tropical cyclones usually begin small, as tropical disturbances. These disturbances are not even low-pressure centers, just a disorganized clump of thunderstorms. In the Atlantic Ocean, these disturbances take the form of easterly waves in the wind patterns that move along with the easterly jet stream across Africa and into the extreme eastern Atlantic. Easterly waves are also seen along the Pacific coast of Mexico; Figure 8.30 shows an example of tropical disturbances and a tropical wave in the Gulf of Mexico and the Pacific. The life cycle of a strong tropical cyclone in the Atlantic follows this pattern: it is born as a disorganized tropical disturbance, probably as an easterly wave in the winds off Africa. Then it develops into a rotating tropical depression, gathers strength and becomes a tropical storm, intensifies further and develops an eye as it becomes a hurricane. Finally, it dissipates over land or cold ocean waters, or else loses its tropical nature and becomes an extratropical cyclone. The vast majority of tropical disturbances die without growing any stronger. However, about 1 in every 10 develop a weak low-pressure center of roughly 1010 mb. Once the low-pressure center is identified and cyclonic rotation is noticed in the winds, the tropical disturbance is called a tropical depression. Some, but not all, tropical depressions gain more strength. Their central pressures drop below about 1000 mb, the pressure gradient between the center and the edge of the storm increases, and the winds strengthen. When the winds in a depression consistently reach 39 mph (about 60 km/hr), a tropical storm is born and given a name (see Box 8.2). Roughly half of tropical storms intensify further. Their central pressures drop below about 990 mb and their winds keep strengthening due to the increasing pressure gradient as shown by the tightening of isobars around the center. An eye forms that can be seen in visible or infrared satellite pictures. When the highest sustained winds in a tropical storm reach 74 mph (120 km/hr), it is reclassified as a hurricane or a typhoon depending on where it is. A few typhoons develop that have winds above 150 mph (240 km/hr) and are called, appropriately enough, super typhoons. After one or two weeks (in rare cases a month), even the strongest of hurricanes and typhoons lose their strength or dissipate. They dissipate because they move over colder water or land, away from their energy source. On a few occasions a hurricane will make the transition from a tropical, warm-water cyclone to an extratropical cyclone (Chapter 10) with cold air and fronts near its center. What does the mature portion of a hurricanes or a typhoons life cycle look like? Figure 8.31 shows the evolution of Hurricane Georges over ten days in September 1998, all superimposed on the same map. Notice how the appearance of the eye changes with time. Meteorologists routinely use the appearance of the eye in satellite images to estimate the strength of hurricanes. Using this figure, we can learn how they do it while studying Georges growth and decay. Figure 8.31 shows an eye is forming in the center of Georges on September 18th. The peak winds in Georges at that time are around 95 mph. This matches our expectations; as we learned earlier, eyes generally form in tropical cyclones when they strengthen to hurricane status. As the storm strengthens to 145 mph on the 20th, the eye becomes better defined and then shrinks into a tight narrow pinhole only a few miles across. This is a fairly good rule of thumb: the stronger the hurricane or typhoon, the smaller the eye and the more nearly circular the storm appears on satellite pictures. Georges eye is hard to identify when the cyclone then moves over land during September 21-25, even over small Caribbean islands. This is because the mountains of these islands, combined with the lack of warm water underneath, disrupt the hurricanes circulation and cause a weakening of winds and the disappearance of the eye. During this period, Georges is strongest on the 22nd, with winds of 110 mph. Not coincidentally, this is when the eye is most easily seen and when Georges has temporarily moved over water again. Judging from the hurricanes appearance, Georges never regains its strength after passing over Cuba. Just a hint of an eye can be made out when Georges crosses the Gulf of Mexico on the 26th and 27th with top winds of 100 mph. In the final image from September 28th, the cyclone no longer has an eye or even a clearly circular shape. It is over land along the coast of Mississippi, and its winds rapidly weaken to below hurricane and even tropical storm status during the next day. Georges, over land and away from warm tropical waters, dies. Meteorologists have recently discovered that the life cycle of a tropical cyclone over water is strongly affected on a day-to-day basis by the sea surface temperatures beneath it. In Figure 8.32, the evolution of Hurricane Floyd in 1999 is shown as a succession of circles whose color and size indicate the maximum sustained wind speeds at different times. Underlying the dots, satellite observations of ocean temperatures are shown for the same period. Floyd increases from a minimal hurricane to one packing winds over 130 mph at the same time that the storm moves over waters that were significantly warmer than the 26.5C threshold for cyclone formation and growth. Later, Floyd weakens as it nears land because it passes over cooler waters to the east and west of the very warm Gulf Stream. These waters cooled because an earlier hurricane, Dennis, stirred up the oceans near the North Carolina coast and brought cooler water to the surface in much the same way as described in the La Nia section of this chapter. Finally, Floyd, like Georges, diminishes to tropical storm status after its circulation leaves the ocean and moves over land. The presence of warm water beneath the storms circulation is a major key to understanding tropical cyclone strength. What does a years worth of tropical cyclones look like? Now that we have a sense of the life cycle of a single tropical cyclone, lets look an entire seasons worth of cyclones to see how different storms evolve at different places and times in the year. Figure 8.33 shows the paths and stages of the different tropical storms and hurricanes that affected the Atlantic Ocean, Gulf of Mexico, and Caribbean Sea in 1995all 19 of them! We can use this modern record-setting year for tropical cyclones to illustrate the seasonal cycle of their development. Hurricane season carefully follows the seasonal cycle of sea surface temperatures. This is not the same as saying hurricanes form in summer! Because it takes time for the oceans to warm up, the Atlantic hurricane season doesnt officially start until June 1 and lasts until November 30. In rare instances, hurricanes can form even in December; the calendar isnt the important thing, the sea surface temperatures and the lack of wind shear are. The peak of hurricane activity in the tropical waters south and southeast of the United States is typically in early-to-mid September. In Figure 8.33, we see that the early-season cyclones tend to form in the Gulf and the western Caribbean. Why is this the case? Largely because these smaller bodies of water are able to warm up more quickly than the Atlantic and reach the magic 26.5C threshold first. A Cape Verde hurricane is an especially dangerous type of tropical cyclone that develops out of an easterly wave near the Cape Verde Islands, which are just off the western coast of Africa. By late August and September, attention shifts to the eastern Atlantic. Around this time, potentially dangerous Cape Verde cyclones develop from easterly waves off the coast of Africa and travel thousands of miles across open water, gathering strength each day. Even though they begin life far from land, Cape Verde cyclones are cause for worry: Approximately 85% of intense hurricanes in the Atlantic begin as easterly waves. In 1995, Luis developed south of the Cape Verde Islands, intensified to hurricane strength by August 30th, and roared northwest and then northeast for the next 12 days before becoming an extratropical cyclone east of Canada! In October, the Atlantic is cooling and the mid-latitude weather patterns cause enough wind shear to reduce the threat of Cape Verde hurricanes. At this time, the still-warm Caribbean and Gulf of Mexico can give birth to late-season cyclones. For example, in 1995 Hurricane Opal developed along the Caribbean coast of the Yucatan Peninsula, wandered into the southern Gulf of Mexico, and then rapidly intensified over especially warm Gulf waters as it moved toward the north-central Gulf Coast on October 4th! Fortunately, it weakened just as quickly, sparing Alabama and the Florida Panhandle from terrible damage. The Bermuda high is a subtropical anticyclone over the Atlantic Ocean in summer that steers tropical cyclones toward the west and northwest. These cyclones often later recurve or change direction toward the northeast, pushed by mid-latitude weather systems. Another aspect of the seasonal cycle of tropical cyclones is seen in Figure 8.33. Notice that the paths of most of the Atlantic storms tend to follow the same arcing track, first moving west or northwest and then recurving toward the northeast. Why do they do that? Tropical cyclones are low-pressure centers near the surface, and they tend to move around the edge of the large surface Bermuda high that sits over the Atlantic in summer. (This high is part of the belt of subtropical anticyclones that are associated with sinking air in the Hadley Cell that we studied in the last chapter.) So, the early-stage paths toward the west or northwest mark the southern and western edges of this high. In 1995, the high did not extend very far west; this was a key reason why this bumper crop of Atlantic tropical cyclones did not cause record-setting damage to the U.S. East Coast. The recurvature of Atlantic tropical cyclones to the north and northeast is partly due to the storms continuing to skirt the edge of the Bermuda high. However, another factor is involved: the wind patterns of the mid-latitudes. By August and September, it is not uncommon to have extratropical cyclones push southeast out of Canada and bring with them strong winds at high altitudes. These jet stream winds help sweep tropical cyclones ahead of them, toward the northeast, while disrupting their circulation and weakening them as the cyclones head out over cold Atlantic waters. Recurving tropical cyclones often accelerate rapidly. The Great New England Hurricane of 1938 roared across Long Island at an estimated forward speed of 60 mph! Based on our earlier discussion of the effect of forward motion on overall storm winds, this means that people to the right of the hurricanes eye in Rhode Island experienced winds 120 mph faster than did luckier residents (including the co-authors father) near New York City who were to the left of the eye. It should be stressed that every hurricane season is different. In 1995 there were 19 hurricanes but none did significant damage. But, 1992 is remembered as a terrible year for hurricanes in the U.S. because Hurricane Andrew ravaged Miami in August 1992. However, Andrew was the only major hurricane that entire year! How do tropical cyclones cause destruction? Tropical cyclones are the Michael Jordan of weather systems: they can beat you using every skill at their disposal. Some cyclones damage and kill with wind; others with seawater; still others with rainwater. A few rare storms combine all three. To preserve life and property, we have to understand how to defend ourselves against such a multi-skilled opponent. Below, we examine this triple threat of tropical cyclones in more detail. Wind The Saffir-Simpson scale is the system by which hurricanes are classified on a scale from 1 (minimal hurricane) to 5 (catastrophic hurricane), based on potential wind and seawater damage. The most obvious threat of a tropical cyclone is the powerful winds associated with its incredibly tight pressure gradient, which as we have seen can blow in a single spot for many hours. Damage is inevitable; a 200 mph wind gust in the most severe cyclones can exert a weight of thirty tons against the wall of a house! Wind damage is such a hallmark of hurricanes that they are classified by meteorologists using the Saffir-Simpson scale, which rates hurricanes on a scale of 1 to 5 based on the damage their winds (and storm surge, see below) would cause upon landfall. Table 8.2 explains the Saffir-Simpson scale and relates the observed damage to estimated wind speeds. Figure 8.34 is a unique before-and-after photograph of wind damage due to Hurricane Andrewa Category 4 hurricane on the Saffir-Simpson scale. Fortunately, most hurricanes do not reach the 150+ mph winds linked with the highest category on the Saffir-Simpson scale. In the past century, only three of these Cat 5 hurricanes have crossed the United States coastline: the Labor Day Storm of 1935 in Florida, Hurricane Camille in Mississippi in 1969, and Hurricane Allen at the southern tip of Texas in 1980. While Camille caused major damage along the Gulf Coast, the other two hurricanes caused much less damage. Why can the damage due to similarly intense storms differ so much? The Labor Day storm was intense but very small; it caused extensive damage in the Florida Keys but Miami was left unscathed. Hurricane Allen diminished in intensity just prior to landfall and struck one of the few stretches of the Gulf Coast that was not covered with expensive beachfront buildings. The amount of damage depends not only on the size and strength of the storm, but also on what is in its way! This is particularly relevant to the United States, where over 50% of the population will soon live within 50 miles of the ocean. This boom in coastline development means that damage due to tropical cyclones will continue to increase, even as meteorologists learn more about the storms and how to protect against them. To add insult to injury, hurricanes also contain smaller whirlwinds inside them that can cause additional damage. Upon landfall, nearly every hurricane produces at least one tornado (Chapter 11) that wreaks localized havoc in a narrow path for a few miles at most. In Hurricane Andrew, meteorologists identified tornado-sized mini-swirls that tacked on more miles per hour to Andrews already powerful winds. These mini-swirls helped cause the worst damage due to Andrew, as seen in Figure 8.21. Houses just a few blocks away escaped with much less damage, due to the local nature of the mini-swirls. Seawater Storm surge is a wall of seawater pushed by tropical cyclone winds onto land that causes most storm-related deaths. The winds in a tropical cyclone push ocean water along with them. The stronger the wind, the more water is piled up by the winds (Figure 8.35). As the cyclone nears shore, the winds to the right of the eye, which blow onshore, push this water inland. This dome of water is usually about 50 miles wide, causing massive flooding near and to the right of the eye where it makes landfall. This process of wind-induced flooding by seawater is called storm surge. It is extremely rapid; in Galveston in 1900, two different observers reported that the sea level rose four feet in just a few seconds. Storm surge is by far the deadliest weapon in the tropical cyclones arsenal, causing as much as 90% of all hurricane-related deaths. In short, this is because water is heavier than air. The brute force of even a small storm surge dwarfs the strongest wind gust. Anyone dragged underwater then is at great risk of drowning. Hundreds of thousands of people in low-lying Bangladesh have died by drowning in storm surges. The 8000 or more deaths in Galveston, TX in 1900 were primarily due to the 15-foot storm surge there (see Figure 8.36). Thanks to warnings, Hurricane Andrews 17-foot storm surge was not a major killerbut it did severely damage the world headquarters of Burger King, located on the coast just south of Miami. Rainwater A tropical cyclone that has moved over land may seem to be dying, but it still has one last knockout punch left: flooding due to heavy rains. Tropical cyclones, even relatively weak storms that never attain hurricane status, are capable of record-setting amounts of rainfall over land. In 1979, weak Tropical Storm Claudette dropped over 40 inches of rain in a single day in southern Texas. The rain was so heavy that rain gauges were too small to use; open garbage cans were later used to quantify the deluge! Hurricane Camille is said to have dumped 30 inches of rain in six hours on the James River Valley of Virginia three days after its 200 mph winds raked Mississippi; the resulting floods and mudslides killed 109 people. More recently, Hurricane Floyd (see Figure 8.32) illustrates the damage possible due to a tropical cyclones rainwater. By the time Floyd reached the coastline of North Carolina, Floyds winds had weakened to Category 2 on the Saffir-Simpson scale. Its ability to do wind-related damage had diminished, but its capacity to cause destruction with flooding rains had not. The floods that ensued along the U.S. East Coast (Figure 8.37) accounted for the majority of Floyds death toll of 57more than due to the stronger Andrew, and the most in the U.S. since another flood-producer, Agnes in Pennsylvania in 1972. Damage due to Floyd was estimated at $6 billion. Tropical cyclones are part of the overall pageantry of weather, however, and they do good as well as harm. Their winds blow down forests and make way for new growth. Storm surges carve new channels and cleanse wetlands. The rainfall from many a tropical storm, or even a wave or disturbance, is enough to save crops from summertime and autumn droughts. From a human perspective, however, tropical cyclones are generally far too much of a good thing in all respects, and must be anticipated and avoided if at all possible. We explore the forecasting of tropical cyclones below. How do we observe and forecast tropical cyclones? Past, present, and future A Category 4 Cape Verde hurricane bears down on a major American city from the open seas. In Galveston in 1900, almost 10,000 people die as a result. In Miami in 1992, the storm directly kills only 15. The enormous difference in death tolls is explained by the vast improvements in hurricane detection, forecasting and warning made during the 20th century. In 1900, weather satellites did not exist, the global network of weather observations was in its infancy, and countries did not routinely share weather data. The U.S. Weather Bureau was even afraid to use the word hurricane in its statements, to avoid undue concern! In the case of the Galveston hurricane (Figure 8.38), the Weather Bureau ignored reports from Cuban meteorologists and expected the storm to recurve as usual to the northeast along the U.S. East Coast. Assumption became fact as the official government reports stated, wrongly, that the storm was traveling northeast in the Atlantic. Instead, a high-pressure system shunted the hurricane far to the west through the Gulf of Mexico. The hurricane did not recurve until it hit an unprepared Galveston. Even Galvestons chief Weather Bureau meteorologist was forced to ride out the storm in his floating house, so unexpected was the storms fury. Chicago was surprised by hurricane-force wind gusts by the recurving storm! Failed forecasts of fair, fresh weather based on a lack of scientific understanding and observations cost thousands of lives. In 1992, Hurricane Andrew was tracked from its earliest stages by weather satellitethe biggest single advance in hurricane detection and forecasting in the 20th century. When Andrew made a sudden course change and intensified rapidly, the National Hurricane Center (NHC) in Miami knew it about it within hours and immediately changed their expectations for when and where the storm would make landfall, and how strong it might be when it did. Hurricane hunter aircraft flew into the storm to get precise observations of the storm more accurate than satellite-based estimates. Hurricane watches were issued for southern Florida, as they always are when a hurricane may threaten a region within the next 36 hours. The watches were upgraded to warnings when it became likely that Andrew would hit the Miami vicinity within 24 hours. All of this information was widely disseminated to the public through government and media sources (Box 8.3). Finally, weather radar followed the storm all the way into NHCs backyard of Miami. Figure 8.39 is the last scan of NHCs radar before Andrews 164 mph winds blew the radar dish off the buildings roof! This amazing level of detection and surveillance is conducted for each and every hurricane that threatens the United States. For this reason, deaths due to hurricanes have decreased drastically in the U.S. in the past century. Accurate observation of each tropical cyclone is one important advance in forecasting. Recently, computer models (see Chapter 13) of hurricanes have improved enough that they are used to anticipate, several hours or a day in advance, what a particular hurricane will do. These models are still not good at capturing the quirky changes in direction and intensity of hurricanes, but are getting better each year. The most exciting advance in tropical cyclone prediction is based on an intelligent understanding of the structure of these storms. During the 1980s and 1990s, Colorado State University meteorologist Prof. Gray developed and refined a method for forecasting tropical cyclones in a statistical, seasonal sense. His method wouldnt have told you if Andrew would hit Miami on August 24th, but it will tell you up to a year ahead of time if the next hurricane season will probably see more than its usual share of hurricanes in the Atlantic. Prof. Grays forecasts are based primarily on the El Nio cycle, rainfall in the Sahel region of Africa, and the direction of winds in the lower stratosphere. A novice to tropical cyclones might wonder why these parameters could help one predict the number of storms in a given year. Based on our discussion of atmosphere-ocean interactions, they make sense. Earlier in this chapter we saw that El Nio leads to a strong jet stream over the subtropics. This jet shreds the carefully organized circulation of hurricanes. On average, then, El Nio years are low-hurricane years, and La Nia years are high-hurricane years. High rainfall in the Sahel means lots of thunderstorm seedlings that move east and develop into easterly waves and, sometimes, into Cape Verde hurricanes. Finally, strong winds in the lower stratosphere, like upper-tropospheric jet streams, lead to wind shear that disrupts the upper-level winds of a growing hurricane. The accuracy of Prof. Grays forecasts is demonstrated in Figure 8.40. In about 7 out of every 10 years, his methods can accurately predict whether a given hurricane season will be more active or less active than normal. This is not the same as predicting a specific hurricane in a specific location in advance, a task that 21st century meteorologists must tackle. However, it is a vast improvement upon expectations based on climatological records, and is helping coastal residents as well as businesses and insurance companies prepare months in advance for the chance of damaging storms. As we will see later in Chapter 13, these kinds of long-range statistical predictions are the future of weather forecasting. Summary Based on temperature, the vertical structure of the can be classified into three basic layers, the surface zone, a deep-water zone and a transition zone. In the bottom layer, or the deep zone, the temperature is uniform with depth. The surface zone is the warmest layer, and is warmest in the tropical waters. Surface water temperature decreases towards the poles. The atmosphere and ocean interact through exchanges of heat and moisture. The oceans gain energy during the summer and losses energy during winter. Surface currents are driven by the winds. Some, like the Gulf Stream, move in well-defined boundaries like a river, while others are broad and diffuse. Water circulates below the surface in the deep zone. These deep water currents are driven by density differences. Surface currents generated by the winds can induce vertical circulations in the oceans. This can induce upwelling of nutrient rich waters. For example, over the eastern tropical Pacific Ocean, winds rotate counter-clockwise around a high-pressure system producing along-coast winds off Peru and easterly winds along the equator in the eastern Pacific. The Coriolis effect produces offshore flow off Peru and surface divergence at the equator causing upwelling where cold, deeper water comes to the surface. This produces the cold sea surface temperatures (SSTs) off Peru and along the equator in the eastern Pacific. In the western Pacific on the equator, this upwelling does not occur, so that the SST is higher than in the east. El Nio and La Nia are extreme phases of a naturally occurring climate cycle that refer to large-scale changes in sea surface temperature across the tropical Pacific. El Nio and La Nia result from interaction between the surface of the ocean and the atmosphere in the tropical Pacific. Changes in the ocean impact the atmosphere and climate patterns around the globe. Both El Nio and La Nia impact global climate patterns. In many locations, especially in the tropics, La Nia (or cold episodes) produces the opposite climate variations from El Nio. For instance, parts of Australia and Indonesia are prone to drought during El Nio, but are typically wetter than normal during La Nia. Tropical cyclones are large whirling storms that obtain their energy from warm ocean waters. They stand out on satellite photographs due to their circular cloud patterns and, in the stronger storms, a nearly clear eye at the center. The clarity and size of the eye on satellite images helps meteorologists estimate a cyclones strength. A tropical cyclone begins as a disorganized tropical disturbance. A few grow to hurricane or typhoon strength with winds up to 200 mph. Most weaken within a week or two. They typically move west or northwest and then recurve toward the northeast, but each storm is unique. Hurricanes affecting the United States often form in the Gulf of Mexico and Caribbean Sea in early summer and late fall, with powerful Cape Verde hurricanes dominating attention in August and September. The location of warm ocean waters strongly determines the birthplaces and intensities of tropical cyclones. Tropical cyclones cause destruction with extremely high winds, storm surges of seawater, and flooding rainfall. Of these, storm surge is the most deadly and devastating. Each hurricane packs its own unique combination of these weapons. Weather satellites have revolutionized the forecasting of tropical cyclones. Today no tropical cyclones go undetected, a vast improvement upon the situation a century ago. Thousands of lives are saved as a result. Statistical forecasts now provide long-range estimates of the activity of hurricane seasons as much as a year in advance. Terminology You should understand all of the following terms. Use the glossary and this Chapter to improve your understanding of these terms. Bermuda high Cape Verde hurricane Deep Zone Easterly waves Ekman spiral El Nio Extratropical cyclone Eye Eye wall Gyre Halocline Hurricanes La Nia Ocean current Oceanography Pycnocline Rainbands Recurvature Saffir-Simpson scale Storm surge Supertyphoon Surface zone Thermocline Thermohaline circulation Tropical cyclones Tropical depression Tropical disturbance Tropical storm Typhoons Upwelling Review Questions Why is the ocean surface zone thicker over the tropics than the poles? Why is it important for weather to know the distribution of sea surface temperature? What is the thermocline? What is the North Atlantic gyre? What is the Ekman spiral and how is it formed? How can surface winds generate upwelling along coastal regions? What is an El Nio and how does it impact the weather where you live? What is an La Nia and how does it impact the weather where you live? Is a typical tropical cyclone bigger or smaller than the following weather phenomena: Hadley Cell; air parcel; thunderstorm; Southern Oscillation? When a tropical cyclone opens its eye for the first time, what does that tell you about its approximate wind speed? Is it strengthening or weakening? When a hurricane narrows its eye and becomes very circular, is it strengthening or weakening? If you could cover the Gulf of Mexico with water-impermeable plastic wrap, how would this affect the development and intensity of tropical cyclones over the Gulf? In some El Nio years, the ocean waters off the west coast of Mexico also become abnormally warm and tropical fish are sighted unusually far north. How might this affect the intensity of tropical cyclones that develop in this region? Only one tropical cyclone has developed in the South Atlantic Ocean off the coasts of South America and Africa in the last century. What ocean temperature and wind patterns might be responsible for this amazing lack of tropical cyclone activity? Why do hurricanes never form off the West Coast of the United States, but can form off the East Coast? Global warming, if occurring, should lead to warmer ocean temperatures, less wind shear due to a reduced latitudinal temperature gradient, but perhaps more wind shear due to perpetual El Nio conditions. Discuss how each of these three changes might affect future tropical cyclone activity and intensity. In 1985, Hurricane Gloria recurved northeastward along the U.S. East Coast and traveled quickly just east of the coastline for hundreds of miles. East Coast residents braced for a fearful storm; a Boston radio station even played a marathon of songs titled Gloria in anticipation. But in the end, most inland residents were disappointed, saying that the winds were much less than they had expected. Using a U.S. map and your understanding of the relationship between total storm winds and the forward motion of the storm, explain why the inland residents experienced relatively weak winds from this strong hurricane. If you lived or ran a business on the U.S. East Coast and you heard that Bill Grays forecast was for a very active and damaging hurricane season in the near future, what actions would you take to protect your life and property? True or false: in a hurricane, you want to go to a location as low as possible in order to stay out of the high winds. Explain your answer. Why dont the damage totals and death tolls from U.S. hurricanes always increase with increasing Saffir-Simpson category number? Discuss the different variables that can make a Category 1 storm more of a killer and destroyer of property than a Category 5 storm. Can tropical cyclones exist on other planets in our solar system? Explain why or why not. Web Activities Web activities related to subjects in the book are marked with superscript (. Activities: There is a movie of Hurricane Danny radar loops created by Pete Pokrandt at the University of Wisconsin that would be wonderful for our students to view. Its 18 megabytes; I will place it on the Web site so that Steve can access it. It shows Danny slowly working its way to the Gulf Coast, within range of coastal radars for a period of days. This is precisely the kind of visualization that students need to get a good feel for what tropical cyclones look like in action. We should have a link to Chris Landseas Tropical FAQ Web site, at  HYPERLINK http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqHED.html http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqHED.html It would be nice to have an interactive Java applet that would allow students to make your own hurricane by specifying central pressure, size of storm, size of eye, etc. and then plotting the winds. Ideally, it could be a kind of video game where you make the cyclone and then specify the upper-level winds. Will the storm grow or die? Will it ram into New York City? And so forth. Can Tom do this? Box 8.1 The Hurricane Hunters Why would anyone in their right mind intentionally fly into the middle of a hurricane? During every hurricane season, U.S. Air Force (USAF) and National Oceanographic and Atmospheric Administration (NOAA) pilots and scientists do just that, time and again. Their goal: to obtain up-to-the-minute weather data on storms that cannot be obtained by any other means. They are the hurricane hunters. The first exploration of a major hurricane by a reconnaissance aircraft, as they are called, occurred in 1944 during the Great Atlantic Hurricane. Thousands of flights later, no plane has ever been lost while penetrating these deadly storms. (There was a pretty close call during a flight into Hurricane Hugo in 1989, however.) The pilots carefully choose their altitude (high, away from the roiling ocean) and path (straight through to the eye, no dilly-dallying in the eye wall) to maximize their safety while gathering crucial weather data. Today, hurricane hunting is a highly coordinated effort between the National Hurricane Center, USAF and NOAA. When NHC needs to know if a tropical disturbance far out in the Atlantic is getting better organized, the USAF pilots fly long missions to search for signs of a closed circulation. They use weather instruments that have been specially installed on their large cargo airplanes. In addition, the planes launch dropsondesthe reverse of radiosondeswhich drop out of the plane and into the storm from above. These dropsondes can tell what the weather conditions are like in the very center of the disturbance, below where the planes normally fly. Finally, the pilots and scientists use their eyes to see, literally, if the storm clouds are becoming better organized. This information is relayed to NHC, where meteorologists use it to decide whether or not a disturbance has achieved tropical depression or tropical storm status. If a mature hurricane comes closer to the United States, the NOAA hurricane hunters take over. NOAAs specially instrumented P-3 aircraft (see figure) includes on-board radars. Figure 8.28 is a radar image from one of the NOAA flights into Hugo. The measurements of winds and location of the hurricanes center, as shown on Figure 8.28, are extremely important pieces of information for hurricane forecasters. They are much more accurate than satellite estimates. For this reason, hurricane hunters will continue to brave the bumpy flights into these storms for the foreseeable future. There is also an added bonus: hurricane hunters often say that a strong hurricanes eye wall, viewed from an airplane circling in its eye (see Figure 8.22) is one of the most beautiful and awe-inspiring sights in all of nature. Box 8.2 Naming Hurricanes Hurricanes come in large and small sizes, some with big eyes, some with squinty eyes, and they are born, grow, and then die. Its not surprising, then, that human names have been assigned to hurricanes for centuries. The very earliest instance of hurricane naming was in the Caribbean islands, where hurricanes were associated with the Roman Catholic saints holiday when they hit the islands. For example, Hurricane Santa Ana struck Puerto Rico on July 26, 1825. The first known scientific use of hurricane naming arose in the Pacific during World War II. It was an easy and effective way to distinguish one tropical cyclone from another on the weather maps. The system was simple and alphabetical: the name of the first storm of the season would begin with A; the second, B; the third, C; and so on. Naming hurricanes also gave homesick soldiers a way to recall loved ones. Thus began the practice by predominantly male meteorologists of giving female names to hurricanes. This practice persisted at the National Hurricane Center in Miami until the late 1970s. One retired NHC meteorologist ironically commented, It is surely only a coincidence that many names of NHC employees, their wives, and other female relatives appear on the [name] list. This may explain the distinctly Southern flavor of the names of past hurricanes. Beginning in 1979, the list of names for each years tropical storms and hurricanes was broadened to include both male and Spanish and French names. The alphabetical naming system continues, now overseen by the World Meteorological Organization (WMO). The WMO composes different alphabetical lists for different ocean basins (for example, starting in 2000 northwest Pacific typhoons now have Asian names). The names of severe and/or notable hurricanes are retired and taken off the lists. Horror sequels are popular in the movies, but no meteorologist wants to tempt fate by naming a new hurricane Andrew, Georges, or Mitch, to cite a few of the recently retired names. Below is the list of names for Atlantic Ocean, Gulf of Mexico, and Caribbean Sea tropical storms and hurricanes for the years 2001-2006. You may be wondering: what happens if there are more storms than names? That has never happened in the Atlanticalthough in 1995 it was a close call, with Tanya lasting until November 1. 200120022003200420052006AllisonArthurAnaAlexArleneAlbertoBarryBerthaBillBonnieBretBerylChantalCristobalClaudetteCharleyCindyChrisDeanDollyDannyDanielleDennisDebbyErinEdouardErikaEarlEmilyErnestoFelixFayFabianFrancesFranklinFlorenceGabrielleGustavGraceGastonGertGordonHumbertoHannaHenriHermineHarveyHeleneIrisIsidoreIsabelIvanIreneIsaacJerryJosephineJuanJeanneJoseJoyceKarenKyleKateKarlKatrinaKeithLorenzoLiliLarryLisaLeeLeslieMichelleMarcoMindyMatthewMariaMichaelNoelNanaNicholasNicoleNateNadineOlgaOmarOdetteOttoOpheliaOscarPabloPalomaPeterPaulaPhilippePattyRebekahReneRoseRichardRitaRafaelSebastienSallySamSharyStanSandyTanyaTeddyTeresaTomasTammyTonyVanVickyVictorVirginieVinceValerieWendyWilfredWandaWalterWilmaWilliam Box 8.3 Bryan Norcross, TV Meteorologist and Hero When most people think of meteorologists, they think of TV weather personalities. This is a misconception, because the majority of meteorologists are not on TV, and many TV weathercasters are not meteorologists. Most TV weathercasters, meteorologists or not, dont make their own forecasts. They rely instead on the official National Weather Service forecasts. This doesnt mean that TV weatherpersons are not serious about forecasting however. They serve a great purpose by using their wide exposure and their personalities to convey important weather information to the public. One of the best examples is Miami TV meteorologist Bryan Norcross. When Hurricane Andrew blasted Miami, Norcross became a national hero. Norcross (right) is part newsman, part meteorologist. He has a bachelors degree in math and physics and a masters degree in communications and meteorology. Prior to coming to Miami, he worked as a news producer in Denver and a news director in Louisville. He also oversaw reporting on political campaigns and the Olympics. In Miami, Norcross returned to TV meteorology, a position he had held previously at CNN. His background in news reporting helped him translate complicated aspects of meteorology to a wide audience. In November 1991 he even appeared on national TV in a two-part episode of the sitcom Golden Girls! The plot: a terrible hurricane hits Miami, with Norcross spreading the news. Life imitated TV just a few months later, on August 23, 1992. Hurricane Andrew was bearing down on a city that had not experienced a hurricane in nearly thirty years. Untold thousands of residents, newcomers to the region, had no idea what to expect or how to protect themselves. Worse yet, Andrew came ashore just south of Miami shortly after 5 am, when most sources of information were off the air, either intentionally or unintentionally due to the storms winds. Into this breach stepped Bryan Norcross. Norcross provided frequent updates on the predicted path and severity of Andrew throughout the evening of the 23rd, and then stayed on all night with continuous coverage of the storm. He didnt make the forecasts; the National Hurricane Center did. But Norcross did broadcast these forecasts with live interviews with the NHCs director. He also provided numerous safety tips to residents who had little knowledge of how to protect themselves from 150 mph wind gusts. To emphasize the risks, at 5 am Norcross and his overnight crew took shelter in their own studio, broadcasting from a closet. A resident from a hard-hit section of Miami called him, on the air, and said her roof was coming off. What should she do? Norcross replied, get away from all the windows, crouch inside a bathtub and pull a mattress over her head. Now. For his efforts during the Hurricane Andrew crisis, Bryan Norcross was given national journalism and meteorology awards. (He even was the subject of a 1993 TV-movie, which won no awards whatsoever.) Norcross is one of many TV weather personalities who save lives with timely, useful information. Table 8. 1. The most damaging hurricanes to affect the United States since 1900. Dollar amounts are normalized to 1998 standards. Notice that earlier storms caused many more deaths than comparable, but more recent, storms. This is a measure of the effectiveness of todays hurricane detection and warning systems. In some cases, intense hurricanes caused little damage and few deaths because sparsely populated regions were hit. See text for details on the Saffir-Simpson scale. Adapted from information provided by Dr. Chris Landsea, Atlantic Oceanographic and Meteorological Laboratory. RANKHURRICANEYEARMINIMUM PRESSURE AT LANDFALL (mb)SAFFIR-SIMPSON CATEGORY AT LANDFALL DEATHS (U.S.) DAMAGE (1998 U.S. DOLLARS)1SE Florida/Alabama1926935424383,814,000,0002ANDREW (SE FL/LA)199292245838,362,000,0003N Texas (Galveston)190093148000+30,856,000,0004N Texas (Galveston)1915945427526,144,000,0005SW Florida1944962330019,549,000,0006New England1938946360019,275,000,0007SE Florida/Lake Okeechobee19289294 183615,991,000,0008BETSY (SE FL/LA)196594837514,413,000,0009DONNA (FL/Eastern U.S.)196093045013,967,000,00010CAMILLE (MS/LA/VA)1969909525612,711,000,00011AGNES (NW FL, NE U.S.)1972980112212,408,000,00012DIANE (NE U.S.)1955987118411,861,000,00013HUGO (SC)198993442610,872,000,00014CAROL (NE U.S.)195496036010,509,000,00015SE Florida/Louisiana/Alabama19479404 519,630,000,00016CARLA (N & Central TX)19619314468,194,000,00017HAZEL (SC/NC)19549384958,160,000,00018NE U.S194494733907,490,000,00019SE Florida1945951347,318,000,00020FREDERIC (AL/MS)1979946357,295,000,00021SE Florida1949954326,767,000,00022S Texas19199274600+6,200,000,00023ALICIA (N TX)19839673214,702,000,00024FLOYD (NC)19999562564,500,000,00025CELIA (S TX)19709453113,869,000,00026DORA (NE FL)1964966253,603,000,00027FRAN (NC)19969613373,591,000,00028OPAL (NW FL/AL)1995942393,478,000,00029GEORGES (SW FL/MS)1998965243,073,000,00030CLEO (SE FL)1964971202,823,000,000 Table 8.2. The Saffir-Simpson scale for hurricanes, named for the engineer and the meteorologist who devised it in the 1970s. Adapted from information provided by Dr. Chris Landsea, Atlantic Oceanographic and Meteorological Laboratory. [Probably works better as one long table with the damage descriptions as one more column.] Saffir-Simpson ScaleSaffir-Simpson CategoryMaximum sustained wind speedMinimum surface pressureStorm surgemphm/sktsmbftm174-9533-4264-82greater than 9803-51.0-1.7296-11043-4983-95979-9656-81.8-2.63111-13050-5896-113964-9459-122.7-3.84131-15559-69114-135944-92013-183.9-5.65156+70+136+less than 92019+5.7+ DamageCategoryLevelDescriptionExample1 MINIMAL Damage primarily to shrubbery, trees, foliage, and unanchored homes. No real damage to other structures. Some damage to poorly constructed signs. Low-lying coastal roads inundated, minor pier damage, some small craft in exposed anchorage torn from moorings. Hurricane Agnes (1972) 2 MODERATE Considerable damage to shrubbery and tree foliage; some trees blown down. Major damage to exposed mobile homes. Extensive damage to poorly constructed signs. Some damage to roofing materials of buildings; some window and door damage. No major damage to buildings. Coast roads and low-lying escape routes inland cut by rising water 2 to 4 hours before arrival of hurricane center. Considerable damage to piers. Marinas flooded. Small craft in unprotected anchorages torn from moorings. Evacuation of some shoreline residences and low-lying areas required. Hurricane Georges (1998) 3 EXTENSIVE Foliage torn from trees; large trees blown down. Practically all poorly constructed signs blown down. Some damage to roofing materials of buildings; some wind and door damage. Some structural damage to small buildings. Mobile homes destroyed. Serious flooding at coast and many smaller structures near coast destroyed; larger structures near coast damaged by battering waves and floating debris. Low-lying escape routes inland cut by rising water 3 to 5 hours before hurricane center arrives. Flat terrain 5 feet of less above sea level flooded inland 8 miles or more. Evacuation of low- lying residences within several blocks of shoreline possibly required. Hurricane Celia (1970) 4 EXTREME Shrubs and trees blown down; all signs down. Extensive damage to roofing materials, windows and doors. Complete failures of roofs on many small residences. Complete destruction of mobile homes. Flat terrain 10 feet of less above sea level flooded inland as far as 6 miles. Major damage to lower floors of structures near shore due to flooding and battering by waves and floating debris. Low-lying escape routes inland cut by rising water 3 to 5 hours before hurricane center arrives. Major erosion of beaches. Massive evacuation of all residences within 500 yards of shore possibly required, and of single- story residences within 2 miles of shore. Hurricane Andrew (1992) 5 CATASTROPHIC Shrubs and trees blown down; considerable damage to roofs of buildings; all signs down. Very severe and extensive damage to windows and doors. Complete failure of roofs on many residences and industrial buildings. Extensive shattering of glass in windows and doors. Some complete building failures. Small buildings overturned or blown away. Complete destruction of mobile homes. Major damage to lower floors of all structures less than 15 feet above sea level within 500 yards of shore. Low-lying escape routes inland cut by rising water 3 to 5 hours before hurricane center arrives. Massive evacuation of residential areas on low ground within 5 to 10 miles of shore possibly required. Hurricane Camille (1969)  PAGE 8-7 PAGE 8-17   d s  6A%%')()++1,//22!3"34445o5h6i6;;;9<B<@@@@@@e*eeef ff|ffffg#giijjjbjbybScef8hijý~{xuro2ZR  ¡ S :  S ?W  x/'p58;;;I?]?@ACJCZCDHgKrNNPTW|[\\\_>b & F & F&dw$d%d&d'd/+D1>bjbybScef8hijjHk[n2pNpru8x}~Ā݀^&d`l$d%d&d'd/+D1&d`$d%d&d'd/+D1$ & F & FijjHk[n2pNpru8x}~Ā݀^<,]0`q֖6{ ۝0@Ķzwrmhc^YTAh;   dg=fz   Ĥ`/  !   ^<,]0`q֖6{ ۝0@c&d`$d%d&d'd/+D1&d`S$d%d&d'd/+D1&d` $d%d&d'd/+D1v+É\|ˑ*>O]̔ڔՖ Йٙߙiwћ') IKڤ.0FNw(4߸jn%?J: H*50J50J6`@c[ԭ2`;{ej&^9Y;s}ZĿzlid_QL   &~    ,?  ã   g D)x   -|k1c[ԭ2`;{ej&&dL$d%d&d'd/+D1&d`($d%d&d'd/+D1&d`<$d%d&d'd/+D1&^9Y;s}Z;VYw-s  & F&d`c$d%d&d'd/+D1Z;VYw-s I6)>HWdlſ}zwtqnkheb_q|   !drS  K   ML~%suwWcN#*    HIJz{57\i %$%%%)%*%.%/%3%4%8%9%=%>%F%G%M%N%Q%R%V%W%]%^%e%f%l%m%s%t%x%y%%B*CJCJjUmHnHCJ560JjisCJU jU j>0J556H*OI6)>HWdl$$)5N`tZsI"ŷyqiaY        d        :  S      O   n oy -9CN[i )5N`tZsI"$ & F$$ =   $     }67sǿ|wrmhc^YTO HIjl      }           p          =   $     }67s$$$d%d&d'd$ & F $$$ & F!6!# %%%*%/%4%9%>% $d$$$$$d%d&d'd!6!# %%%*%/%4%9%>%?%G%N%R%W%^%f%g%m%t%y%%%%%%%%%%%Ŀ{wrmhc^Y                          !  (  -  1  8  @A  F  K  P  U  Z  _I^">%?%G%N%R%W%^%f%g%m%t%y%%%%%%% $d$[$$T    ֈ~ %}%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%&& & &&&&&&&$&%&-&.&6&7&A&B&H&I&N&O&U&V&Z&[&a&b&k&l&q&r&w&x&&&&&&&&&&&&&&&&&&&&&&&&&&&&CJB*CJb%%%%%%%%%%%%%%%%%&[$$T    ֈ~ %} $d$%%%%%%%%%%%%%& & &&&&%&.&7&8&B&I&O&V&[&b&c&l&r&x&&&|wsnid_Z            $  )  0  6  =  GH  Q  Z  b  i  m  st  |                      "& & &&&&%&.&7&8&B&I&O&V&[&b&c&l&[$$T    ֈ~ %} $d$l&r&x&&&&&&&&&&&&&&&&[$$T    ֈ~ %} $d$&&&&&&&&&&&&&&&&&&&&&&'' '''''$'%'.'4':'B'|wrnid_Z  E  K  Q  Z[  b  f  k  q  v  ~                                      "&&&&&&&&&'' '''''$'%'[$$T    ֈ~ %} $d$&&&&&&&&&&&&&&&'' ' ''''''''#'$'-'.'3'4'9':'A'B'G'H'O'P'U'V'Z'['c'd'j'k'o'p'v'w'|'}'''''''''''''''''''''''''''''''''''''''(((( ( ((CJB*CJb%'.'4':'B'H'P'Q'V'['d'k'p'w'x'}'''[$$T    ֈ~ %} $d$B'H'P'Q'V'['d'k'p'w'x'}'''''''''''''''''''''''(|wrmid_Z                                                  $  )  ./  7  ="''''''''''''''''''[$$T    ֈ~ %} $d$''''(( (((((&(,(2(7(8(<(B( $d$[$$T    ֈ~ %}(( (((((&(,(2(7(8(<(B(I(R(X(`(a(g(o(u(|((((((((+\.^03%7|wrmhc^Y!#              '  -  6  =  C  GH  M  S  Y  `  f  lm  s  x  ~"((((((%(&(+(,(1(2(6(7(;(<(A(B(H(I(Q(R(W(X(_(`(f(g(n(o(t(u({(|((((((((++0033'7+7,75767:7;7\7]77777777777777777777777888888*8+8/80838485868J86H*jUmHnHCJCJB*CJ\B(I(R(X(`(a(g(o(u(|((((((($$d%d&d'd$$[$$T    ֈ~ %} $d$((+\.^03%7'7,767;7L7O7X7]77777777 $d$$$%7'7,767;7L7O7X7]77777777777777777777888888+8Ŀ{vqlgc^Y  h  jk  z  }                                      "  '  0  3  D  I  S  XZ"77777777777778888 $d$f$$T-    ֞ lH!888+8084868<8K8 $d$f$$T-    ֞ lH!+8084868<8K8L8N8b8g8k8m8q88888888888888888888889Ŀ{vqmhc^Y                                                1  34  C  I  K  O  T"J8K8M8N8a8b8f8g8j8k8l8m88888888888888888888888888888888889999999*9+9/90939495969G9H9J9K9b9c9g9h9k9l9m9n9999999999999999999999999999CJB*CJbK8L8N8b8g8k8m8q8888888888 $d$f$$T-    ֞ lH!88888888888889999 $d$f$$T-    ֞ lH!9999999+909496999H9I9K9c9h9l9n9q9999999999999999{vrmhc^Y                                    4  67  F  I  K  O  T  e  gh  w  |  }  "9999+909496999H9I9K9c9h9l9n9q9f$$T-    ֞ lH! $d$q99999999999999999f$$T-    ֞ lH! $d$99999:: : :: :!:$:.:3:7:9:<:K:L:O:_:d:h:j:m:|:}::::::::|wrmhc^Y                          0  34  C  F  H  L  Q  [  ^_  n  r  t  x  }        "9999:: : :: :!:$:.:3:7:9:<:f$$T-    ֞ lH! $d$99:::: : : : :: :#:$:-:.:2:3:6:7:8:9:J:K:N:O:^:_:c:d:g:h:i:j:{:|:::::::::::::::::::::::::::;;;; ; ; ;;;;";#;);*;.;/;2;3;4;5;F;G;J;K;U;V;Z;[;^;_;`;a;p;q;t;u;;CJB*CJb<:K:L:O:_:d:h:j:m:|:}:::::::f$$T-    ֞ lH! $d$:::::::::::::;; ;;f$$T-    ֞ lH! $d$::::::::::::;; ;;;; ;#;*;/;3;5;9;G;H;K;V;[;_;a;c;q;r;|wrmhc^Z        $  )  4  78  F  J  L  P  U  \  _`  n  q  s  w  |                    ";;; ;#;*;/;3;5;9;G;H;K;V;[;_;a;f$$T-    ֞ lH! $d$a;c;q;r;u;;;;;;;;;;;;;f$$T-    ֞ lH! $d$r;u;;;;;;;;;;;;;;;;;;;;;;;;;< <<<<#<$<'<2<Ŀ{vqlgc^Y  X  [\  j  m  o  s  x                                               ";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;<< < <<<<<"<#<&<'<1<2<6<7<:<;<<<=<M<N<Q<R<^<_<c<d<g<h<i<j<z<{<~<<<<<<<<<<<<<<<<<<<<<<<<<<<<<CJB*CJb;;;;;;;;;;;;;< <<<f$$T-    ֞ lH! $d$<<#<$<'<2<7<;<=<@<N<O<R<_<d<h<j<f$$T-    ֞ lH! $d$2<7<;<=<@<N<O<R<_<d<h<j<m<{<|<<<<<<<<<<<<<<<<<<<<<Ŀ{vqmhc^Y                                                -  01  ?  B  D  H  M"j<m<{<|<<<<<<<<<<<<<<f$$T-    ֞ lH! $d$<<<<<<<<<<===== ="=f$$T-    ֞ lH! $d$<<<<<<<======== =!="=1=2=5=6=B=C=G=H=K=L=M=N=]=^=`=>>>>>>>>??????"?#?&?'?*?+?-?.?0?1?2?3?4?5?6?;?>>>>>????#?'?{vqmhc^  \  `  ab  n         !  /  1  3  7  <  I  LM  [  ]  _  c  h  {  ~      ""=$=2=3=6=C=H=L=N=P=^=_=`=a=>>$f$$T-    ֞ lH! $d$>>>>>????#?'?+?.?1?3?lyXL$$T4    \$  $d$($$T4    $$$'?+?.?1?3?4?6?Lv_7ЯKS\8t0 WplCakΒp ֲ+I.4zzġ1umZETRTG zOXSl$q kVa4. ,{A|.dm_qmv!8`W.$ٵv5FםG)vNH\:݅nz ފ 5sJ+9](`1:l獵1HnD4dnBJwƵԎIUlpؽ`Sp^t5q!.MC-û%0.ZT iȰvܓWgޏP%fYZ z_6#UĎFi 9V7UBJsAw5xl\lSHw)J "3X2= [@@eunBn/=Ӗ;D/IvBWI^ >R;!% JBdQiuJ%1ZJ' OY:u̝8$|:jHɾr1P#^nh$9՛A8,ݵ&BY)6V#ڛ>ukrz "ՆVlLST ;]\A?x' DѮ3ď7롤PT7Fo٫`I N_N@ރq9DҼSV{'@xccxEc"*%~-HTfA&qgԀc8]L%BSqAnA&N;n,mVF@h5s\!70]lPp &=x@4t\5DppZ 0QaYcke:zڇh2(m gZP֬MuhwZقH#C#l ,Z㢉=b BK01IpAv.T[t+䲲*jk[R EQY@P5B(64YvojU2)%iR+I*$Z )fU1z@% +&#&Ƣc7\pq×Kw@D7ͭjkV}$C0&XVnX.jsM^HE=PmC@z0;cs.Ĺ@lhӠm''  +DWY{0@bwU k!TsPNXMY]lQM+9SuIؙKq<P !ٲI\*5>ס A %'d^G ډAK-uցFjSZj{iK(^. E!!4Rql04pD7P ϵ>^kr,64gt =^`֙Th6j29" VoqHWsv1X.JO *zEfwX@qq&=L?<7P w!jͱX7uI֡[lfg*c_]Сt5vb jm7Uj#]4FNر{شۧSn- L@jdX$M]BYSY{8!xca R} 5>_jj*O)w vrᣦ"uœW 5y*bjw-ӮiA T&aV)C[,Ciڃ(fIK-RC*.TY,Ԟ+=ϑ:Nh=`(.q,uℛVb ]Zl'o-ybUb`:G,B($ Vеj"@jS\Ŭ%YOy3=mja5nVEO `H}\ ]+0]NYqbUr>J h /=鄒1B?=ohlձx z=-c*1`5t\\7bzJ]xm- f5B % ]%tIpQ]?T,#i&qHU8v-zr[r8%td%@tjS*XelHX\z0JC{mm kll%+hU5ı0WW\NC$R+-6ClkB͖W]]w9)q@up̓mSPZz`ةtÉ'4b M,DujǞBѺ%b@>ҩB.Qp:vBz>guVF'9뭩5ͅx06 .Ґpj>,@vG.$nָh[ '`E %0x(xø:'sb7"?@Tv5r<vރ)ջHZ"Bs(mXŝN4H mMDz'x/1xpq]ŵ=uzO5@5F[{Dkc\cF\rAA5N-+k=h?i]C}N1(צޜw=@ݘ U+@(ael4!HVQ!.tx=A9Ur5 V /v&$Zv\URt85ol:nQC6{fc5I Z $0Fҗᘴ{˂T=yBh?`!LB SZ{$V=JrOdB jvΞc.hu>2.[zreu*WK2O-n;{t,=Ez`]!XH  =tPB{t< ;BCCHim55yKR Ө.yP_yӭ}4iC?0ُT<h.b'f.CsxhclqOn' v =jC $0\<kPAh( _jdN1i!Ѓʂf̰*`P۩vA\*;ХBC PRD-`"..Ĭ`,,1CkW9$[?5_/^_4"{?ZYvMEAJKKwPSo'Ej`B*HSXe-T@̂D,@^ N+Fݥ(A.փN r , N py2BB踸KU5ՃxW 3v۬Ŏ deӫ]rtn[>31+&ɺJуu@ [Bw1uNT@`J!t#XKX wOCWL Zqb,#F)X[CA k4-u>eRS`a@w1e4VR)nbVMR`T2tt\%(ԜNHO{D%8֥^]#" `YɯLy%bfs;]M=mXḱ zuU CSw_rU^83jdq*ÐYV0X$ ߍj`jE Ed-(gW1ɊLM&grтb˩+0g'32b#cllJFͬmJ.ȧpd†Y ;rCy 2yycY6a6 qXba(w&\W5y*_^3 j EQ#Ϙ-36chQ-fLVL OL33VB|H!(ife#4SfEUE XAg-:}kQ}k)OL̓BɛM꠩ͭLkF-Ho(*{n"i7bI= -zZi.kSj]::Ւ$)Kke5wwDjg;k+2c-VzlNZtl O'+bGy^:VƝݳu>Uv\2Ŵ`̔b )A6S0)g`%)>ٲڃJ``#Lo82cB> 0 Ƭ3v|vJP3_c鏞:CNw͞-ڻ~=<*5%H13 Uyկl_u+z=q3RgB@K kpnBW5˥?yCX~F6L5i6S]&-^t-8I*wZms+k_[N)ʶtkO&٬?G%ZH:ս^F÷fhאY@TKƶ()㲜R]Mg_΅<^XgYc65O*ӛmjC+'~^ bP绂s9='yexߧ[j.yzwݥXJq\ܥ&F"|gIcgTH0bUƓ6.'n=7oϫ[%ʝ0Z]⇯^|c`Y: 4X4*nV|nƌ童xO w2<_=noY?:s&7ӰZT˔|Q::&vTU;i[Lcl.O맾#4?ztWi.XRx,f\}soxn^qKˏf*得ĭl>{䝻H>?k/zʿ9K!SE8hCւ>h-fɔ\@U\)Y??q(7;8`>)맇_Ի/~Ofqpg{7zY ?ձT1oQ5hR8,^2ROVߑ>8[C ^x_z#x,t}z̅5YS}Mg#56^f9-]lWkbZEke'"n8#`ӨˆdʘpGyן |/|[7ק[Aڻ }YڛRnPe%}W^ .ov8N衋kr )1qNF-m^y顛x`;bVҶ)UaVl0 PA&Aq}d!ƨqMr@z O_xԽ퇾7o՛E~7z1W7Gύo/ݲ_xj&$z']`_y<[O|=y<ټVxTkVz[͎[ QNfV4cjM՘\Ȩ U5\n/̯?urf}OO|#ndz%H>;]^SS[ߙ~󋟈^j._>r&J '׉.|`:Rċ>:Vn?4 UӢ41Ǔp_`&nK;Cd6yJ}Ó_yp=?{uf:sΆҼu?7n>⯽Byi*v;.O>ܺ˿/;WK/JwƕC~ ׻S̥I|tެ=?իz?מѴ=BXwW{XDMRE'R -$6oֻ':O)E^}ɧ{O~׶;'w0ch`[n[;Wo'r7r?W~:x^i~z~ٌ衫A衇.#l:sogs,kFYvDoo%۩ԭ}EaT7rx\>,ښ`s{87MeF/phgeG/w=ɏ=u[_w~=zϿ]bIӌPL6}Ye6fD?ߜ_xtI%7<|4ٌa9X5T)fOwO= ˊvIDAT?J-4_pdW(lC!S~f|֦Å)6̔KVo|3mu=egZIg{CLu:+Oyɖn݊n?^9 ZvϿ5ŗAZB/}MJozm{ݟxsY뗿9MoدRf0/'W^`G.43+Ы|Q X<ω$S 0Go^"3!,ߗe=vWLc~lX0G'3 WnI{{#|?}nz"8͋8ToOBI|8^QUƌFfV;J. hδcLܘ!8$?;ծ0|"Tq9[2Al1̤մ>^(Ic#R/X"-JkU2q?lޠg9۹nJ?~vzxg/>kZ<4TxdehO |l.Ey_揾=.tX==6PDu#UAɼvuVL2d=·aVx SC1rK}|PEypNӱ̊GmL\,d9{S?4"K%@'մ?-k\ʝ̧?+ۢaj>z 9,iqx ߒ'-(ZTOg,Sx1^<%ޝ׵SEQjsbEbЃdEQhm/Gr=߼Z5ͥثLd}"%sz 8[>;7<}ߦkmhNlH3ogGؤlMKΦnrB-ZPzkz6cޡkF^yyעKj=7uEFƛ44u[Gǩ ӡ.ob%L]͛[aK䋻wIU&CMĽ.vr*c[y(VLO:z&$Mő\M{dzu+x dSZD: lޝlWFp6;-G^{Sm_[%%|-$Ä1ˆO #tUuٹ3Ok}h"v$tmQ=Q>N@b%丫LXb= moHtoz|u}{9 .>>~~ݕ̆'.̟g?chʾ/׿zٷ:rl7:0b40+_x9C@&49oyOْ1vdC:&w/}K1{cf{ػ?!И oorCd]&?aw\Y^o .g8j¶ӏ!49z>qS ƂϟzrnHl4v.NL$ƺ;gGCEBo!w]0}vl}?o@GD&~'G$e=QI/ϿXi}#zc ͱ=nqXr.)%-ma Hp7Nv}3&  >\pd՞g<_!1`0ӧދao{`]/ +ZgqDPx-sNK. my@j|_zڝ7,zXS;:GНyKcqd" 7ѓh[H>d[Cu=+Ni"+)Nk=u{|<=Ak9 #YΒkLiL &@kٷ))y;'q#*f< }[8!V߾yp6(|A qbGX 3T[91sgv›%΃w{`wYd) ط,bѓ߈*Y^@O!l{T%rxT0[&TUTsKirӏE)Ιjғ3p=\r׽%*\2Exޞ I$#Rd-*5A#ՙ0* I"U8QފzA~u zD#$!(U(Ub`,%GGfD")m_{fR ,@ո`JWM+]z &GDHѠI*HiVOl蔍^fGwCĶ}CSC|~_F5%q˻|7BB5zD00sJtⲘryzU%+ /veb_xZB#%8͞`\0yl_1O/=&M`B7]Ecӄ D!ec]N}ŴHDG@&SӾ??12?"|KNjj8pg6#[^xn;cz\٠'9 p8< n9[]=_v t}Xgy_!D]=q"> exJ6(9ƺ6 2˘@`7݇,9gNSw,C<'5gKG`c_ ; v ɜr: J)-%Mn J!.UuGxx3` {%{z^~)<|\q*8[Ck"(}*r2jBkY Ԕow»6F8,q) A`fpxII"dJвjjLMB$nL&.}`Y-ZRǡz.=6T*f&an%pj\Q-zr\yyV*,ym4I飸YaEN唂鍘Jf ;|ZA`*CE0o3|,aBR}hq+ȓ,WDc9qPYut`ÌkyYy͗!$%O$0(ofvnTL#SE!& ri֨ 6n&,xbPd)I3BH?*߀}3kFTX)DOԘJ;N#)@^! (')Ii^= Pk1 3@̏X>%Вܽ۱pmˊʄUtH<qrW5`)^e><!i,h^krTW.ףFPzE4UooL*ЅwrzBy1w-6)8Ou;0_n].<=g8;XtQ:m`( QK"oA\;7,aJY`$y˯*xpAaR1zKPky,$k_h5#tS,PҲ^U&(EЊioDǑ뇸l\[7S3+Y֊pD8<*'Ц I*C pOR<_Yf:VTI0+K=ZnTU ,t.fP*24SOQ hɉl\sUd^M%H L*mMPqERŠ_0xNXD(p'r'FMn0Fɒ'U8C)J䳙EMUa>{VD xc~X&w +wo!(Mxl $e7eN r^+ *R5dxCuf^h@.!m\xn0\+X{ѮYTً\2K)mͼdѓ'g HBd1$:5E[qZ}*% 4:ЃtJ^Sh,)Nb+3epxxVXR$?Ch"}P]f̷u8rba rL-{7de1EAp\IRII )5QԤE5-ʭ 4I-lXKt^Q?ɚZ [^Jk{Bt^ .DubYM$8zLIENDB`F9!Z >JFIFHH.Photoshop 3.08BIMxHH(FG(HH(d'`8BIMHH8BIM8BIM' 8BIM5-8BIM8BIMAdobedC      s!1AQa"q2B#R3b$r%C4Scs5D'6Tdt& EFVU(eufv7GWgw8HXhx)9IYiy*:JZjz?*|?i4keaffh,VG~IH̒zjm S"ump[~YyhFx|=epF{ YOOip lcƓhn@tſS<N8ߑ<IuS a+l~zUߓe;O+*nWPTqp+ˏ,3.ZX+R3-Ⓡ3H6m#RmޟB?bjM"j_% ۍ#eį-rIuha8SV^]tCHbmTT /aɚ8yDf"WƼV!XcM6Ey*=a V A wK֯a冏ʌ-5F#uaq;imo{*[+{đsՁ^\S;%MsSȨ:Ok& +xeKXɐ(;4]`˶C*LFZ\]=b r?2)`YJqeeq /=60lG$Sٜ3?wsTY}bRy!ƀİPږiv->ORele<8$m3+Of8v. &.ʣ1+e*5#GF!IU}k(+Pʼn`MpQw 7ݡ@#"֢fDxHdn&է<3JEU!;\ɱ~#HyKe؝>}v%P>AЦɧGh~6oAM8I]DMy2.#?u#ōek <f6$~?ȍmŧ[UnBX׈jTt\I2T4W~ \ dz-į!a^v%ri jLE~q[FM l'^ҟ{*5?cVm5:̒=)]9QR mիKx&ArZKY6dCFPz':^9L۝}Bm>)U|3К>kڕad+ņm?WSKXiQ£x|-KPie CQ@zI(Nt tx4~p{vO!5@ '`7aXȍ4޽0BKkF``yՁIN1xRPuvQ˽5+.\Uvvu_9Կ.ŧOz:M>:dъ #ah'Ђ/QP<^(KO>2( o $7]Y׿%5o[Jҝ+i'~idSZyI? 쳱#M(ǧY d)_l$.ef'` g MbhM vk:bVu*jzt铛C!T ֻwav&TJw6I?* m"ݣ7~:$A)J Dž;?Rt[߈775(hcD(T ،y exc"}"z۩Gpѳ֢Ӓ?˙/,ayao,q'q֙OMzCPY|嚪 m!\a20oAA#BroQ&^=?|Z,?|Co@)BA5>4JkQL9'NHjp1K-6n n;WEbcˮ~^niBS@řOjᜏVv/N B~ĽP{JaiYE-rN:Ƭ|飊Q{g`GK~yW>JE_DN?C$ׇhh*Λre7·ڑ +Q5)nǸoC̤XWژ,O١I惘u_[Z3T ȖCħbv ^fK:]OLf Nu4TRiЮ+1G.hAђt9loR"R^nxS>,ȧtnν?֏O]E2ImFs(XUz*pESz"kW^*Y)QH͡I>$zX cҦ ;O (֬i'bIc|`ޝX~yyS`zhs(aP{?aYzPTuq!ث~cmu-~_~_yHL:q%%7cO1;Fi]VNDp2ի'!~f`ޑԂ{bAuo$t+eVMrSГn^Zm3vn#:jl,^rky\FªgcOMis~Ey9}Uj7r ›g[G̔0FB|5`EsiFg2!Xm5qun5_"I>[(eIᑇH bxF o=b\)gɾ/M*]u%R^ v?~ N  ?W.*K:u*Z{W_% xY9W ;dV49?s⎥OK9;h\$iwx9<)$id_V NQq%^`wih~P$VO+wh# e~9ʤPP=zCfӡ x#D;+I`Y9n)\4OĩPS޴Ɖsq/ˬ0>&<"h=BԀ>=՝W. ybeȥiżr+d+ͷe&+(]en_Ӛ6u']["V+$gDӴ}1&V &Mpjw9|+H#J_grBM) '+tQ]Z)ھb`;܌]-]4)= :D[Jr;ޝ6[!7I]¹"sL:>hZ5)%/ڕՏ?5!G#C>!9ɼKyDcլ5"~~sy#=՗,4KHU%s=k˲Lr'?^mK5h~Ǖ.e%4z{w `)Oj 8&2v(vwo,DWD~UɹbbHS ۙل׾ǾMl:Iex®篶CT)SZ{pbIb)p˿QKɴ}B,/TqpPT?$c]gi>i|F0,+^2(XcmvM>=sy#ӗ~_^Z=G/~IQHTTc5c w*g4%WvsOQS-hI랆#iLK̚Aɕ4s}n0]Gr^Nf*w랅ֱKCʤaBŵعsCȪPv; 6iH/s#lc/-]; %(O.A2EHxVP @#ƙ?X~E_ӮKryBnX {lyLKհ뷲X} =SI Joŭ뜳̗jPH^ʿ1cڽʞ$UMMwɷ}v-;XjpQd6wK5_:\5=co.Ֆ~_8Cߏ/ ž'~,^5ܺmœ4y'B[Ґ$d;7WbkHS̞cKo34W`*pARUvR[Iz|uU6ŷA[@\L[l{}2+}O:Eռˮ갈eg]JƮByB8nc1ExGgABl+ |B? S9{SW/>d Ks!<$6%}?GK֗_?0"/>PV>?1tbcJyHoO_澂K9@+n 'V:HĵMkuX ]i*ZZZkK#k+H y~5녊Vir\xjz/q;?"qqXLtއဤ,+:z1D8 Y}uK"$gGgk[Ϸ9]OyDkP?GXೆ~Ph?Z#Ж^[]layEb5 mvvU{hܣ^R?Q91|yxm_O!hZ :JMOq'SԎ&3//w܅v4?:Nr"ɾ`8w̑ʐۘɢGA.oHߧr?vzODW+j-*AA{t&WsɎii7L"HԳ~>y>a\j9U9M`'_$tay]OӖ (?wT5Aȯƃ 'W/$ݘ~lROuKT6/+h>`pi[YVF5hUYE 8Hއ7Bŷr; ONaZ'ye#cծvv`\e<1i9 Q?/y;[V;˓0(gҼy=&-a!$ɏ0|?\C1Dbj*}㉯у(ZTR݉.{Z_,~cgֲOZv #p6UUh\do}è_ G+yI5 n|i(̉n~jRGW'G/5]AR֟9?; rgA)~=m?ԟkRC>"@A#tvJ*վ_J|V*ʐ[~NK{-KXBE :*UV/?kUF[^jzͻͫr#|ծ8 kx) QH<.E$?WwZz5NoWwZ _)sjPFSgEjSY'` ;W/*վ_J_|V*!?KIWW [hӬdifV";',5 )S+c?/~/~/~/~/~/~/~ ]_ >AtisDd#&|  c XA4figures\chap9\bengulf.gifbr?( D urDnmrᏮ \`O?( D PNG  IHDR%PLTE3f3333f333ff3fffff3f3f̙3f3333f3333333333f3333333f3f33ff3f3f3f3333f3333333f3̙333333f333ff3ffffff3f33f3ff3f3f3ffff3fffffffffff3fffffff3fff̙ffff3fffff3f̙3333f33̙3ff3ffff̙f3f̙3f̙̙3f̙3f3333f333ff3fffff̙̙3̙f̙̙̙3f̙3f3f3333f333ff3fffff3f3f̙3f"""DDDUUUwwwPbKGDH IDATxܽKv#I%mRC ZmBlĬVѓT9}l ӻx?>q.mmw'm_wy>ڍO40&K^qЦq:c3d%o&tzs}>K!}L_)>B9>{^kG\RpZM\86z _7*xĖOx8qk~_%٥2jBT7n=s`mx? b W^]ms|˧Q*i'365Σ}O&j:C*'r65,JƧwe6.vW[ +C䑟P!>ty-^;} ocӕ-{{•울]+̄%bҟ2jmX֍[{d*Gr'={wP~y8z+Mq4:C`+ Oy, Gq\1m[zE yq m,w-.U(I۶ySWNif./ܵթr!Z<Nn/1v3_g% zX[oqy=Zdۭrڽ}Z6݅i"g=P: /doB^6[ngS" $(U.ҹAniA]+S:A>yxB1(2ȵ\<مӼm`b'm홒;,R Pw8k_1K˗?0W M7|XDqyGxGߩP0cH6xg[ uV%́ZfR&~JM-7_BjW}.D@W(b+>]XțkBy|{ m;~ܿfR+ɶhbQG߀\]2IStׇ+Lv~0vZKd(&V~ƚ-2!oiTg>ےY3zI ͢IMѴ2[0L/X`~kN߿|O;A_4XT2G=O) TFWa01;nMvXqŌС Z@{Z0uZq'(ݽ(i(˻KP2xU;5n7;ey+K_${n[yݳOX$5qS& slD;4l[R3=W ^1kvPqo^oW=خ߱s;V~ޜw7hT1 rkAYo $0?H&JIҴ.R vh|n)@ ,0qd.娙6#>x!%BtĴ_R\WOc(`6+?L|Y%-m,bWvg%n \km-ŵ7&n9<*Z7A!$ي;|$Ť5 ;;¥gWgM-"BWo,+"%!ڿIh@q{LJq4 ,jy*'Z=X:2ueM .% {{ Sj]йZ=*fDF%3u]'<[iTO")cԁ2pC\w]śh|VDŽ?θoiJ jyb^h YŜ㥼dJw^ ĻVZĬ 1yZdf--("Y(FAFܐl2b`j 서|c7.puHo31U@kf&.3 !)uTIhcNy_;dswJКm-t{ 0,eZieJI9N4OB'h[Br}dx0%6xOϦH!*=y͑h0@WoK$h{K\y,(D9!\ׄQ`,tX&x%OR* lO[9,$#"&u~S(#r/Z~@" ႙=1|'b\6 ܅-߅3"r}a ,=Y]ޙ!`ٲ֚H)vfkfЄ'q[۴04B$kL ƴҖ]`f֓jB`H53,{PcԱ !ɆVEǫ&epBojIƶ@K{*J363rӐv8 ,qB2̼y"`; oƣ)}$o;1ӲIjW 5PI[WOxf8$ [B`v#v܋ªge[#3)tD=,5]6gcƗǩzMe啶`^</Mc]%4jJֵѪC<_iik6 >2?OCz)-\!'L$JxmRlәxI(5& /VB]Y[hk]e8FO*N1eL )楴H%WC#)Ӳ"t%m!R "=hw;% C ,|#duT\e lq!;>7IG˪<)dF3?~Vo ;B.T q$FwG"U3-X9*3j]^ߚSJ#LksWUF B)ٝM'D)eb)L֓¼mi=VǗYaZ'jV(/Ɉ!*8J0֙Fi.ؾA@m9%pģ(}B -"?)XUiSJށ(1Y+\8(2<]ۣ@-l@-H2*C$Q3RJe BϬoy}&~0#>rM *N+oH§ iuuZqՉ ;J¶^@n Y˜/}RlD+ȏq2q5a2z%gQ^=+bR ̝',ڂ?ZBbOd'\yf@ )4<'^Ej_Oߙ+!)߆Xlx2 '7j?-[ڐy}26-LC_R7D 0-ΨhPPτVLa2@XaZrTlW7q%i&1+Óa)3|P:F'8lҷHs#֋6ᾷk_.R^gC3{pNN E/+Qn.{2<[}d$sqh(1ynm%c=a} 0?۠B@_0# >|Ӊo6ITtxmX9a7gt:O؎izƶyiHK+ŦyAd8 q"̒*@&b?}y|ᇙ F,naCo/|=&nC+$WDIe}4&Ԃy'G8fL WɷḁrϮsyhQTVOˉ8 ?Z6,tkDXC!e3CxE0}?Y/DLLG_ؤ$Q& #$W9p #R e$Jog'|| g O|ɻ7L+8#+O>͈h=!E8bKޱ'I ̜b3jYc*597(EM<8OkvLx v˔)!674z,DatIMnTm^\U,*ve8JّԻ {Wo,]',w){%Mb?$O8 3Yy2K# WU ȤJ -F t DI^٬28+\'k s `;/9U;Sv@mPmKrWB@iDC3[@"Q>UFs2&jnh OC,O{4 KGѡkWc\=(c /Ybn,Xc陟1 E z]rWM -2ɰ/@SY ᫰hGy+&`})MnRxȼ AA ֚(`,Ls@B;3+\=D4yT,a)&3 wr{[hv";b @(V=/DTrq\.9A-hYsfb^Νk!9YDO5< FV*6>>Ui٤+3M@5hbRW|ƒp<= w,_vsq_?.UO|z8'n6a:RzyJ@mcEkYQzh{?RF.P9 -vvJfa!V%U)ywO׫62 ycpsͣ[x*?v<ޏITF[(-lSF zK)%=%8HxUr4W)`RQ&2)@ 4<19 0""GijPHeW>sZݒ&Mׂ|pq AUW` ]W H[M*NlxLUǮ Z֗Sõuk8"TeY6<ДO@ 5!Q:eGq7IU(єFNC`_O  T\($PUY v,Tөz€I2N@4ecu !Rjtgebrܫ+ 4@|a  wȄʏRGDa7-gaQv!`eM>C6'CDIYee)\HG H3P[WNi q%qJ:5;VտGPgި^qruÖ^o[cI517*dEg졮EG{Gω[$+x^X )~Bo.0`>79#,y\q l቙K4ȵX͸ J@(rsԿv1C&IMzEwT\`A ǙtL+XqOKH,be5={8 Bz tSbޔ,n 7, a<$C/8!)WV,fGDJJܡ wſd:Opbyfdgm0MsPðڮjz`!7gp{pOYUk"qQj(F?@Ә2N; IDATy" l\©i$u&!{RE&4gl Lu3O;ENZp c¹Z֖YF.68ߥmvVOHJ!1*Cdz$z%faTeܮ1 f*z]SYoemJDյ {El!P7d9 ߁{@m\,|5Yx:%WBاgX3G3xF} J 9f^ W;Dm!kip"Y\G+ 6KfMhS/niׄ!UC^DwH@O/D7 ͣ!]Qj3+UfhUNRBR2خ$ .|L*3RܪC֍c< sH􅇀%S˽Qj"K4b$R$gMDU|ze.NU&(܏. ^,ITJywZ?tN7 kA{oW9@_d*K/ZVfUMv_"FlZ1C'و" 4YAaMj]T4'V]6-*RBnJ0wD..i>݁[7<+ZHgG#8q/>L{A3qAa2e׏apzOklrvgč}KpѲ:\aQyh6DG`Iǥ#8hF8߇L^&R'DD{R{DFM!4&D5֢󃩲>cfg@T-bϳ@4Mhѧ)lqA)TR4+*WI>J3@gn`T+= 4?Do7x 7M^ƤLyze$ RUcĎu7X][>)*RԻU]&񧽮nw{=C4X^zM}Z[ b a􂧓w ƍD`pJol:bbߌ~lŘȎIf><`23a@C H)"1(KX+@6+ }P7+< ڽ%O[mJi1 ?BFbL`KRp&272]u-&0Czw9=-@(l!/KP9j:̾8QO,ʹe=IO`%Ezy`!fn=sSFHS8,rH4&,⿳j#֧/ސ%{"$#;!>#B"`rZK Dk *y]OU)ϐQ+Ay^ܺى]%В2RF?T knsHDQg)'Ih4Es"'b%eMz|ɣ ,'1x.q@ V"4 x>#Ah11Έ>8H"%[W+r,ί*3p|PnLi=䐳%krQp'p/DKpĪg*=N# JdMeL C kz‚{_2jvޏȗS}un^,Te``ئGIP)SqEM)%yE4%m^vJ8qVH3&rnzHl >()R^ Q#/YRĩsʘh N){X`k`KMWBfM{SzU=IEJoRR?JsĻzښGWuoEnlN3lf`2 9D칈\²_Ua|*hNbPEC-, MT3"9d{[VrvxK4lA/tnHf%L RnTgsbF]/U]+jYA6sLӺ4*lRtF w$+y,3g9LM)*A7YXݚU t4ȁ-Ok75%щjƽԑyEL(G-xƃM\QSx*:sAscn֢ ):YKH#& T21\u"ϤɢhgPO룧qwB—HB΃2틬z32jnO``B<}3%8qy65o6zgy1$"fpDq'M/mwaa[hpY~+>>|YDYTM 5Qz4O9HJ,eWǥ%2ac\D6=3o4S:d_SV;[ࡣ5`bI=5^EZ⮒BẌ́0|FPŶ-ܮ,)P+L:Qj@@ Y Y@`NY!V5Dn jY8 Kɝ=[[rƨ TuZo{3&n Kp=Ux]Փ+\|k cQwdYwfx" #z ^p^zCV*Иf9]fRCTǙ"%Skl3m^j'6gRhB<SlzDs_7]\ͱGf9qbQל: fd f7Em\0& &3sC!Gg򖕉(>$8l:<ɭ%Ϙ7G}0!JW6,4IʳR\NXyR?YbXJsqgkӪuyHbNIÂL=.KܡJ^v*7( *#mw oeGlqf_PDa}~ @6~W̡o$Œ֤,L>Rr,yەbV'8dj#ۛMc*n4'D{}۰ R8נBJÂ\F_@M#nlT^á62YQ΍xI7IgszT@iǢsxv̡+x?W$ɤ}ǯ;żt*{&Q4F^յOcD"ln|CFȺ'U1cJscn6uq= #i]$-SǠW 3lΐ=]qP9HƼ&DMh0 9|uXA}=7#9<$\Y]fYjeO9rZ|<"l#&,1j`FOx/nSPDSU*r61 FUӵguu2ӘQXGhHW0X"Ci-V#kQ/8_C53e3|{|Bȯ,rqص,.%\U(3ZMo@cH.13 FT}$`4Fhex9>'q {#QW68y?3KSe~WDb5#@SXZnCSxݚ5ɀN lW $w:_=y?~쫰- o?NˈVky#T(@~">2-^B3X7V6a)@ӻ7UTGnd8g[VDygm_ K%(0yXlAi;i8Xwa&3,u:Nb{EZoì/;dg$2ȎS  䩲'jS+Bԙk^Ƴ51lÖt'`VQ*-kӹU;wʒ L\uKAjh;ֱ5&u>$71)ŖZ= $d@`;VY|v u:)ӄթu`ß8o 1en}\!ON^笄$ h"02 `p5]˱\IO*4ɜd=yS2%Ǫ@EӄUTce2MI,w#݃73pvg#2=;כ'nqގp# o/YMnIc'`gҳ20R֘;vy 6ۖhEQrHU윗H`Umә%f^Ѥ|z2k|ebvO6N_TZfMsw՛U۠-  b -=QYOa>Fpa>8큯\zy(&$ީpא'έ?i3`J#*5%s>3<\C+kcj-TqBRb$y=X[hwsF:3Zq$ּMJg2K(rpƷ#%-}/Y]AUkPm-:!VIb 5hX4c+֚ğwM%`WZz,czeE.(C݉TQK'tnPjVטSKo_tJ;{qkGбq:AjBEHDcxڝ`ַG}5IХY57=^_sfUi9}7  q_H%<ݼX\ZxG i>4ks|٧SbZK~KHٿ::}D)\)R |ЙЄeFspr6{yOiߞOWcXɆ@=]4Mz$6belʹO> :)!:HQ9Sx߈q:GR`H8iLF>ތYׂ 8|? N&5\(/uV(!=C>E4OMK`9r[z~X<8/L?*D~L塇YrASl5F+G^Mko%I6nx慰k\bLѼ63R`F_?NDD8lȀ [`1#raˍzL ^{,n^`3 oaY{RF yq ⮒D+; 63JECpS,TPݙ.̯[cTAls&ۑUzTnK+uqPXnSVN;G!KR5S0+Q$υ><~elh [F݁u,:ռ8/%Gʢ2&Ug 'b4ͶsJ9U0{tsQtS!xH1Xza"e" D Gz!J `ȃ4Y9X|"8QVSZpӓ8 K@8hqIr3X8W{qۋnAo`jQNe}5ZdrJfWQg@Lk~]DUsEl6GD̥&Xn[v@b9#hLQ㋏h$)$ {$ Ҟ<ɍFMΓgqSRl' 'X.Y`bX HuJѤ1g@϶ip㌠Xv8f g3"رC|g,3$80 iNOfw,8G E]@G1rEZt&2f*SI˂`>&x~ kF?|1H!aM&-J՝ĪA~`Ťe)1etĕiCy> BA쩍=8t,z3%soP̠fvOǑc[MBH#&٥T^N`PB5La|TuLi4XL|~1ãqE9%\@8Zμ`TZי)al(r,mfCȻʡ -5 DD1^\TY$?mN)xL@&!w"1x@i>qdS; tW"ά09Ӆl̞`˜h,eB53=H8FF2\I<2 T<p1US@ۂGT{QYlYY"q[⹲{a3{u.5ioAűIlĮN*yhZ$|plȧgףxzJ= & ZQ3Ę(Ţ(xV|@j2*#tigv̝ɢ(M}X2N95I ~$j"IF锲0g( A D*To*@X^n9WK}ЎjܢX'ɩ [oM syGaE'?9:j ]$zhSc87ic&6gR+0{1d/_`,#5 UAb?{ӫUk~L#'5ui O&X(MwͭωtJϠ)X6'l֜*-]'?k:a4(\QS \[׍.N۶.Gpjfo°fZҵF0q!| AfKsTY9,'+XYa9K AM2HshA{-LS5Ki'2};Ƿ ,oW gqRӊM o1bMf-ݔhm^-Gq~go`aDŸj0wƍ0P%@kȂ FP q(Ę'%X-8#9Occg ]|Ct| y֓7c@lSIĨ sMf׳57_`\T}.bCIK#s.!jhh^f\W}:?؛Ew'ćx9;>=8>&l["x rK?!d.KeRz4m 8<$FLKSLWF fS(=+@,Q):vã{ \/0 9(gdMjؚL>O4iUKڂwj#d?gj 0P/ɋ%#pNI,)Rb8$pzR/b̭t 2psظ:*}z~3گӢ%N>m$S] Jc"?qYect&­Niy4ja%yin 3FqpggO~&ªa\4x IF=NA6+s{Ll$*[i<[HkOn~z"!V+U&\w>V=rA˫neHW-,\7zMY ()@Ky ܆:=p6%){wt䙫rܾш U?:*Í5$1ĻX:TvF0w 4ʁ" fLΫ, 6#Kf-RV>:y{ Χ^݅qNs!3P2cTF57: šD?5;2O??WFX9 O(!#Z/8XіhD܏iw_g$/l5NxJejɅG?7YMj)N.I8\>Yo0'Vgz1:'r73C˨_oa~%9{zŨ0LDJ+=8Cvxa,n܀}lc@j73WƮge 79^Xso 1ٟp+≜-Q1K8uᔗM{Ɗ$C^$Mq-HJ0#8dyG05b 8w3sRl.1oTה5 .V?4oP>*Z:=$9ڲXWSve-qeK˯ j*6Xڵ*~IęC?; нd;U@bfm{K̂"ȎS`Ui A|%m"v^|zl>[V%08 4'_9ceu^gPN:F[vCRKHN :'5rtECjX ФeY(l'f;S 4Z±X|q!G}i<)r.ڼcy8N?`yth"6)j >Zv-&`o#x<%nSwԳ7E3ke(G|3l/+-9>lJ cKjd\Qصd_ &}|l;} ?)s0뢳`jF ӯ0b9au\$̟FghPb)̓ ~sL3N>CJ ?ŏ2wCa֜lN\w:<7=X]I8s2OkXƃć+o9VkHfc Y=S@'<+?2 6UTWx6aE' Z4}R;@5*2uܫFkB}1*J]5<ٖxMGXT|70Îj,mw;@%2:f w0W' G7Uv O1M$m{3+?}y"PJ(O}ym$8 '$bz:ۺQ[M|k⧖tw NSE+zyGsH y7M܅ܦ4jSx뷣l>dhX#+^c iGn1 ẋG"I;Yf`7%,<˪CEo/6vDy"mJ\Ƒx~0GЉ&\¢^dl6˔VquD$'ݼHA0WtL]Ʌd\f 37Ј.c2 4HzM- +s*9#:mN %fJɁ,-f6ig#>Wk9 :jvIն.||vIgʁ&6м$d)\q \u=4 >bcRREQA1[ze@wu7뺜MssL0zUv1cv^.)"Z K7SO ޲錎Y 3xv_T 35yxbF)FRCh>΁j5b^ůYg}:4rs^$Rn4<ϫsY=1I<,_Y wߪ`uׁ:b~Tհ8' `|+`I28gWUm(RIͧ 4Ii)ua!dXIBZ6.C4 -NNO 8{GDA!Ӽm1 Opfм\ f)?r6IɎez+**Mk\&,S fւNS ybO/AfqPSvZ6f-屐i쑖 2߼E4hTyNka րY[PfMB!yRJW|YR}j:Gc@NǪ#8n+!:vZgYؼǦ.?B<ԛOR ^3h:;j^VY*E˨#!HQkHUA_R=^fF-fa&ʟs^5|!HP;5 |V/GUu<{+)\43!x!_px83)FnobK4j=Db7a sSuAf &IW'<5:FqLkN(uuXʙ&tC4K|)Kj0K o[Fc3-Ճ G:kP')lV;B~ƶD^vX#|6㏢.,LU"IH13@_th"GHc} {M)9fӆGujG%jA{V*F̑JXC.IY~2Rxel-c1{ 4SӘWuL)6}FEPG܄GsTw4b4?4D4XEԨU9h#j"D{D&<'4̿Lf?◽9MLS QνE*IAt1O]iGgrtIk,̄7u ]*#vmjU?w`V8WNd\KW/g2L<Qd%{luVaRR5NF}m^ҎȠ?hΕee()B'0#Zrcwg.W\K󴨳Yu2t#QP.W'2}؁6̓'09^[=*%QSl Νm)FHvXbmGMqW} Ҝ^j LNiyL'iܾB%q$\թdvNi"tĉՀQ@Ґ 35ĸA "3֝I|qׇHo}OzEPm˹=S|4'5U).[::Q3e +'{QM̀E,}kv p9xTfpD)c Rwv͑o#}U̴=Tԧ%B(w >ywnl/{K3tYui!;5Wȣ*lvOW D]PN3 A WZK5DUk#r:96 ⤺E'~u%:H[g0vЍ ئ 01n8{"ϭrd@V6C#CS$kuT- WM!9>-sޓ;A=(t`C&BU.+|Ut!\؍{ϭ.x.#zrjtߍk$茅ʗ/ h/Ibޠ88a#qBAW!);S(z򨦋' u 3d]Z`yRrWZ]"tTi(~="b$@9䏫[N= om"?2EM.򒳤 b}n/vb3ë.WWLfs{3!"|34B{0^,s\s;%¬H[-/9X'U ͝Yz4*CG}Xa>Ed72m^Eˇ_L~=4=ډUJ7v/FW &+..hFYE`+?-;Ϝ0p.cGQ:4_~qb`D)l'MǬ'=.ME[>0mv&4ZL!D#s;%=۳Z!cZ}Xxw^i6V# ͇aͼ⎤-U`85 veeNǞ_Hvh-#旊vA,{92Hc6W\]MT=ff5Q%ۗ@qt8nRq%5NT-"[V:ޫFOpM+I)Kpr+|QYj|lFʩg#_N%|}[}isPgMP%hJ(24/~AkI*\ iO$ 5[{hg*eXU4~"Mdʃ ɵnNL6,al6@\}'@$L'Z1MⱓB }lEcd+E$ʢHd2Fz5Ek]ة gt>?᪏?Y,TcFs_S`Z -/#WRرWjp%O-E#q11:Wt!:Jj7Лʕjdݥ{)G79pW(h*ZnBV 232GCAԜolWE tR9U8 @<Egy O/OڎWlm"xe'"f$[g*վHfs.FJ˚C*#Y==JHK 2m|:S&KP'0qtӧϴs`ca:Љ ɂuo!9"M4@~KyLS)FjTcWыەzG*-ReK91/͂f.hL׿*@Gpc[EBܽ&]GI<>u[*sx,=x~5N0nRP\ɽ#B Fxm"P)A(&OQȪ9;]‹ vsUSl(E9)ASM4ۇ;OELȿf#f16Ql"C3*gZ,͗FPW7+̲%Jf=}h9aK. ×"3!Ӵ5w ; @L. /.JZN ioռ$}Nha=oDiIVt^[p4rD.^h&&K0xzΎB>vC%<t磽#XٍsDuzu] HOvHZ݂""oY8YE#щR#Y+qB.^(Oސ$V1+r0" h:,A ȡ sԑ@kK`pvmqv{#팴#\;ʅtn؅ >S[6NHa\K7iI1d˂-q94 r=ƴD-.k (,GCR ;`)'ʠ6K tArX,ɗgg@^P.)ȿK& {m'pn|@]HW:wZQ23=k8-,ٯܮnN\d?cRa ==x2NH j'Y }r:-ڮ\xޚ;eRs'YN{^8I>DzCC܉VQ (KgusT۩͝ڱjw~uԎz^8ѠF:d޴b帩"|\hG0'' rus/98)Ep dB%.: {=}_ &a]9S,Vn}d| "gʛP5Oc譫锨-g3q5yi3H{Xs"i Tu72ղW`s0D FLcG];Zа3/ZNlm*Z!U=i4̦T=:h YJS4sa%8ҙHx=Y=#az`ӈ2g!08ڬf^ p̟TX? 6S D2+J)K: %nMH * EDѐ)a|&Pߙ|0jVSȽT1 -I,)4]Q"dkt3Fd.1t-tڭ>5zlD1(@މ&w-`12>J)E3@ | CEE& YgL!Wnrs 6ST\O`Ф-Y.$Kfu|jUy$UZ)9g0ŸW%}gtbqY?뵅osiYB{fHi :'ܪ5#߆>̇[!z37}b]>aT 0Lk9Reen|x M紻R0uK96/ Zc'J ol/M(DtǾu-z\AeJzRDύDrm jqf_؞6l1d!ZsP/6AxnjMON!ANcA,a+vjgt^+ )[7p=었T9xUY&q ]Lt2g5q(;u?noDw5)H1zG'1Hl@XP}NwYL UabHjRV eLf%1XcjV 0SBIcUyz[uÄ);6rq)!ȯEW1ա<|sSPWj!@ݰsͥ2(Js *\ؽ(v9` mrԴRO38&(ܴQU$V"Pi\Mc+eE;pO'D\D,ˮӱeg5 aYNuU>{t5f:Ka2DCxۯEVd,`cO?B*V+MҀlݧ$4Iz[I`Dګg/{D1#dTF T!0׹U9y1-sGi3e0l9'¥'0BswrغjCRsْS-0>\^iVoݏ*>H!i"UsE:oeYmR"[ДnzLF)8/kaZjQ2b<+] NNc] +!`qJ;WѭL3 V.T"/rX wwGfh!?+WT*f0gLegP{; Јϻ~$˗N&I& '"'|ǣ%-^4@N) FʛvQ\]WDsed.FwU& :6זZ7J3[]=!+Ԕg:seA; ?\,!ğg x%!%M:+w6o`D_d{b=nMu(ŲK$Е#5Xqkp_.b~aX]?pʙsryxaw/4pp:Fh3>moSW݅+|*.˂tP,pHS; xF:"V,ZKΘ wNx؊Èt%>Xc6-h׌oo㯇'+ANܦU,: \ Fֽ-5GiEd:EFZw}dw&5/oVk@?S2[ň%˨?ˎ_b5>wRw_VĎq ƀf|G$L'dyla(xjn/q#~f*0u2DTeIJj3`,ޭC&'8MdIB9 UC0 ˆ#r=b諴$0pMQW~_EzEHO{׏l`s%N~(ZD $ƿc?sALp(5,@GzQXzc|Ō/`X=9 =LGU`(>u鸻QaTF- í4L'K07(_g1!g8˓(ʹhǸ'`sQA=۝eUr5m-_"FҌoPު%Τ vudJJ>$G(&$ȫD ;(Il`39Aehؠof#0zV։, UShIoU!ֹy:Bg6(N$DޏvԿ<*I!=( ӯ@;}i (~u_~|l 5).3 G(K |D]Zry g"3z'M4? .nP x} v‚ɉ)扈mM#.V8QZkh6 5+:ѠaGyG]bw^ ء>$@ S\'L/Wt44L\\Zlb3ibK~D:X*[je&zG VNgv9;;uHoQxq7$55Riq W3׻14Ots>  Lߐx@ϹJV\U:m4^05:[y!Xaݯ6,@]=܎A=qDSEgq~j%Xov;Hm@{Wl:?E΋$~{ 9D,{_qzMT.#( gJt 5]LV=) ]'[|0WysGQbTCUd%MpK/(n!~a-ڔgZf8.P^R "$0ֶ/@*Jn0׃t=N!UuXM>c+$߀ΐ~`Tֵe>D>^!t^֬LYkH1l:=WDt.itGr5d_YX׬&Esv FBpSѝHS̰Jk잺ӥT`j@%ndlQai-:vVݳa[C59V!Y_ (y( xZ\jzvWσi8N'g/$p ;鋢kԖw_q$`R${SD.l.]M&$Ѯ9t P吓t7 pkF.JJOB+U >Iԋq;U-k =r18<mӷ/|gU*Pr-EfNtMz49#aIݬ"Uy#a`(%Hqi%jYHa0OM٣`%wh菞lZh==uӋ9$(cV(y kGJRu4iI}j,ݯáՐ7\i)Lg+U#+Kϧj9Qy?t.h̒2UѳEjC;tt] YSvjC2^\I]rht^<,EcM h/XE㘩cçudn!:B#6uڅ]@c N♴'kl3vdxv5'\-r Ӣnd+{>hy-|,Fޅڰ[ f|QYY4N T<:޶|)W)aoj@m9{Y5-2DJn$H7&ew7Z#G:U) ۃ3AbaYՠ[Qixo <5Ѳ|Khy!2 #=(w o4*W9Wc*$f7ѫ:ւ)m[[E߀MI E? \1szMt~~cgY*q 3 9TwyYO[d󥓡~[/׆Pŕ3D}@ #73 D9q;M"sUK$;t$4i4u(Q|LEVEXb'lRyF29 ƜǩjKG̴Ң/sE&5zŲ̻2&[̄4rdz݂7y5~6}e}} 4w<8` (Z= NC(4.E /=AAej:egաmN(mUlJەۭ[cT&ƽ;^Y~u'c:m?&:"=֑u`<KkiAxMjz&!Ɣ0DҠݮ[7#fTo"JE3@nwƶ#K5R_V/osQH&|&z=X1$Ћ$&q kie``g:P Tt:MI.'оgH,x)/^CFs)#~)ln;h>U_6R#q។wff1DoV;N1> ߆'zD3+dB䟧qJ tp.1k?hQWhnlw$@D^1Evwȓ+d-̃L 9+Xx I':vOXk%\Fӛ9m~[i}i3HlN/EoaQe4E 'Y3W]ݙ[dp5ai6]QdD3CfDh=P3$.{>foSӌ[*;e+J4 F&B"9A+3x mJV@M9#V*ϥםꌴ"1V*qX Ds'n&0L{ɋ2SfOkqI@B+¤(& lKrf{oUC˩ 9 RCKv + $#JZ%2hGXB.ںLVP+R`LbwNC;OC IDATԙ;GO8U#T(XziXHԠ:ftUζf**dz{Lo"eNJG-yAR"n4!ّ$(SDn?11hd/jGXUJ\'ȈH/݄D]JϞg)̓F&iCeRzZas~3u4k̹#ӃkdkKƴ!?$ :wZ2Ks!EgE)we$}H Wg HBs ^ˮGKtF!^GS= iR'xN"_H~b6A+;J :}7gY&IO?,hӨT<7 h;mI_^$Վ]0p2NjՕ^=#M(8bT4L8vI(=rXyHS.Q5Y lH%eCGQQXN[4eC+6LD9:iXqinnf4_®oܽ,^ϛGZ(][V{{USK[|cK3[l`V+)2=$+ё} 7g7,!}@F3ҝC秣t<^{$ 3MǷx4ZkLV]߰%ҼS:(RwR8Um3ٞ̚^eH Fӆ?]Z^~6_K zo!bh'Uf~YH[`#bѨA:k,vRD1*ւ@lΤd%aI:p7\G*ևz@ӼxYF%g:>_ϼFTchJ5,7F˫CL|s1ro?,;;]b/:L(M2@b]v*,&3hƬ3qEn+j ]Kr0XWx,^\U ĺĤZ]+4Q GZ g.SjYԔ*IxQ<{Rqt{N=.W$c_H¼<,^D]r/R`Fɪ09>&z C5q " "!ͬ_+x~ǫx¢xwOVvs7?y= L]rA XQ-CKkIOzw;M A+$l m-sP,*yyx_]{ŏoGM{5l8'40]K?,byۇTBqtś;5/kX8wP35d&̊WjaO"ҵ3|f}||Q6pĐMI!^ *CrO!ZmWO1 b"hn#³l*8d ׵P.Օ!fd%Ŷ|NKzȱCZ pM8J7䊈ar?>tYaLI3ztMSYhjzzά>O[~f[zJ՘)G6t}ܛFZ?t:myr7 4𪅕 ⠯~%g1W2>eS9SAYd]Ko(1BDj\'w[׊Ckin4^xc3,;p')f/>?Xܜ~C$1 wSY9)2!jZ  Z̯\iZ>,wG'{=@Cb7Y lYePK / bmQS'% WuaE;N~E,-R\?o' TL'X=&>E"02#]տ[S?a -;b$c앬v=5%, qECN_9I( rpd*9 s9pkUw*TTq2Sih:.f@V-䮣ԮcQsa%etlڍBgzd+70kDt S~o"B̫f]J-&gKd!~QHdB" cB ~U(VhkEԔeۏ¥l]MK0hAx!'lmU}bאr`PRu&c ;8 @нPїHsB<Ԯ:u^~I"7@M_ZM kh-OO?I:ƁNv cO'6LyQ\& b,~y4[8cuql@]~MInBF[ BvEIY_Sv*zFZIGQ Z5Q72T2n4Ɵr;9ʄ~bz[[~u$T'K{>HRbuHqnRqpVL&qcI@*'r1Rgyv⸵S1fLg{@דu[=0PK6\ZWjt(,bMɃ#s@>>ASjpuX8e@gM!,DR K-# ܁|=WkH̪S  k _j ¥LJN,Red割N)g-j22F*Qk%4h3(pw2 :Z^>/i~l#ia&ޘW.P\}5(c O_{oȤ@]!8k n\iƹR/ nG"HBFˀ VkppvZ\O FWwN0aY$i1] J+(U\Z@XtyE3܍,N!$J]7Yn)@(>OE 5!.ٌYpҕ]$wxT>^8E女_ۿ)fDtws}P^3 g)kO6K.;˃nu<Ō(&7 %IDh| y!)n9QCEYL*$I H9Z;f ,`e_CmǟŨo) 5uæv+%]'m }6A81שeAS2FG K~xѪVRpߪͿk.oF:l$|bfmgB")i˓VBb3NJHES 5DzG AүٍՃrnVͣpV *`ezaOHF8zow ^ .ދg}^V<_8K>\pdoW9KYޮg1 bi(VLjEDݱ]o*\!&yFr '>)kgTfoj`6A$]'{vN'QJۉ $PYFB|[7G1<&m̩^5xIwMn&)e*-u\qV\=nw-变%QJ yU"Ї7W"d\>l};o/Ì6 O|7* ʖ`T9Fbr Dn\W4,Y@3;VqUME[W뺊SQ+HDU2Dy5klndVW0Y Klŀ}0u`}> yPuf[V^^Ej'dĞcfACS+i<lCR;?&泃'J@UQuE//]Ǎ(HoB\)wD1_B$? &MC! ߄\4;=;x tYF7]4mˢvwa^ժ(eGX(4wh 3XɛVl2ڲ:vg1eAŽF]FK[ tN*j)"Ay[K꠵&7tq.9ouB+1B#UI1MΏ8% 0TGFF\3Ftgx#&jxh)7c?CZ,WCd65\,, ՠr>THDv犋Ic!o:2#8_A8v:z[J }])Av~>\-'&xڷD EЭ2#Z]8):+#ttB^UDRlnd $H OR*_L051PQ}#@y2k_s݁끢]k.AACLw4H" CuEc&}%gl?J~){oz\Maѷ4"$,XiEs_7]ey.aᵐVCJh'l+c-VnU&8CDiRM3|DTQW"B$A;voF 86)}Q-IGDt=!= s!>߬C+,gI[{7#Cĩc. o6YZOmPl2LTcЄo#`4ZY=]⚂Ĭ DIJtul壡F#oapvr+:3Y7G˝ bZJw߁J%+̕)cG#qd'e-{\zAc|$~IZRV˰ Ter}t DvCp2'tt/}Bk{رPu$:72鸂i4+K+y|!eцCnZ*pZTYBX{QKթި-P>Qm]ZI~]WsQ)wZB;>/* ).8=IJ[R.@DVAߵXtEp'z*sm?QnTU;B^ZǶ-o.ΒRrcW )W> qcb?}\pI5Ymr#h PV]LCy;h>iZϖݲ$wEtY^^[y%|hM85叟Q;0,'j26 8ZcXޝ\j3e_F9\0IWAxH zι>0g"j 1EcnT^n~7ֻINÊЮ蕠SV5UB-\w+70V$>GAߕwDee,Jխu;*c4#v.P5MjE3Z [O-d؅4vCC;ڌfRoE)*V;+zང{igZ]c!E 3S8=AO6BȻ=6bIRbſpxgqՕUaf "k"|K#Ӵu7*{BFފt KP!/g:x|M3£b߄a< ѧy=/w=H[pbퟷ]iO"GaE $(WzWcCmPvuvPCcwn 7V&7z,caz"'p]D.W~G?.S*;Uˎ{#!ri^%CbzD\<\&OI2pBUN%N`s Om|Z4Mc{#Qoal{8VdL}S;+y^/Dq{2VA'G\D2+D2X-.x@Z] Y EIHSܹ(Sr)c!^oQ\xxQkIqz)1C,v.vH J,CoToH?V B^[̱6@peR_[E'E"2PLod7bOe>h0SLM#V-( qz4&>[Vcc3V\~NlHҹ8vʢTHnOO|lXaR%x{RdclX ώXFs*Rϯћb?LW4.hn^<]}˰/c2?Z%d]Y-Uӡe-`0H`K{`&qIEd2-ટs @G5X7U|w&vi-+&>u !x,KId%Ƴ)XpO}ѵ">^Dp IDAT P!N%Fe3ԿwF޻KwP HzVSq;Rh#b5q)^)`-b𴔅!TQ`h&/H@yzؕ!ĝ 3"0i+  odsj_2u||jrCG`8"Dd[i$O;ɁT [|p.E9+/ JBbm7L>X\+PcWE;)0>$%l"PУD 1f;J;3XE%t›bw  L¾ Y,Z)mOqP%aIO4} ppdËܞ6I`ҴgѬ뾅2Vp_Z0hwVCPJQ-<.֋g]Y<i$ 'KޯrbvuV_Kr/L+'Te3:j* 31k]㑬 րX?잗9J#?hh2+'YȰ^dU0-Iz5&vteO<<٘?h a2L$B _\ (¡U@=Vt(,EmcSMJ"Xd8AQ9`w';:kµBJ[n ITM[4."+w5QMcx29LFmա?8BJ%wQk+rwct7?ǾĒ-g*[ iE:<3 LtڢF  $]#"8 6b``7f_bQ`r  ģz3T7q&ALyӨO1MA0w \vC3KyލwєQ)ٝ O ?`mfDH ,3^+܇\+Vնo[_"ecsI9v7o A Dʍ<0m*L%-+-4aIo$MFQlv37NeY׉ﺍAJ1{RKqϳJصw0p%G,@F5IAك#Mr1*3QO>"WeF7@anWUu,?n ,pp8h"#uo(,~ew@Ɲlhc47&IgE=1,eX21h.r28eJQ'\!qlYoIs^p*|{%kfn[dqhV |-]ٵYi )H-o~!<\9έFwTsuUfմy:7hZUl|Y>\;/T֫g9: z8ӍQ9KI`'!Q˶eEspNDc/zMIk1WqUZ}.HhJ n3[Ė0& %^pVZ> X5X>x?T6$-&ɱKQ\V*IE`wqf$[֓IE"y\*Ԟ&k1V (\!JiA^TW6Iqލ,ű:qyy?^.SMjNwwV:,l>n>;>,oEMnB'|2JQɢOўO@AƠ~"V1 AgG@NPP*Qacd_ɞ%5&/) Zqgߥ;HЎb^c_gR8i +!b¤gMnQA $;: :tНܷzt &L/͢@٪RgJiHN^JO-)C+}~P=+by>›sNwQp:npy3+HCDXV8gE1q(mJڭUyƤ&&u"1R\*7#f<^ :L6'IyiT@66Ƣ y]hʎzA^ECNX* E8A;_Ջ) pg'z_RY9.{V?=^5$`~@9r: C)/X`G 5%mZqunI[N!]s6 sL*3DeT~ i+pJt)9k}54 byδmIȮY\[Zȕf'뾹E6zJG_9RxV`sKɵ;!݁pITl 4h"֮ Xqlp:\&[T-h$/v`IwK2$rR|U!s|xVpVӠwP-:^@*QXO ,\̜M >Ct|ghSCN7*za:*oe=؜=-# .)ro7ҎWj.inf,IhO\Gv7v%p[9deCa\&nAn +d:yKMC# `8/Рf@3DT<"U?Qm㎊,-2 l+}>} y6%O}v1wfނ}ZcaxP38cu& I/`n~S( kP/"pWEO=0O^(i򪕒&Z n+7loyP"ώzș*B؍!Z:Lfl3cZiwȌ!U>htrU{#F>>]ga{{)G's+8ڸF>q~/j6q=Xe}R詯P~FG@¶`,_=Μ!Iغ{dJ2IVy;FƮ'nu鰽oj~sz褱xsȻ *&jX0^pI2=TagW<$h 2FxN3yq-ĩ%ż?JzcikYq5EkH"ՏV`27O(q7ΡG[cz3D4JgtMU)2 "W{>^\iXvLjͷ+jTshShR)y45Cd5IUQKLGL y8V76jt.z-v :ѩsw]檵;! Bf=3hKqaϹKׇ\dxpsE[}/tfS(CN|=ZF6|Zew7&{+oYY8Sݹrn~̶ b)5Ye\T´ % v>ӫmO rlJ;@`tdExoE]M &g* ęՒv A/2"I-\ȟunz0ך!j[n>%Z0Ih,P P,qdyxr oz1NPڅ'9+lVUEU(.=")?xwvgbUyrqUTݣѝK|0fKٶZp+2؎n4科cHhq6c.$MٳR%s1T=K.率꦳EcFFNڴy2 4ܒA됹v[GT{ ?Bɓ4jB3WVs) TT+rK$r؛Bװ2+M2J?nuVFY5p85C6~xcGHҌA@gRlR(e BtzhTЩ&JS6RU漰 3W.I#Ip6 8zh7.B 9ì,>dycf_Sؤվ_/WYMk19K r>[CC9PAtIhJ mϽb}9U挷ƨE/mb@#,81 Al`I.Lf2.)g_i ǣvOLרp*!"5 ?9 sU7J~dHF+G}ZF{+R}vtLґ-vZyf qC˻Ҟ/Px3fFxSgZT'˸r*b?e?DHJ ̐Y4_R?7J-ZX8/G(XDRyФ+YGV!( ؖ&d]l̓ fr78ˏYbQ'˟|Gw_Y_kH0 rE FehJc+"זzʝe M虅汘LOp\;a@-~$yEϛ֕r'tg =?DV몆YYa>Iydt{dž:p[f"XjsTCXnNfsea 7c&Z ~[ֺJ?5A}ꦐ2y!B+OvQH %M͔ ӡ4x92{d4OBҵSHHU-{nFMߎ@hfʫ/pI4< a\:ijg5uLx.x{@f!b(;YWl6uʪvmK{${aZ0Dg֋l'z'o:eb3Q[\wIZnT@e*a} fmR{!>=]|/Oh^c%{ϗbag"5lsM&[.dM{E Zu]:J.vuĂ#ö6: L<4K\EbS۱k%8ؖn7܌nՏBBhJ.'m*}S%MxF/6Uh8D-t?$.@Ӷ.-0aǸ~R@e ^CxP{KSTZ`AYKCL2$j4Cu"E:~!J +^5e4=>B!5ɥ1U!1X(OBLxO$lFaJ)JHvfUDrPQDž6.\`<|)g߬fWI:% #Ms|(|Rt s*OdrD(^@3'QESݰ;Hmb<:0O0Wܧbt<д}򳦑ÕX8!cc5 ^=7b/q'K9!Ē ԓ݄CP s"dPp ʩ[Ҽ^HP6<."b'SYr%*%%KN@ZB[kwZ8\[^ "r%CЋTA=ǜ )nKܔdL"q5Mj|N@ RBR3%PI  JE& m?ޛX:`tT4)BifUo^sf+,n!k$ WIʡfq*kO. RjL템`ڡ<;Pdr~\JPrv؋Nw˦ɣߊ^O~wTiIO0GʾG2_)6ݧ'Cr+vO=iX>~:O)r;ۣmce$m(7; IJ,:ܴ]z8oZ5\wbhxXAB[piXͣhaH{uk=.&S{+zVi Л'0cObQ ˤ -f)|̒%x%/rP5d;DD%ʑqӒQH,\΄2iW<{{"GwS0Um/{-Eq77m3Si4mQTuAZpG$ejt::\@4 Jʝs;+/<Քlk`+ # GڕV,78>U{ߊ%6~jZ̈́?x6o=T ~&v];3hf+4g?vrd$:Yt+7,*g1ʫ$Icu`1RכJ4\ˬii9UcX'= َJb'*ԱyݢlPπ]P:\^TܲP+j{E KzN8ٝ s |&EoA HoFޮbğFc':pExBJ hLvb/KpVx{;Z0gw!kvK^w+YZ!%1KٽI&p.i!Qy!rpNU4FJU~6xK 5ZMCM;/#H]ր~5̮&JLrGB~Jz bfk-@f*N$w:2^j Fc.]՗A>2 dX1CU%E&=UjxBwT3;+&U>S %.i"IT>v̎&‘vw() E:c⮬^#qQb~Ws^I۱6Qg6'B`Qc&*\d}h"@ƤxYМ&% KG 8dɩ1\ޮ[)^N%\GfCAFNqޟ n5@Y}d&n=N@{$ ̡ݓ;cTe6Q(,%=Kw!?W&'DI`0fCՁ# O ![U 2gXz'm;Z$~fV|/(ѺT‰ |8XVvoZN1 wAAega.Urt%hbR1NJe座a9Mك^O Sa(SY,$S8!w> fye J8m?^!jsd6BIV5. }{wM`רg)+_yI~1M؉ [g5a27$R$_S3K^5#:p^֑ (zi숧wEK?wg$E~ >IP<*ְI¹?BA돀UFaX" S|l_.R11Iu'^0J!0|9sbm8B($Ȁ1 M@7RPLEPi Z#G\;pl-*iTUaVÙgÙejTZ%XeNBt{*e,,/]:ع=ef|9TnWʨO\l}2L}e=8{M0~c7G==2/G HkI]HEwE.!Ǔ4鿲\d%+}d5U銌#V F|] GwD zk*!W8ik\s pR5Jb= ka(4FcO8"}\ڌ=͖7fEeΓm  wQeh!WÃe;+dqLeܩ n(PR ?Aϫ%!!(CH/_֫ h}wnt0!bZ*pǎy`JV"@ilR{1.?|Ѳ\l8YifbLQ!s_,dV%t` Zd~*6:`tpBNֆ9qWDu#|JI/e,(W[Y5Y|tO.6Wm4$[ g.5PQo U`d7MGHIpw̟wk %QjEY2Jp4@r[hMKzAAh.e/gkd^8^W5hzOs4H-=)nDu#`NJiI@jt7[{}ڗ] V«[,6D?+~La;tPJg]%{GT+9OV/+̓ύAeHqO]{g(|PT'/E34,Xi rSwAfʈ]a#j;J+6#է!i4<ҞșB'fF>GRR:ߤ7b9=VaIF6<[yTzHW0ϢHիvjcCLY߱HnmkI*|&7paS /˿bstKE-G-˭*.6^=) 5=6&' f U҅D"V <J'q"v8h+*㐈FU2բ=S$Mf ~!C[5*seOv4Y3҇po"7!V-_4E?Ij$A%ekuV<%bOM\(y /73Cy*X~]-N~?Ek(DγhϏ۩ZEe ]n݈= 'rUUK E;bW1[y/Zm$ VԾR1NH ]-_#0VH'k0Ԅ%5l'=U|}ҚI ך.j~Hyj>IYpOZ"_S-]X.† Xs:+m/|Ze&ep*99W_9R (\>K\ݵYѤ،4{@IKQEt 1&~iv.Fk?s3OڧVw f2MYiN:C')zJ\uv;{&kj %KhbUD+cS,bkY=ԝR1 =B LmXCCEF|Gѡb\p"郜MA"it˂DajJ+RHcg{Wc'MwVT;;Ǎ1a0%(?ȨTml&W3h+C=5*>8=KZ^)0 "l}3DãyM[wW0|,4u#|:,1Cy %9K- GݓȀiwo|@HCQ] VEazCFL<uh9kzBХʼљxG$_[ |7Z^v[ Mvdܲ+'L[vxal9l2yv,3j~xb|/!;0]hjp ۹0\+#C.֡S+ՁwaTEqzI"=Igy]D`C!~G3a*9)څ"S'gNSV.gj>qo9lOn``b)%K9)~f cNF2OBO.׎9<jX@ջG4a MDl'Vvy ]mdjLVϙ~ܸ;dDӊS[V8Ђ rXѲ9qTHSt͝5NHlο}"\NL_`KRΜȃQw[5.\0pK{fD>Nkf#e1{?Kֈ7^sZxG(0$*Cj rRyЪzED~<VmG6Au46F8b7RVR -LJJ2M.Xw@'q<+,j",;SEfG%<{J|RARQMhS C BY``ȃƚPeZc Hig ]]@N[)nUk.<5>aooiۣ k16rI%GgB .9DEW3WC'D9J\1?Tߨ1Wb+A1oOk'?1;xw;ZV/|ÝNOLt`8. aօEM\1P lb:27%@?LW]fdL "3Q*l˿l;GOibķ]e2sҮ`v (5u^UE,4KZ"7a$\փ!Q"~"z`?2'aG3A9&Xz\s/I:;VL ^U)x%jY; ݉VoR?u{0_9h7FkKM>OC֮l'˃*+?Lm{gM̍5s@N}sf@2"!t!OS]F&6PU['5224S"^]xl1>$[aM!:w wq>nxOoΉ12c4#Qj "(fܕ ?o+gc0,?ׁ9f,A%xW#h>n4f0I0乊M+J|=87%)DqVwsO<$‘@"Sm͗x)f3PkkNi-1OK,oFK<'5,W{η2_e5:3R5y}W6{>Li_+=+y>5_]64)QžCӜޚq7N$q(Dfi-n/Bl\;EJxIVdID%fq']9-pFEk5< .Ү"7)K<$Uz:mKhD2#9ؚA=?7ũwBڗh 'p%\,V24aǯ|jLWIy bX}|57D :#"jRywy-ԣe/-Yn,(\_dJkNXߢ4H6iպ̦ԏT(n;;!SUw*GN2@v̗>Qm =8@7a=$({*YE}Fs녴 k[ Twt:"*uB@Pp h [t 9䪌WH -cSaKUB6Yf>ݏ҉(Z䤦 2FK+H[ ra\`1(kGBw"ƅӣ:f_5^g0 ,UZŝjĆA39B*sYMݠ͜/@,0jLMq&"9BUY;oc{I#J69R"_&f(ed*%}:e@Y\$;Ew9//{W#kl`W$-{Jz^d;RDퟕ-zTEA6c!G%{t+4ŵ5dٵ]/-*NbDHE4#L6=MU%jõڝnL`y|0XjHk@۸0j{E.A$0|@IoN$ݩj7mkSB")}X|#jOyvt2.:Gт9J$G! 44k:L h=(tbɘ¿?TǾD>b%hVN nxC&!"B&@BԒ#hZtR2y"BWBL .Eo *^ibTY.:Vn#0S MLVL=Ajy2i9_*R(+ȼԨ&ƒ,.TIҾZe҅և~vp9`=u:Hu=GcD<U:<In4JX6s`ْveXCɇAx#=>਱Fzkpl;HFh.lq +rF*By43M X3\!s:hE;YZw6|M]Q+|U4'.KD 94R="I_T1Zխ7"N^&|:Wʼn'wi'5Cgʯ߾`d7 a> CFҌXFfo-RICr꠨25y$H=]]d ,^_ ;%˫[.EY2r%(!x-C@W-?k0ɥ=!dKŴ]<-#kWA'C!@aYQV3ӡkĹcz$Csj7ܝF4fRRΊ!G R*-LR]3uCèX6g }̌1JXUzz\km,ڃ 46$ЂVEf:O;@TP Ocx2PKEs1V#AH9:P)~]6B4ydY!m'yS$vbP$M",AG\&SCCXaj}[cuXf^Q:H]436rRQ4Qbz۳q.{P+]i%:N-\e5(olhwTTѶ>gHQO[t:?% Gy$SĎֳ$ua']YgT1 )Hqj 9&G؇d.{hdd! ә׺u`RԩubG= i!CC|O4?mxY]~pC()Lv SGTkl(vS| G5UiՃEAE0$0AG_ꪧl$ - 9 *65/ [izI)~.?"܏kɪCC-p-al猢OXG$R"T$$n#ى.KdcRԡ pz;.lexm+`SDBqVe,PI |kU T̅'/qg!' :N5hl$x =Y5EվS="Q^!+p+xk@ h mUD<ViϷZ@zegIG01us ^T_WsCe<I+Ԑc<#7|L R$ kql-A Pl ppͦ(qOPU{QT-!xe3.xZ\H^?P( }:mO ێkI2q bKI^8ㄋxdm=iYaB mxӈ4F'Df꽢y̦WV=R9;XrjM,8O_ix:=S zNu,D0X9#E '6G7TOwrNyW'Z#ƝdhLAˌvP⚫Yʽ ?ZDGȃ v{EUfqTOG& Lqw'P\k'a\U4j$-!2<<P+NH@C@I.8|Μ.,`r;.`[+')Xt4WcYqc+v|%t-%dk>rvQưk$Ê 2E|dhRbDʹՓ3[[pIKGP R*=Fu{UBf? @mՒ$ IDAT3x0 ީwΡvb8 J^v҇w)MdqU#ݪEbK8B%Q-EiX*@.1C SR-) M?VA.V;JSB}z ;ڕ#%9XdDy C`P3swZ3Jby))Z SQw@Yڇ8›@47Ha̵)I"]nL#S cq EםF`Uy")s5)In*"],i76;5n5D c%~Fv"ye QTI>{EOyJ!K('Lp$b[3ĹGG 7ե-ʑ) ,-A sݪqe; eI]Knmݙ(/1VOgsnke>8!;z&@6SLk~+(cVvbU!U=֍/RC>K5 u`#F/u9NCaF? 9 &˃R)ZHϛeؗ'a?->ӣj0=R9~<ǞT /b4)+S1Hi!clzSيw+d2^.(6R7}~lVsJ|`=[TkcnO47d* oް^X|VPStj pҫ?K\7\y٪~S4L`!B,(VFfs4v>Ƣ'Kaf7(`{"2Yp+K5*Y|u&2  E,RȈ-[ZyNnq12e~DQlAêeкKZNՎ;v @2XÍ^Wxʻ?qHlHN.!SOlUqC!]KfW 88̡ *\US7Ŏ:MviA =-Ynp {~$f?_!_oOiX=S3O3Vx?5_ af-4cVNCL~_*^364S#EA۶{>,Z*3 *lJ?r& Y =EOs'GAEZe-_!$(qJέ3{tȶ܃G!_{U1R$NVyԀ W30jL_Q})\;!gMk)H{OwUzT'b|84{1G5X,Cg1* 1v8{\kW.x(xkRaS[@܀@@ۆ~T%X]M YuG'-gh{ 29][LYbw܁f(bٽRsJTNҍ1 UB)2z3#?[KweCTM;,7?'@4hb pwB+ T! х-Z#M,EQCܫ{rkIUr .y@-q=Y{:I ڀ3j=IY 2'bP*F3ٸ V&V+Q:jJJØػ6|Պnagxr @z)[Kb  FH=:34ACpB-Hm:5vn hUK*3XϮN4y x ZA|"$Ϊ3zHQ0\lɼ--1UCbd&Wx롈' xn|Ȁ[Δq{ ORb/RcŮBKt6U} gXn$D2w\x(^*꧰0d4b\-bBNP:e8_`(ͩBu`V,; GYSa`R(&03߉`Jj^ !Eg-<#kMl~7#NdґXy^B6+1V" f0WT1r3f ÇA .(=;N~-PM$' |E6S$S \W>Jڤ̒9j. )4 2{-Z$Yg~:l/ZPAgN.II> X,Ŗ z.AmHDڇ+T*AdwiX5\NUki(gfl,0^y& H 9`w)v`7H %} ~ucAnZq]-WdN}*beVtKoh)фl%I)FR$j8)ERr*QhG` nl~ǩ% hggDxG̤ xY1v?keUj3}~@j\׃e1<;d&JHI؅EAzhǰ;4?̠Qy_B̂'M\)$i̬/XFz!'ՒäƧŕj+^Z$]6|#{? e@H,fϹtj"SޢբhZK~k`MUG=|=V>k8'\NQֿ-i)rDz2UW;ZP녯˚,`80W/1.hc!X6$E Q}G\&m:_4,6YArh[.raR#|"$~2}_۱MܜU1_LJpW>0|Mc'^%MǨA0BPF8Ȯ7r2DwI [#RwqIڪ߾hYUxFnz=ń<`!Py5RWD(ǟ}ix*:!i}* T3$\dgHDRK3=1sXxbpՓt=e&%s>Vpǂ~,ѵHU֌ ѵ"H*w-}rҏk'w*~v?@ EZ/U/yӫ!;%۷oQhS_Z"U?a]K++կfL lk 0+R>Y| WE,怭 M-Oi4XB96I˹tvPU& XXP2qt_T/;w`wt$G}qQXxp5V%d7zr{eqi#"wb r?(Feܸ ybhJV諨!%c$.Ff  O0p5jE<.٧7+vR ë́U"<ڡ*FPO=WN)HHc`]׷Ŧ=DTRt3c_oU{AS%N7᮷\Zy?I\Bp2bWY,Sk]!#l7ѩ!jcf=9#ġ@\9jwխrc.pzC(C>K@ktpKiLN]̢g}#DH֢C5Y].L"LEqcy y!@ǼNtF&/3 N&ÜiM¹5*ƒ֭ct:V(9+?9gE5O$-3 DW؝:9_g)/$WVpss̡ʕ˼=ꍺ.tD IDAT#\2K.K88aAT}-5T*¹YvDH8jJ*] $mxhWfJ'{]Xo.w{MƓl Kl7~"šSb‡/|v l% B*]-fz>zvMAeM ݐUN(P!AfT#?+gL*k\:Rqd 5זTjǒՎ|Ԉk2H8.VW|lČzL=$mE##_?_ X6rxO#"QHgZԏJGxpT;DPɶu}X!oiN6_0r@kPț9FXڣ3%k5;qFEpi d<K 4L-Kb A%15T=NS|TZ.{p;'5G"bt8CE虉rnV_'8n}:mE6a|.BX?b!DYL^Y8#IOח,e%jEpuObfJ[gkv [CNX^ɸ]khuGm*EI";4s3?gIP;aQDbk"09:>^_أ!@Bpm*7yCE$HAQ֮ f%9SNYڒ?bDZR|f 搾8Zed(A'ZSJ< rLwE 2#@ھ/~f:$Ru4 >>`j[ٕq C%΁sۛU  ʅ]aPh*DeG=}vM7 y)VDJ޺eT"TR@[( .Ȇ`"ߥ hQ (Oح(B3`puzKwi"{㒭AVqMHB]}B`BD8IfLPMN.#( geޘ~RSkzʾ&Hb54D*[X&!N<]OrY=5HQgD-IncW<j"n4rA:V!Jto6w8?^KDY"Lɓd YHRi-f8mRROmϳL" >bh2Ma=rLNjUG/6Q}Sjl㐌UVE}tR0V=ᆨ(Q. G΅Dv2vO&ӂdM9?@T읃YN6ip^xOjF(Gkg<9"pi9 ַdf3@$~6^ 5=4] vG1-'/G)j7(4 I}2aI\fIϮ'Ssba\-hȨFl%ҝ| ǕJC%Wf~ITv#EC{Lr-_v9LfՆՃE>ȘzWCﰓ>9?!&4{9$ʒSW0~ MA"BkE [Kɫav8$W 8z[՘Tr̼!*W0#]QIvZs= apNj{Ʌoj-Cnhô_v 1x>,:&KjJ |/Bt^]8*pӪ)r$äٶk ɍi̓]^_n "X;\ @;HɚE_x>0fBhӊBծ6䄖g.A+ 5rv'QxoeE#q-Rcx䶈-<cE; 9}"W׌|%M$D@FMS zkK%*`W̅VO2vpRۙ=UcK̪ LRKkz89P3kqY#D{+׹/jww(|3r~9yUHn+ɪID^{xG(M p5-gObV+Yz+fg9h dkJT. &'b*XHnt,Z,^xkg{RGM« P s , b1!$jK(MJVNpEX]!8މ[I}&J6˞߮ ]Im6qlj"jY_3[W=`K.YI,^Ήb^eNx!,Zɽpz{]ϳs)H"⻜\5R'2Q'[1#>Tjk/&G/kS}-JLAKFm:^ wRKuςGkib`A'$O~ؑgw#ŲoN1՗gfC J7&C9D̛LDBŚE0`[ӮW}{ZrHmn$J~pnht©TOSѬh`<'w~Rʣ#P.֭̒m1U"oakltqLS݂,8K Yݿ։{鳫>ЬHWZx-GfOWc{gնL( b;BȪ"!(]jKjx `U*,d15L:Дj*kj_d{w+Y-:ljRM0[>J`w(CKΏׁg扃|R2nxs3=o)SsU{YV4j$^~heeo }S9-׊RjtlBf%tj\z |wKAiǶWofhkȒVBT̃n[isPbV GKEޮ9O~ JvBXem#*OIMG[)K݇T^vV&RNiNwP{ySBBHZk3BAohjLNVPp.l S+uGL$ʦ(?s~PAl[dϒFT2M)%2tΥGˏSi>xb+,:Vid\MUeJ;hb.K#iBSJ6U$#IV=ƶ59"#+l1u$Al '<:MX^Wc-gls1#fmAaYjtA pI3tSy5 .˪9K3ݒ"~~3֎+dkn_kJ&M|r,XVfi|FsZzg=)i5Jpm$_,= ?G$wi.11@'YQ|%ҡz~' bΔSbJh.,\N T|aP s2^&ʻ4x)&lʐ5pk޾akWP#w2jӆ XmeX,P1Dn>COMPE lq@PJnؤhx*+r@[L58?LʛI[Fٓė/(u 3 . 1beo Eva_ҝ֖Ae%W d[;mR>F,'vPWt,i }42t1zmw7Oz)\Im{p6d#Djka؎[f2^vѥ.P47>x6 :\1E7gFDa`tm~Cw8ځ*E%2V26mj;)2ÌԺ̽! ;a=YzT2|# 憝{#%쮢 @x{ BBifkA96+dpAqnes2T2qj$*\tMB:G?XR*m]x!)C]~Vy6b zr <ҞG/L6>>QdWڪ &>=^K販T Gosi7@U,Z.#jIiߨSG^8m?R o`z6o}CO 6%-  {,$6DK{W? čdaO-&(Mj@U,u7dm'2N7, =ni˨u};H ,cʗb";,dԫ .;"2Ha0J;7B~[,U^\̶ASUet?WY.e&uKQow6B'32gT1|RЕ Kv8WYdvE5>zi0>h`翆llϯ6a̭-9- MNEPfM.Lr^8Z˛q!RA \1bN\]y# z#Ȍum"J0dBDӋa\d^*#`ەL0J%0æ(i;]) R[w6/1;'Fbi"?ɳD{gOhp:Hyg3K񄠉̐T;G)cГr߻r%ЎD$?1lͅksY+5\5\Wr\< 4 =K!m^%JX,?(KE' +#Dţzqd++0۵VBaU1,ig4հuLsWked'd֜>s FC@=>=NIRA6Fٱϳ>/lZ[0~PS hM$eef+)CE)$Ÿ6 Im=9) I>4j5у" egپYT wοH12=KaYUSpQ=ʶ #= ugm%3C,P^ʥ9p.FN7K)"MN1QX_[UlJw.iCGi[C ;0Yl Pԍ|d &E,j^`r@sg`Uua=55`4 24V1r> sdhԎ>6a~~#%PY3jFA]'&hȦFB$AʬWeB)Sbe$4vOV*7(]}TaY̙^pxPj KrӦ tcH='^GaP!da+V19Q*dWe%Pi{dhki5/u?m$hlBgde&Ae7 vVlrEv8-K2t e0{Ӌ&`Njh_[8SR3\GX,5MY*  1B.Dw DeNDkո]{<( SB}F ݌J­r'f $Ȟߛyds=(|SIؚR*z LCƉ0K' aE k˪(7/*]yUcd SgZ]o`1 !X>8೯ӊ~X'\}}\$R`ٵ*FkvKhW%/@@&7{vgɓCP75l ,xaI\vIDAT9TOØ ^dc1VK˳YH#%KJd<@5)˧J:X,`M4G $eD8)Pw zftHb.P ɬ(:L'%E{,Rjw,t[UoNR0ls^C&yO_=n._`X9Z#Jq6If{#+BOL&%H S%fwѦDo0knc8C] ])4zVɊPElBeөOͩJo_gf 6}KTh=z1~T`bFK/؋4S:ofi;\a50׈9uɎgsӟ(ɵSFѭvã_LF}XeJBŮ.yS^ ύsY`hL"G (2[ e{գt{@&R/RO]Opfr^V'OH^YE[[e+^W\Be8aa܊0ꥃb:kOzUČ3BdSB˅¬Ғ!: CtT%u_ƓuĞfY`B%Co $R͡ 4OTƈW2q˔"h{1Vog.Fv/)zq7e(^/[!(e&/DCh5m\k@`WqCn9ܩYLm* EV) ?-;ݤ%EYf#7Dl.~EnegŦuQ_#w}D (=BgEj{ÿ8Dc4,2^5 SJponzUvVxP[XΜr%<OBM1?FLwPK]g;im|jV?9iXf(;DMHL0:(EQ#R"7 D}'_B6$fr#9 3ZHA]r-IR,[BTxAXvQ#a͘:gC/tK lp'ɟscttVhBu!<*5V\1)=+|,(&wD |>!lG]YB9ߩX{6..r>gd;c+4rl&JIY@lnb̹*,ʥMi W"G,]GQjYD25 J*ϗ51er@aBA6Kg[0n7*9 #%^myMg$N,)WcG0Vс  +h߳/' i:ٲlRZiՒ1ьԾS*Y{n藷W?RH)$m=UbTbe"}dWyTHΊ^].. DR( h'pPMTZ͜ H2?:Vܻ,ҠgpF^">?39Xx M0n L", sک=' y+OI(Ҥߝ8L0[Q!|& 9NhWgS"_B^CD 3WÈ)7^ڡpU< ˃@>@ۿXDHUί{`Jz %7鷠_1mq۵E` ܆&i\j#N$"- C. _u BRh؀(X5rRΣdrƮە!ä@.L!Ԏ1e^XVG>.$]T`iWhl0J: #,YVŎQ58F/ρ`6Ewm~e՟_֫aZߟhn6Y'pa\~#0>amnW*!JֶJڅ}:6[ ?y?מj~j;޺ ?ډ/k/9+?j2G[FP~&{cs\~^ D+}:>z}y]{7{GzESl[|[\Zs/z{[F}}[|-s?x2eq[Hb_H;՗+}|q'Òb=z#Z&Gt^~ЭXZ~(|[4C,/_Dvv'LчGt?)w#Qc5ܩ KIד˖q˧ k1~/˻=1s_^p|~kha7~_LSC_'nh&I[-sLgϡnv9=ʅƗIENDB`:DyK 1http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqHED.htmlyK bhttp://www.aoml.noaa.gov/hrd/tcfaq/tcfaqHED.html [>@>Normal dCJOJPJQJmH f@f Heading 1,H1 $$ & Fd<@&5CJKHOJPJQJnH V@V Heading 2$ & Fd<@&5OJPJQJnH V@V Heading 3$ & Fd<@&5OJPJQJnH X@X Heading 4$ & Fd<@&56OJPJQJnH N@N Heading 5 & Fd<@&OJPJQJnH R@R Heading 6 & Fd<@&6OJPJQJnH R@R Heading 7 & Fd<@&CJOJPJQJnH V@V Heading 8 & Fd<@&6CJOJPJQJnH V @V Heading 9 & Fd<@&6CJOJPJQJnH <A@<Default Paragraph Font(O(worddef15CJ@B@@ Body Text $dOJPJQJnH (U@( Hyperlink>*B*@O"@applbox dCJOJPJQJnH &)@1& Page NumberF@BFHeaderd !OJPJQJnH F @RFFooterd !OJPJQJnH NobN Normal (Web)dddB*OJPJQJ0H.-.(-.W-"""%v%&(J89;<?L0LP$p5>b^c&>%%&l&&%'''B((78K889q99<::;a;;<j<<"=>3?e?? @.@zFL0L()ij@Z%&B'(%7+899:r;2<<'??zF0L I z 0HX %!!OR$9!Z >|6b$d8kMW6M7BSb$d8kMW6M7|R$9!Z >%@( `d   c dA . `wp3d.gifS`TS`T  S A C:\My Documents\My Non-Research Publications\Textbook\Tropical\norcross.jpgB S  ?'0H# @@# Ta!l!Y'd'-- H.H1Hx==-N5Nʆt H.H1HSteven Ackerman*C:\TEMP\AutoRecovery save of dickchap8.asdSteven Ackerman*C:\TEMP\AutoRecovery save of dickchap8.asdSteven Ackerman*C:\TEMP\AutoRecovery save of dickchap8.asdSteven Ackerman*C:\TEMP\AutoRecovery save of dickchap8.asdSteven Ackerman*C:\TEMP\AutoRecovery save of dickchap8.asdSteven Ackerman*C:\TEMP\AutoRecovery save of dickchap8.asdSteven Ackerman!D:\ack\book\revised\dickchap8.docSteven Ackerman!D:\ack\book\revised\dickchap8.docSteven Ackerman!D:\ack\book\revised\dickchap8.docSteven Ackerman!D:\ack\book\revised\dickchap8.doc+q{ Z',d@o"X Um8ƾl"  ^* LX>i!b5%Mds_(~sq/+*G,u+ -,2/ Z)0^^X9]0 {VG7N;H<4OI8`}J@+_JlXtxT9T]Y BcZ4V* f[FC}*v^vwvbIbg ,[hi!;Y`{BcZ f[@@833;H<;H<|n3Ysq/+sq/+3`}J`}J؊3T9TT9T(3* * 35%5%L3  p3bgl"mm63 Kn Kn73$ko$ko3Z)0Z)03}*v^}*v^PR3.)p.)pR3XxXxR3n3vbvbo3-,-,lo3_J_J73 3 @hh.D3 4@h8)3 @h hOJQJo(\3 @hh.3`b@hh.3`g@hh.43`d@hh.ċ3`b@h8.l63`g@hh.63 @hh.(3`1@hh.T3`@hh.3 @hh. R3 @h8.\R3 @h8.R3`s@hh.93 @hh.n3`i@h 8OJQJo((o3 9@hh.xo3 0@hh.73`,@hh.+@!!̊!!0H@GTimes New Roman5Symbol3& Arial;Wingdings3TIMES?5 Courier New"qhLIJf-t/~A!20d.L H(Chapter 8 Atmosphere-Ocean Interactions Richard MorelSteven AckermanOh+'0 ( D P \hpx)Chapter 8 Atmosphere-Ocean InteractionsordhapRichard MorelosichNormal Steven Ackerman102Microsoft Word 8.0r@N@xO@:eAt/~՜.+,D՜.+,d  hp  Strong House, Inc.A.Lj )Chapter 8 Atmosphere-Ocean Interactions Title(RZ _PID_GUID _PID_HLINKSAN{CE778A80-8BB1-11D4-98B7-C3DC19B6F366}AII1http://www.aoml.noaa.gov/hrd/tcfaq/tcfaqHED.htmlCH figures\chap9\bengulf.gif wp3d.gifE{LC:\My Documents\My Non-Research Publications\Textbook\Tropical\norcross.jpg  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      !#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~     Root Entry F$PeAData "t1TableEWordDocumentBSummaryInformation(DocumentSummaryInformation8CompObjjObjectPoolPeAPeA  FMicrosoft Word Document MSWordDocWord.Document.89q