UNDERSTANDING SOLAR TERRESTRIAL REPORTS PART I - MORPHOLOGICAL ANALYSIS OF PHENOMENA REVISION 1.1 Cary Oler Solar Terrestrial Dispatch Box 357, Stirling, Alberta, Canada, T0K 2E0 ABSTRACT This document is intended to aid those who are interested in interpreting and using the material presented in the various solar terrestrial reports that are posted over the networks. This document has been written under the assumption that the reader is unfamiliar with the terrestrial impacts of solar-related activity. It is therefore not intended for those who already have a knowledge of solar physics and/or geophysics. Some of the terms contained in this document are undefined. For definitions of undefined terms, request the "Glossary of Solar Terrestrial Terms" from "oler@hg.uleth.ca". There are two parts to this document, split into two completely separate sections. This first part describes the morphology of solar and terrestrial phenomena. Part I is fairly extensive and is intended to give the reader enough background knowledge to understand, interpret and apply the information presented in part II. Part II discusses the format and proper interpretation of the solar terrestrial reports. They should be read in proper sequence in order to be best understood. May 7, 1991 UNDERSTANDING SOLAR TERRESTRIAL REPORTS PART I - MORPHOLOGICAL ANALYSIS OF PHENOMENA REVISION 1.1 Cary Oler Solar Terrestrial Dispatch Box 357, Stirling, Alberta, Canada, T0K 2E0 1. Introduction In March of 1989, some spectacularly powerful solar terrestrial events occurred. An very complex and active solar region erupted with almost unprecedented levels of activity. Flare activity broke records that were held for over 30 years. Intensely severe geomagnetic storm- ing induced electrical currents in powerlines, which resulted in a total collapse of the Hydro Electric power distribution network in Quebec. This resulted in the loss of electrical power for over 6 mil- lion Canadians. Telecommunications equipment experienced powerful voltage surges on the power supply lines of transatlantic fiber-optic cables in excess of 700 volts. Oil pipelines experienced rapid strong variations in pipe-to-soil potentials, producing electrolytic corro- sion at flaws in the pipeline coating. Radio propagation was severely effected by both strong PCA activity and severe geomagnetic storming. HF radio signals were completely blacked out over many global loca- tions, and remained at very poor levels for at least 24 hours. Auroral activity was easily visible as far south as Florida (and even further). Some satellites, unable to withstand the rapid fluctuations in solar wind pressure, began tumbling out of control. It is well known that solar activity has an astonishing influence on terrestrial Earth-based systems. The events of March 1989 will long be remembered as a prime example of the power and influence the sun can have on our environment and activities. The solar terrestrial reports and associated alerts/warnings have been posted over the networks in order to aid in the prediction of terrestrial conditions which might be expected from solar and other related activity. This document is intended to help explain the nature of these various reports so that application of the data con- tained therein can be properly applied to the various fields which can be affected. Part I of this document will examine the basic physics behind such solar phenomena as sunspots, flares, and coronal holes in terms that should be easily understood by the layman. Following this, a basic overview of the geomagnetic field and some of its important features will be discussed. The characteristics of radio wave May 7, 1991 - 2 - propagation will then be explored for VLF, HF and VHF signals as they relate to geomagnetic and auroral activity. Following this, the characteristics and behavior of auroral activity will be considered in conjunction with astronomical observations and magnetic fluctuations. Concluding this section will be a discussion on the impact of severe geomagnetic storms. This discussion will include the effects of strong magnetic storming on such environments as electrical power dis- tribution networks, atmospheric circulation, and radio communications. Part II of this document will delve into the format of the solar terrestrial reports. The proper interpretation of the predictions and various charts contained in the weekly Solar Terrestrial Forecast and Review will be discussed. An examination of the flare alerts and warn- ings will then be conducted, followed by an analysis of the geomag- netic and auroral storm alerts which are posted when necessary. Con- cluding this section will be a brief overview of the material covered in parts I and II of this document. Hopefully, this document will be cohesive and interesting enough to be of value to those who are seri- ous about examining the relationships between solar activity and ter- restrial impacts. 2. Characteristics of the Sun The sun is a dynamic, complex object that we are only now begin- ning to understand. It has been a source of study and wonderment for centuries. Although many questions have been raised regarding its influence on the Earth, aside from the fact that it is our primary source of energy, only during the past century have real achievements been made toward understanding its intricate nature and influence on our environment. We now know, for example, that the sun has regular, fairly con- stant cycles. Through persistent observations and meticulous record- keeping, we know that the sun runs through cycles of activity with periods of about 11 years. We know that the sun also has a longer, 22 year cycle in which the magnetic polarity of the solar poles actually reverse sign. We know that the sun is a rotating sphere that com- pletes one revolution approximately once every 27-28 days. We also know that areas near the solar poles rotate at slower velocities and therefore take longer to complete one revolution than areas near the solar equator (this has been termed "differential rotation"). It has long been known that visibly dark regions often plague the surface of the sun. The ancient Chinese noticed these dark spots at least 15 centuries ago. Early solar astronomers noticed, over time, that the number of spots observed on the sun vary in cyclic patterns. They also noticed that the number of aurorae that were seen were posi- tively correlated with the number of sunspots observed on the sun. It wasn't until the first satellites investigated the domain of space, that we began to realize the intricate nature and varying forms of activity that occur on the sun. We discovered and studied what is called the solar corona, a great expanse of superheated rarefied gas extending outward many solar radii from the solar surface. We May 7, 1991 - 3 - examined in great detail the morphology of solar flares, one of the most powerful natural explosions known to man. Our investigations have revealed a great abundance of radiations emitted by the sun, much of which is filtered out by our terrestrial atmosphere. We have developed new techniques of studying the sun at different optical wavelengths, which has given us a wealth of new information regarding the physics of phenomena seen on the sun. We have witnessed many forms of activity: prominences, filaments, plages, faculae, granules, and many other forms of activity. The first step in understanding the relationships between solar activity and terrestrial phenomena is to obtain a knowledge of some of the basic characteristics of the sun and its attendant activity. In this first section, we will attempt to cover enough material to explain some of the properties and relationships required for strong interactions between the sun and the Earth. 2.1. Sunspots and the Solar Flux Galileo is credited as being the first person to discover sun- spots telescopically around the year 1610. He immediately noted the presence of black spots on the bright surface of the sun. It was also observed that these spots moved across the surface of the sun and gra- dually changed shapes from day to day. These spots puzzled solar astronomers for years and resulted in some interesting, although incorrect, theories regarding their origin. Through time, we have developed better instruments to resolve the features of sunspots. This has increased our factual knowledge of sunspots which has allowed us to refine our theories to model sunspots more accurately. We now know that sunspots are cool regions of the sun. They are regions approximately 2,000 Kelvin cooler than the surrounding surface of the sun. The suns surface (called the photosphere), is about 5,800 degrees Kelvin. The cooler temperatures within sunspots are what cause them to appear darker than the surrounding photosphere. In actuality, a sunspot separated from the sun and placed in the black- ness of space would radiate a great deal of energy, and would appear as an intense source of bright white light. It is only due to the contrasting temperature of the suns surface that cause sunspots to appear dark. The dark central portion of a sunspot is called the umbra and is most often associated with a less dark region called the penumbra. Sunspots vary greatly in size and complexity, but are generally around 37,000 kilometers in diameter. Sunspots almost always form in groups of two or more. Sunspots are regions of intense magnetic fields. The magnetic fields originate from deep within the sun and gradually propagate out- ward. When they reach the surface, they cause the gases within the intense core of the magnetic field to cool. Magnetic fields forming a sunspot often curve around and re-enter the sun at another nearby May 7, 1991 - 4 - location. At each point where the magnetic field enters or exits the sun, a sunspot is formed. Sunspots rotate in the same direction as the sun, and at the same (or nearly the same) speed as the surrounding gases. Sunspots near the solar poles therefore take longer to complete one revolution than do spots near the solar equator. Sunspots generally take about 27 days to complete one revolution near the solar equator. This increases to over 35 days for sunspots existing at high solar lati- tudes. Sunspots have distinct lifecycles. Although the lifetime of a single spot can extend for many weeks (sometimes months), the activity cycle of sunspots is very distinct. Sunspots begin as small specks (called "pores") which often grow rapidly into larger more distinct spots. As they grow in size, they develope penumbral regions sur- rounding the dark umbral core. Nearby, other sunspots often form and grow simultaneously. Unlike the cooler temperatures within the spots themselves, the outer regions of the spots (outside of the penumbral zones) are often superheated and appear brighter than the rest of the photosphere. These brighter regions, called faculae mark the presence of strong magnetic fields near the surface of the sun. These are all characteristics of a maturing sunspot group. As a sunspot group ages, the spots within the group begin to spread apart and drift away from what was once a compact cluster of spots. This spreading is caused by the drift of the associated mag- netic fields away from the central spawning region. As the magnetic fields drift, they often decrease in intensity and diffuse into weaker regions. Sunspots begin to fade away and disappear. Eventually, all of the sunspots fade away and die, leaving only a brighter patch of faculae marking the region that once was an active sunspot forming region. Over time, even the faculae disappear, leaving no trace that sunspots once existed over that region. This life cycle is apparent in many sunspot groups that forms. However, not all sunspot groups behave this way. Some groups of sun- spots never reach full maturity before perishing. Others may sustain mature configurations for weeks before beginning to show signs of decay. And still others may exhibit multiple cycles of growth and decay before finally dying. Although the morphology of sunspots varies dramatically, the general life cycle above applies in most cases. Sunspots are sources of enhanced radiation emissions. For the person dependent on long-distance radio communications, sunspots can help provide the enhanced radiation necessary to provide excellent radio conditions over long-distance signal paths. The radiation emitted by sunspots are most often concentrated within groups of sunspots lying in relatively close proximity to each other. The radiation covers a host of different wavelengths from Gamma rays all the way down to radio-waves. The radiation emitted by all of the sunspot groups visible on the sun are measured by a variety May 7, 1991 - 5 - of instruments. Satellites constantly monitor the radiation levels from the sun which cannot be measured from the ground (due to the filtering effect of the Earths atmosphere). Radiation which does reach ground levels are measured by sensitive radio receivers tuned to those wavelengths. One of the wavelengths of radiation which does penetrate the Earths atmosphere down to ground-levels is the 2800 MHz band (or the 10.7 centimeter wavelength band). This intensity of noise (ex. the intensity of the radiation) emitted from the sun on this wavelength is measured daily by the Algonquin Radio Observatory in Ottawa Canada. The intensity measurements obtained from this observatory are broad- cast world-wide by radio station WWV (and all other related stations) in the form of a solar flux. This solar flux represents the intensity of the solar radiation being measured at the Earths surface from the sun. The solar flux is fairly critical in radio communications work. It has been found that the radiation intensity at 2800 MHz correlates fairly well with the ionization levels at altitudes sensitive to high frequency (HF) radio communications. High solar flux values generally translate into better radio communications. It also generally marks a period of better long-distance communications using higher frequen- cies. The maximum usable frequency (MUF) during periods of high solar flux often exceed 50 MHz, providing long-range communications capabil- ities for operators using very high frequencies (VHF). The solar flux (and hence, the radiation intensity from the sun at 2800 MHz) is very dependent on the number of sunspots. Large sun- spot groups can produce steep increases in the solar flux. Solar flux values in excess of 300 are indicative of extensive sunspot activity and may coincide with very good long-range communications on HF and perhaps even lower VHF frequencies. Low solar flux values below 100 are usually indicative of periods where very little sunspot activity is visible on the sun. The solar flux is therefore an excellent means for monitoring sunspot activity. Increases in solar flux indicate the emergence or growth of sunspot areas on the sun, while decreases in the solar flux indicate the disappearance or death of sunspots on the sun. Since the sun rotates with a period averaging approximately 27-28 days, it is reasonable to question whether or not the number of sun- spots visible on the sun fluctuate with a period of around 27-28 days. This is in fact, true. The rotation of the sun often causes sunspots to rotate out of view and then reappear on the opposite side of the sun about 14 days later. We say 14 days later because by the time a sunspot group has rotated out of view, it has already completed half of its rotation period. So it only takes 14 days for a sunspot group to rotate to the opposite side of the sun and back into view. This cyclic behavior is also manifest in the solar flux. Because the solar flux is dependent on sunspot activity, the value of the solar flux often fluctuates in tandem with sunspot activity. As a sunspot group rotates out of view, the solar flux decreases in value May 7, 1991 - 6 - (sometimes dramatically if the sunspot group is extensive). Approxi- mately 14 days later, the same sunspot group may (assuming it doesn't die) rotate back into view on the opposite side of the sun, with an attendant increase in the solar flux. This cyclic pattern can be easily seen when the solar flux is plotted over time. The Solar Terrestrial Forecast and Review plots the solar flux graphically over a period of 60 days. By observing the cyclic pattern, it is relatively easy to determine approximately when the next peak will occur. Using this information, enhanced general radio conditions can also be predicted with relatively good accuracy. As will be seen in later sections, however, the quality of radio con- ditions depends on much more than simply the solar flux. 2.2. The Sunspot Cycle Just as the number of sunspots fluctuate with periods of near 27-28 days, the sun exhibits a longer period cycle which directly effects the population of sunspots that form over the entire surface of the sun. This cycle has been called the sunspot cycle since the primary effect of the cycle is on sunspot activity. To discern this longer cycle, it is necessary to plot the number of sunspots observed on the surface of the sun persistently for a period of about 11 years. If this is done, it becomes apparent that the number of sunspots which form and become visible decrease to a minimum over a period of about 6 to 8 years, followed by a fairly rapid increase to a peak over a period of about 3 to 5 years. This cyclic behavior represents the sunspot cycle. The solar flux likewise follows an 11 year cycle. But since the solar flux represents (at least in part) the quality of radio communi- cations (ex. distance and stability of communications), radio commun- ications also follow a cyclic pattern that is in phase with the sun- spot cycle. There are many other aspects of solar activity which closely fol- low the sunspot cycle. These other forms of activity will be covered in later sections. 2.3. The 22 Year Solar Cycle Superimposed on the 11 year solar cycle is yet another cycle with a period of about 22 years. This cycle is primarily magnetic in nature and can be seen only by observing the polarity of the solar poles. The sun has an extensive magnetic field which reaches far out into interplanetary space. If a compass were held while standing on the sun, the compass needle would deflect and point towards the north and south solar poles just as it does for us here on Earth. However, unlike the Earth, the suns magnetic poles reverse polarity approxi- mately once every 11 years for a total period of 22 years. That is, once every 11 years, a person standing on the sun with a compass would May 7, 1991 - 7 - notice the needle reversing directions. Another 11 years later, the direction of the compass needle would reverse directions again, point- ing back in the same direction as it originally did when first observed. The characteristics of this cycle were first noted by Hale[1] and Hale and Nicholson[2]. It is a fairly important cycle as will be explained below. As was seen in section 2.1, sunspots are intimately linked to magnetic fields. This 22 year cycle affects the polarity of the sun- spots that are formed in the northern and southern solar hemispheres. It also affects the polarity of the interplanetary magnetic field which is detected and measured from space by spacecraft. Near the minimum of each solar cycle, the polarity of the solar magnetic poles reverse sign. This does not occur suddenly. It can be a rather slow process. Often, the solar poles become the same polar- ity before the full reversal process completes. When the northern solar hemisphere has a northern-magnetic pole, sunspots which form in that hemisphere have opposite magnetic characteristics to sunspots which are formed in the southern hemisphere. After the poles reverse magnetic polarity, the sunspots which form in the northern hemisphere likewise reverse magnetic characteristics. This cycle is important because it affects almost all of the mag- netic characteristics of the sun as a whole and requires changes in the way we examine sunspot groups and their behavior. 2.4. The Solar Atmosphere The suns atmosphere can be divided into three distinct regions, or layers of differing physical properties. Each of these layers are very important to those who expect to understand the phenomena which occur within the various regions. We will very briefly review the properties and characteristics of these three regions, and will note the types of phenomena which occur in the various layers of the solar atmosphere. 2.4.1. The Photosphere The solar photosphere is the lowest layer of the solar atmo- sphere. This layer resides between 200 km and 400 km deep. The pho- tosphere is responsible for contributing most of the light that we receive here at the Earth. It is the photosphere which produces the so called limb darkening effect, where the radiation intensity emitted from the sun decreases from the center of the solar disk to the edge (or limb) of the sun. As we look closer to the limbs, our line of sight approaches tangency to the solar sphere and therefore travels _________________________ .9 [1] (1908) On the probable existence of magnetic fields in sunspots. Journal of Astrophysics #28, pg. 315-343. .9 [2] (1925) The law of sunspot polarity. Journal of Astrophysics #62, pg. 270. May 7, 1991 - 8 - through a greater volume of the upper photospheric layers. Because light from the deeper layers cannot reach us from the limbs due to the thickness and absorbing characteristics of the photosphere, we do not see as deeply into the photosphere when we look at the limbs, hence the limbs appear darker than does the central solar disk. As one would expect, the temperature of the solar photosphere increases with increasing depth. In general, the effective solar pho- tospheric temperature is calculated (using Stefan's law) to be about 5780 degrees Kelvin. The photosphere represents the coolest region of the sun. The temperature increases as you look deeper into the photo- sphere, and it also increases as you travel outward away from the pho- tosphere. The average density of the photosphere is relatively small; even smaller than the density of the Earths atmosphere. In fact, the aver- age density of the photosphere is only about one thousandth that of the Earths atmosphere, yet we can only see to a very small depth in the photosphere due to the high absorption and continuous spectrum of radiation which is emitted by the photosphere. The photosphere is not a particularly smooth surface. Through observations using powerful telescopes, plumes of rising and falling gas in the photosphere have been found. These granules can be seen over the entire surface of the photosphere and range in size from about 200 km to over 1800 km. Their average size is about 700 km. They are not a long-lived phenomena. Average liftimes for granules are only about 8 to 9 minutes. Sunspots as seen with the naked eye, are viewed as they appear on the photospheric layers of the sun. Their domain, however, is not restricted to the photosphere. Indeed, they can have profound effects in the suns chromosphere as well (discussed below). 2.4.2. The Chromosphere and Spicules Immediately above the photosphere lies the chromosphere, an area of the suns atmosphere where solar flares originate. The chromosphere is much thicker than the photosphere. It resides between the upper surface of the photosphere and extends to about 12,000 km above the surface of the photosphere. There are basically three regions of the chromosphere which are defined according to the temperature stratifi- cation which occurs in that region. The lower layer of the chromo- sphere extends to an altitude of about 1000 km. The middle layer extends from 1000 km to about 4000 km, while the upper chromosphere extends from 4000 km to about 12,000 km. Temperatures increase rapidly from the lower chromosphere to the upper chromosphere. At the upper edge of the chromosphere, the temperature can increase to values in excess of 100,000 degrees Kelvin. The chromosphere is the home of another type of phenomena, called the spicule. Spicules, when viewed at the solar limb using an appropriate monochromatic filter (such as an H-alpha filter), appear as grass-like protuberances that project against the black background May 7, 1991 - 9 - of space. They occur primarily in the upper middle and upper layers of the chromosphere. 2.4.3. The Corona and Coronal Holes The highest and most diffuse region of the solar atmosphere is known as the corona. This is a region of very low density gases that are superheated to exceedingly high temperatures. It can only be seen during a total solar eclipse, or by using a special instrument called a coronograph which automatically occults the bright solar disk, in effect, simulating a total solar eclipse. The solar corona can only be seen when the bright surface of the sun is completely blocked out. The low density of the corona inhibits its ability to give off light, hence its surface brightness is only a few millionths that of the suns disk. The corona has no well defined upper boundary. When viewed using a coronograph or during a total solar eclipse, the corona can be dis- cerned to distances in excess of several solar radii. Indeed, it extends to great distances in space, out to a distance of several mil- lion kilometers, where it gradually becomes the solar wind. Whereas the temperature of the photosphere is only about 5800 K, the temperature in the solar corona soars to values in the range of 1 to 2 million degrees Kelvin. Pressure waves propagating outward from the suns convenctive zone in lower levels provide the energy that heats the suns tenuous coronal regions to such extraordinarily high temperatures. The solar corona exhibits several forms of activity. When viewed using a coronograph, bright transient features become visible. These bubble-like projections called solar transients or coronal mass ejec- tions (CMEs) are relatively short-lived and expand rapidly outward through the corona. These disturbances are associated with radio bursts that are observed here on Earth. The high temperatures in the solar corona are sufficient to pro- duce copious amounts of x-ray radiation. This wasn't discovered until the early 1970s when the Skylab mission revealed the intricate nature of the solar corona. When viewed at x-ray wavelengths from space, the inner solar corona appears blotchy, with many bright points and exten- sive areas where very little x-ray radiation appears to be emitted. These areas devoid of x-ray emissions, are called coronal holes. It has been found that the passage of these coronal holes through the central solar meridian are almost always followed within 3 to 5 days by increased geomagnetic activity here on earth. It is now known that these coronal holes are regions where the magnetic field lines from the sun are open (ie. they don't immediately curl around back to the sun, but instead escape into interplanetary space). Since the charged particles in the sun naturally follow the magnetic field lines, the charged particles are allowed to escape into interplanetary space when the magnetic field lines of the sun are open. For this reason, coronal holes are often locations where escaping high-speed streams of May 7, 1991 - 10 - charged particles from the sun are allowed to impinge on the Earths space-environment, causing increased geomagnetic activity and occas- sional magnetic storms. Coronal holes most often reside near the solar poles, where the magnetic field lines extend radially out into interplanetary space. It is believed that the density of charged particles and also the speed of the solar wind are increased over these regions. The Ulysses space mission will hopefully confirm these theories. The Ulysses spacecraft is on its way to Jupiter, where it will use the massive gravitational pull of the planet to slingshot the spacecraft at high velocities in an orbit that will carry it over the solar poles to measure many aspects of the environment there. The solar poles have never been seen before in any great detail. Moreover, we are not able to determine the characteristics of space over the solar poles, since the orbit of the earth never carries us beyond solar latitudes in excess of about seven degrees on either side of the solar equator. Hence, there is a significant amount of interest among solar physi- cists with regards to this mission. Near the solar poles, coronal holes do not affect the earth. Their primary effects propagate well to the north and south of the Earths orbital plane. Not until the coronal holes migrate toward the solar equator do we begin to notice the effects of coronal holes. When coronal holes migrate to solar latitudes below approximately 30 to 40 degrees, the relatively high speed streams of charged particles which emanate from these regions are able to begin to impact on our terrestrial environment. Coronal holes rotate with the sun. They are therefore capable of producing recurrent activity each time they rotate around the sun. As they rotate, they change their form. Sometimes they expand in size. Sometimes they contract and disappear. During periods of sunspot max- imum, their forms change rapidly and their recurrent effects diminish. The numerous active regions which plague the surface of the sun during sunspot maximum are blamed for the rapid changes in form, appearance and death of coronal holes. Coronal holes formed during the sunspot minimum, however, are often long-lived and may last for many solar rotations before they finally fade away or migrate back toward the solar poles. During these periods, recurrent geomagnetic activity becomes well established. 2.5. Forms of Solar Activity Among the various forms of solar activity are plages, facula, prominences, filaments and the powerful explosions known as solar flares. All of these forms of solar activity are associated with active regions (sunspots). However, their manifestations and trigger- ing mechanisms vary considerably. In the next several sections, we will briefly examine some of the properties of these phenomena, concentrating most heavily on the aspects of solar flares, erupting prominences and disappearing fila- ments, which have the most profound effects on the earth. May 7, 1991 - 11 - 2.5.1. Plages and Faculae The terms plage and facula are often used synonymously. In fact, Deslandres originally introduced the words with the phrase plage facu- laire. Since then, the terms have evolved into two similar, yet separate phenomena. The term faculae is now used to denote the bright regions seen in white light surrounding sunspots (as is noticed when sunspots are viewed near the solar limbs). Faculae are therefore, photospheric phenomena. Plages, on the other hand are chromospheric phenomena and can only be observed when viewed through an appropriate monochromatic light filter (such as an H-alpha filter). Plages and faculae are not separate phenomena. Rather, they are the same phenomena manifested at different altitudes in the solar atmosphere. Faculae may therefore be considered to extend into the chromosphere, where the same phenomena is witnessed as chromospheric plage. As a general rule, the plage outlives its associated facula, often by several weeks. Both types of activity form around active regions and can extend to quite large distances around the active region. Plage and faculae do not extend as far north and south as they do east and west. Their east-west extensions cause their apparent shapes to become elongated. They typically follow in the steps of the active regions and are always the last optical phenomena to disappear when an active region dies.[3] Faculae contain bright granules which combine to form coarse mot- tles having diameters of about 5,000 km. These mottles tend to string together into chains. These chains of mottles are what compose the faculae. The temperature in the upper photosphere where the faculae form is higher than the surrounding photosphere. Also, the tempera- tures in deeper layers of the photosphere over the faculae tend to be lower than the upper photospheric layers. For these reasons, faculae do not exhibit limb-darkening when viewed near the solar limbs. They also disappear from view when seen away from the solar limbs under white light, for the same reasons. The associated chromospheric plage can be viewed against the solar disk or near the solar limbs when seen through an appropriate monochromatic filter. By observing the chromospheric plage through appropriate filters, we have been able to determine the characteris- tics of plage associated with active regions. For example, it is known that plage and/or faculae which form away from active regions do not live as long nor do they attain the sizes and intensities found in the regions surrounding active sunspot groups. .9 _________________________ .9 [3] The magnetic fields associated with the active regions are ulti- mately the last detectable remnants to fade away. The magnetic fields therefore, outlive plages and facula. May 7, 1991 - 12 - 2.5.2. Prominences and Filaments Prominences are structures seen protruding from the relatively cool chromosphere into the hot corona. They typically extend to heights of 30,000 or 40,000 km above the chromosphere, but can attain heights as high as 100,000 km in some cases. Prominences are only seen near the solar limbs. When prominences are viewed against the solar disk, the name changes to a filament. As prominences rotate into view such that they are seen against the solar disk, they appear as long stringy dark filaments that can stretch for distances up to 200,000 km. Although long in appearance, their widths are usually relatively small, near about 6000 km. Prominences and filaments vary considerably in dimen- sions. They can be very small, the size of chromospheric spicules, or very large as was mentioned above. Prominences form both near active regions and away from active regions over apparently quiet areas of the solar surface. Prominences which originate away from centers of activity are generally known as quiescent prominences, and are usually less active and live longer than prominences which form near active regions. Active prominences are those which form near active regions. The fluctuating energy out- put and unstable environment cause active prominences to display some impressive forms of activity. Active prominences are, as a rule, associated with sunspots and occur in the earlier part of the life of a center of activity. This does not mean that quiescent prominences cannot undergo sudden changes. For example, sometimes a quiescent prominence starts to rise slowly, but rises faster in the middle than at the ends, thus developing into an arch. As the arch expands at an increasingly higher velocity, attaining several hundred km/sec, the material disperses and fades to invisibility. Such eruptive prom- inences have been known to reach heights of 1.5 million kilometers above the solar limb. When seen on the disk as filaments, eruptive prominences are represented by the sudden disappearance of the fila- ment (or a disappearing filament). Disappearing filaments (and thus, eruptive prominences) can release huge quantities of energy which can produce terrestrial impacts here on the earth. Erupting prominences and disappearing filaments are one and the same, only viewed at different positions on the solar disk. The majority of eruptions or disappearances are only temporary. Usually, the original prominence reforms over the same region and in nearly the same configuration within a few days. There are many different types of prominences associated with varying levels of solar activity. Prominences of greatest interest to us are surge-type and loop-type prominences, which are manifestations of unusually energetic solar activity. Flares often produce surge and loop type prominences. The typical surge-type prominence is a confined jet of material rising out of the chromosphere with a velocity of several hundred km/sec to a height of some tens of thousands of kilometers. After May 7, 1991 - 13 - reaching a maximum height, the material usually falls back to the sur- face along nearly the same path as the outgoing matter. Like most prominences, surges show fine structures in the form of threads of luminous matter. Several surges can occur in the same region and using the same trajectories as other surges. The lifetimes of most surges are short, usually lasting only a few minutes, although they have been known to endure for several hours. Loop-type prominences are likewise, associated with considerable amounts of flare and coronal activity. The prominence loops often form from bright knots or arcs at considerable heights above the limb, perhaps 100,000 km. Material streams down along two main curved arteries, and soon the prominence takes on a true loop shape, with the two arms meeting in a single point near, if not in, a sunspot. Loops usually last a few hours or less. At the end of their lives, they fade and disintegrate. Quite often, the last visible features are the high, now fainter, knots from which they originated. Loop prominences exhibit a peculiar spectral line called the coronal yellow line. This spectral type (made from Ca XV) indicates that the temperature of the medium surrounding the loop is high. Moreover, the spectra of the loop prominences themselves point to temperatures as high as almost any found among prominences and bear a close resemblance to flare spectra. Although there are similarities in activity between quiescent and active prominences, active prominences are always more energetic and have higher temperatures. Quiescent prominences have kinetic tempera- tures of around 6,000 to 15,000 degrees Kelvin, while active prom- inences may have temperatures that exceed 25,000 to 50,000 degrees Kelvin. Loop and surge type prominences most often exhibit these higher temperatures. The average lifetime of a filament is about 25 days. Compare this with the lifetime of a quiescent prominence, which can last up to eight or nine solar rotations. Quiescent prominences are therefore, considerably more stable unless an active region forms near a quies- cent prominence. Filaments tend to migrate toward the nearest heliographic pole. They form near the sunspot-forming zones and proceed to travel toward the poles. As they travel, shorter filaments often combine with longer filaments to form a very long filament chain. Many filaments do not manage to make it to the solar poles. Indeed, active regions can completely annihilate filaments which wander into their domain. Filaments can also simply disintegrate over time. However, the gen- eral tendency is for poleward movement of the filaments. The high latitude filament zone becomes most prominent during the sunspot minimum years. During these years, the polar filament zone, known as the polar crown, continues to move poleward during the new solar cycle. Polar filaments are characteristically more stable than filaments near the sunspot forming zone (nearer to the solar equator - ranging from about 30 degrees latitude during sunspot minimum to about 5 degrees in latitude during sunspot maximum). May 7, 1991 - 14 - 2.5.3. Solar Flares One of the most powerful natural explosions known to man is the solar flare. This relatively short-lived explosion occurs over com- plex sunspot groups. They can be immensely powerful. A large solar flare can release energy equivalent to a 10 billion megaton bomb. Solar flares can be devastating to our terrestrial environment. Among some of the effects which are experienced in and around the earth are bombardments of huge doses of ultraviolet radiation, which have been linked to global reductions in the ozone concentrations which protect us from hard ultraviolet radiation. Flares can send out vast quantities of highly energetic protons which can penetrate our Earths atmosphere to tropospheric heights. Some powerful flares have been well correlated with anomalies in atmospheric circulation, affecting our weather and climate for relatively short periods of time. Flares have completely knocked out radio communications over long distances and have caused significant disruptions in ground-to- satellite and satellite-to-ground communications. The massive inter- planetary shockwaves which can propagate outward from powerful solar flares can create exceedingly intense geomagnetic storms which can cause a multitude of problems, such as a lack of compass accuracy, loss of radio communications, and heavy currents induced into long conductive elements such as pipelines, railway tracks, telecommunica- tions cables, and electrical power transmission lines. Strong geomag- netic storms have caused electrical power transformers to explode, large-scale blackouts for millions of people, and a great many electr- ical brownouts and surges. The shockwaves from solar flares (sudden changes in the velocity, density and pressure of the solar wind) have caused satellites to begin tumbling out of control. The highly charged particles which engulf the environment of satellites have also damaged the electronic components in some satellites. Indeed, solar flares can have a profound influence on our terrestrial environment. Solar flares may be defined as a sudden release of energy causing a sudden brightening of the chromosphere. It is important to note that flares do not occur at the surface of the photosphere (the area that we discern as the surface with our eyes). Flares are chromos- pheric phenomena, and as such, occur above the photospheric regions. The energy released by solar flares comes from magnetic energy which has been stored and accumulated over time in an active region. Generally, solar flares require strong magnetic gradients. This is particularly true for the more powerful class of flares known as pro- ton flares. The process whereby flares occur is basically as follows. An active region forms and develops. As it developes, the magnetic fields associated with the sunspot group intensify. Gradients between opposite poles of the magnetic fields associated with the active region increase. This process may be represented by an elastic that is stretched over time to near the breaking point. At some point, the elastic suddenly snaps, releasing all of its stored energy in a very short time. The sudden release of energy that was pent up in the May 7, 1991 - 15 - magnetic fields causes a sudden and intensive explosion which superheats the chromosphere and nearby regions to temperatures of near 5 million degrees Kelvin. Particles are often explosively ejected from the sun at this time, being accelerated to near relativisitic speeds within fractions of a second. These types of flares are known as proton flares and can have a strong influence on our terrestrial environment. The extremely high temperatures emit high doses of x-ray and ultraviolet radiation. Within eight minutes, the x-ray and ultra- violet radiation reaches the earth, causing instantaneous and abnor- mally high levels of ionization in the ionosphere, which consequently affects radio communications. Within about an hour, the highly- accellerated high-energy solar protons traverse the vast distance from the sun and slam into the earth. Many of the high-energy particles are redirected by the Earths magnetic field to the polar regions where they may penetrate to ground levels and cause a polar cap absorption event (or PCA). The unusually high proton density of the space environment at satellite altitudes are called satellite proton events and are responsible for causing satellite communication disruptions and potential damage to satellite systems. The massive explosions from flares may last from only a few minutes to many hours. The huge conservatively rated class X-15 flare of March 6, 1989 maintained its explosive power for ten hours, com- pared to the more typical 30 minutes for flares. It was an exception- ally powerful flare, perhaps the most powerful flare ever recorded. Flares are not usually visible in white light. That is, we can't normally see flares with our naked eyes. The majority of light released by major flares occur in a region of the spectrum that requires a monochromatic light filter (such as an H-alpha filter) to be seen. Only in rare cases, during particularly intense flares, can they be seen in white light. These cases are reserved for the rogue flares, which superheat the photosphere and cause simultaneous bright- enings of the photosphere. These brightenings can be seen in white light, but last only momentarily. Flares are therefore, not usually seen in white light since most flares do not attain the high tempera- tures necessary to superheat the photosphere to levels that can be detected in white light. It typically requires approximately 36 to 48 hours for a powerful flare to produce significant geophysical events. By calculating the time it takes for flare-related impacts to affect the earth, the velo- city of the travelling solar material can be calculated. Generally, the higher the velocity of the material, the more severe the terres- trial impacts tend to be. Flares which eject matter at speeds suffi- cient to cross the sun-earth boundary in 24 hours are capable of pro- ducing profound terrestrial effects. However, particle velocities are not the only aspects which must be considered. Interplanetary mag- netic fields and plasma densities are also important factors, but will not be discussed here in any great detail. Suffice it to say that plasma densities (that is, the density of the cloud of material ejected by flares) that are relatively high tend to produce strong effects at the Earth. Likewise, the magnetic fields contained in the cloud of particles ejected by flares have effects on geomagnetic May 7, 1991 - 16 - activity. Interactions between the Earths magnetic field and the mag- netic fields in the cloud of particles can cause field lines to cou- ple, link and destroy each other. This process releases vast quanti- ties of energy and heat into the Earths atmosphere which causes both auroral activity and intense magnetic storms. More will be said on this in later sections. 2.5.4. Polar Cap Absorption Events Perhaps one of the most astonishing influences of large solar flares are the polar cap absorption events (also known as PCA events or PCAs). PCAs occur shortly after the eruption of a powerful proton flare. The proton flare ejects large quantities of solar protons at high velocities towards the earth. Within a few hours, these high- energy particles arrive at the earth. Since the particles which arrive at the earth have an electrical charge, they are influenced by the magnetic field of the earth. The Earths magnetic field effec- tively steers the high-energy protons to the north and south geomag- netic poles. Here, the particles slam into the ionosphere at very high speeds. Their energy permits them to penetrate to deep levels in the Earths atmosphere. As they penetrate, they collide with consti- tuents of the Earths atmosphere. When they do so, they ionize it. This ionization prevents radio signals from being reflected by normal ionospheric refraction. Hence, long distance radio communications are severely inhibited during PCA events over the polar regions. The intense ionization which occurs during strong PCA events are usually confined to the polar regions. However, the latitudinal dependence of PCA-related ionization is strongly dependent on the intensity of the event. Particularly intense PCAs may cause radio blackouts for regions down to geographical latitudes of near 50 degrees. Thus, middle latitude regions may also be affected by PCA events. The intensity of PCA events is measured at polar stations using instruments called Riometers (Relative Ionospheric Opacity meters). These basically measure the transparency of the Earths ionosphere. During PCA events, absorption of extra-terrestrial radio signals (ex. cosmic noise) is enhanced and the corresponding decrease in signal intensities is recorded by this instrument. A PCA occurs when the absorption detected by the riometer exceeds 2.0 dB during daylight hours or 0.5 dB during the night. PCAs usually reach a peak absorp- tion level soon after the flare and may require several days (perhaps up to several weeks) to return to preflare levels. PCAs also produce ground level events (GLE), where the penetrat- ing solar particles actually reach ground levels briefly over polar regions. These events are detected using instruments called neutron monitors. When the neutron monitor trace increases by 5% or more above normal background levels, a ground level event is said to be in progress. Associated with GLEs are phenomena called Forbush Decrease Events (or Forbush Decreases). These events are also measured by neutron May 7, 1991 - 17 - monitors and are defined when the neutron monitor trace decreases 5% or more below normal background levels. Forbush decreases and GLEs are usually associated with large geomagnetic storms (discussed in later sections). 2.5.5. Significance of Sweep Frequency Events It has been known for years that the sun emits radio waves over a wide range of frequencies. Although solar radio astronomy began in 1942, it never really became a serious area of research until after the second world war in 1945 and 1946. The years of research have yielded some interesting results, some of which we will examine in this section. The sun radiates three types of radio emission. (1) The constant background continuum of the quiet sun, observed throughout the radio spectrum, caused by thermal emission in the chromosphere and corona. (2) The slowly varying component, most readily observed at wavelengths of 3 to 60 cm. This component is associated with sunspots and plages. (3) The transient enhanced radiations, including noise storms and the several types of burst radiations. The radio burst radiations which we will concentrate on in the following sections have specific characteristics that allow them to be separated into groups or types. We will concentrate on the radio emissions identified as Type II and Type IV sweep frequency events. The term sweep frequency is used to describe the behavior of the radio emissions as observed at earth. These emissions consist of intensified regions of the radio spectrum which drift (or sweep) from higher to lower frequencies. For example, during a major flare, a Type II radio sweep means that during the flare, part of the radio spectrum observed intensified (ie. the noise became louder) and drifted from high frequencies down to lower frequencies. This is what is meant by a sweep frequency event. There are basically five major types of radio emissions which are commonly categorized. These types are categorized using roman numerals and depict different aspects and phenomena occuring on the sun at radio wavelengths. The following sections very briefly cover the slowly varying component, as well as emissions of types I, III and V. A more extensive analysis of the slow drift bursts (types II and IV) will follow, as they pertain more to the occurrence of major geomagnetic storms than the other types. 2.5.5.1. The Slowly Varying Component Radio frequency radiation from the sun has a characteristic minimum base-level which is generated by the thermal processes occur- ring in the sun. Over a period of days, this base-level radio emis- sion can be observed to increase or decrease in intensity. These changes comprise the slowly varying component. In 1959, Covington[4] _________________________ .9 [4] (1959) Solar emission at 10 cm wavelength, Paris Symposium on .9 May 7, 1991 - 18 - showed that the monthly average of the emission intensity varies in phase with the solar cycle. In fact, it was observed that the slowly varying component is closely associated with sunspots and plages. Shortly thereafter, it was discovered that for the 20 cm radiation, the maxima of the solar radio emission represented the area overlying the brightest plage areas rather than sunspots. Since the slowly varying component occurs at wavelengths ranging from 3 cm to over 100 cm, the range in height of the originating emis- sion above the chromosphere can be considerable, from 10,000 to 300,000 km. At longer wavelengths, the slowly varying component begins to interact with radio bursts which originate higher in the corona. The greatest effect of the slowly varying component is observed over the frequencies of 7 to 60 cm. The slowly varying component is not particularly important in determining potential terrestrial impacts such as geomagnetic storms. They are, however important in determining the potential activity and intensity of specific active regions (or of the entire visible solar disk as a whole). The solar flux (at a wavelength of 10.7 cm) represents the slowly varying component and is very useful in deter- mining the activity of the sun as a whole. 2.5.5.2. Type I Bursts and Radio Noise Storms Radio noise storms are violent increases in the intensity of noise originating from solar coronal regions. Noise storms are gen- erally comprised of many (hundreds to thousands) of discrete bursts of noise, which have been identified and named as Type I bursts, or storm bursts. Radio noise storms and Type I bursts are associated with the intense magnetic fields in active regions, which rise to coronal heights and interact with the corona to produce the noise. Solar flares generally do not affect the frequency of occurrence of Type I bursts. They appear to be somewhat independent of flare phenomena and are correlated more with the magnetic fields in active regions than with flares. These types of burst radiations are not of particular importance to those interested in predicting terrestrial impacts. For more information on these types of radio emissions, consult the many books available at your public or University library regarding flares and solar radio emissions. 2.5.5.3. Type III Radio Bursts Type III radio emissions occur almost daily during the solar max- imum years. Both Type III burst and Type V bursts are associated with fast drift events. Fast drift events are those where the frequency of _________________________ Radio Astronomy, Stanford University Press, page 159. May 7, 1991 - 19 - the radio emission is observed to drift rapidly from higher frequen- cies to lower frequencies. It has been determined that the drift rate is dependent on the frequency being observed. For example, the drift rate at 200 MHz is about 150 MHz per second, while the drift rate at lower frequencies such as 25 MHz is lower, near about 4 MHz per second. These sweep frequency events are caused by outward-propagating waves which travel at high velocities ranging from 20% to 80% of the speed of light. The duration of most Type III bursts is about 30 seconds in the low frequency range, but varies with increasing fre- quency. Burst durations on the higher frequencies vary from 3 to 10 seconds at 100 MHz to less than 1 second above about 500 MHz. Type III bursts tend to occur in groups, ranging from a single burst to as many as 100 grouped together. As the number of closely- spaced bursts increases, the intensity of the observed emission like- wise increases. It has been determined that about 50 to 60 percent of Type III bursts occur within 10 minutes of the start of a flare or subflare. The greater the number of bursts in a group or the greater the inten- sity of a burst, the more probable the association with flares becomes. Aside from the facts already stated, Type III bursts do not have any significant terrestrial impacts. They can enhance atmospheric ionization, but cannot produce geomagnetic storms. 2.5.5.4. Type V Radio Burst Emissions Type V radio bursts tend to follow Type III radio bursts. This type of radiation consists of a wide-band emission of considerable intensity, particularly at the lower frequencies near 100 MHz, with durations from 30 seconds to 5 minutes. Type V bursts are usually confined to the lower frequencies, and have been observed from near 25 MHz to frequencies in excess of 150 MHz. However, most of the radia- tion remains confined to frequencies near 100 MHz. Type V burst velocities average about 3000 km/second. They are very highly correlated with solar flares. Between approximately 60 and 90 percent of all Type V radio bursts occur within about 5 minutes of the start of a flare or subflare. They are more closely correlated with subflares than flares of greater importance, but are also fre- quently observed to occur in conjunction with flares of greater impor- tance. These radio emissions are not related to geophysical phenomena produced by large flares. There is no real correlation between these types of radio bursts and significant terrestrial impacts. May 7, 1991 - 20 - 2.5.5.5. Type II Radio Bursts Type II radio bursts represent slow-drift sweep frequency events. That is, the frequencies of the radio emissions decrease rather slowly when compared to the drift rates for Type III radio bursts. Type II radio bursts are important to solar terrestrial physicists, since their occurrence can increase the risks for terrestrial impacts, par- ticularly if associated with Type IV burst emissions (discussed in the next section). Almost all Type II events are coincident with flares, although most flares do not produce Type II bursts. In fact, Type II bursts occur rather rarely and are generally only associated with flares of greater importance (ex. major flares). These bursts consist of emission in narrow frequency bands that slowly drift from high to low frequencies. The average drift rate is about 300 KHz per second at 100 MHz. As a rule, the bandwidths of Type II bursts are quite narrow, sometimes only a few MHz in the lower frequencies near 100 MHz. Most slow-drift bursts of this type fade before reaching frequencies near 25 MHz, although Type II bursts have been known to drift down to frequencies below 25 MHz. The drift of a burst from higher to lower frequencies may be interpreted as a result of the motion of the burst source through the corona. Methods have been adopted to calculate the approximate velo- cities of the burst sources through the corona. The methods relied on most heavily bring the average burst velocities to between 1000 and 1500 km per second. These values may be in error by a small amount, since the density of the coronal region where the burst source ori- ginated from must be used in the calculations and this value must be approximated from models of the corona, not from actual measurements. Type II bursts are often associated with the expulsion of solar material into interplanetary space. By calculating the approximate velocity of the material using the method mentioned above, the approx- imate intensity of terrestrial impacts can be roughly determined. If the Type II burst is clearly associated with a well-positioned flare, the probability for increased geomagnetic activity increases dramati- cally. Moreover, it has been found that magnetic activity tends to increase between 1.5 and 2.5 days after the occurrence of Type II bursts. This correlates well with ejected material travelling at speeds near 1000 km/second. 2.5.5.6. Continuum Type IV Radio Emissions Type IV radio emissions often follow the slow-drift Type II radio bursts. Type IV emissions are primarily stable emissions which do not drift in frequency very much (if at all). They have very wide bandwidths, sometimes more than eight octaves and often lie at higher frequencies than those occupied by most radio noise storms (see the section on Type I bursts). The greatest intensity of this radiation occurs at frequencies below 250 MHz. Often, Type IV emissions occur simultaneously at high and low frequencies in two separate areas of May 7, 1991 - 21 - the spectrum. Type IV bursts frequently occur in the low frequency areas between 7 MHz and 38 MHz and very often follow Type II slow- drift bursts. A high percentage of Type IV bursts coincide with solar flares and burst emissions of Type II. Generally, Type IV bursts occur in conjunction with more powerful solar flares, which is also in agree- ment with the behavior of the Type II bursts which they often follow. Further evidence of their association with major flares is the confirmed association with the occurrence of polar cap absorption events, where high energy solar protons penetrate into the Earths atmosphere. They are therefore, also associated with the powerful proton flares and often accompany the expulsion of high-speed solar protons into interplanetary space. The correlation between magnetic storms and Type IV events is exceedingly high when Type IV events are preceded by Type II radio bursts. In most cases, a Type II radio burst followed by a Type IV radio burst indicate the mass ejection of solar material into inter- planetary space. This material most often causes geomagnetic storms within 48 hours of the observed event. Moreover, the association of a Type II followed by a Type IV radio burst is very highly correlated with the occurrence of major solar flares. This is extremely helpful to the person interested in predicting potential terrestrial impacts caused by major flares. By calculating the velocity of the Type II burst and noting the intensity of both the Type II burst and the accompanying Type IV burst, the potential sever- ity of terrestrial effects can be predicted with moderate accuracy. Given the typical lag time between flare occurrences and magnetic storms, the forecaster can generally foretell the occurrence of increased magnetic activity (and therefore radio propagation and ionospheric conditions) to within a 2 to 3 day period. 3. The Earths Magnetic Field All of the objects in our solar system have magnetic fields. The earth is no exception. We have used our magnetic field for centuries as a reliable tool in navigation. Little did we realize back then how vital our magnetic field is. Without a magnetic field, the earth would be subject to harmful radiations from the sun. Life probably would not exist as it does today. The Earths magnetic field has two poles. As any boy-scout knows, a compass points toward the north and south geomagnetic poles. But there is a third component of the magnetic field that most people are unaware of. This is a vertical component. Not only does a compass needle point north and south, but it also tilts at an angle to the horizontal plane. As one moves closer toward the magnetic poles, this "dip angle" increases towards the vertical. At the magnetic poles, a compass needle would point straight up and down and the horizontal May 7, 1991 - 22 - movement (ex. the movement pointing north or south) would be unde- fined. If the magnetic field of the earth were drawn schematically on paper, it would resemble a spherical shell with lines of force pro- pagating outward from the poles and connecting over the equatorial regions. This shape is characteristic for a spherical dipole magnet. A dipole field is a good approximation of the shape of the Earths mag- netic field. However it is not a perfect representation. Some anomalies from the perfect dipole exist. But for our purposes, a spherical dipole field will suffice in describing the phenomena which occur. The solar wind has a profound influence on the shape of the Earths magnetic field. The solar wind is analagous to winds that we experience here on earth, except that the winds are created by the outflow of energy from the sun. Just like the Earths winds, solar winds can gust and fluctuate in speed. Solar flares can cause extreme gusts in both speed and pressure which can affect the stability of objects (such as satellites) in space. The pressure from the solar wind transforms our Earths magnetic field into a comet-like appearance. The "head" of the "comet" sur- rounds the earth and the "tail" extends outward away from the earth for millions of miles (well beyond the orbit of the moon). The region near the head of the magnetic field where the solar wind first makes contact with the Earths field is called the bow shock region. This is a transition zone where particles of the supersonic solar wind are abruptly slowed to subsonic speeds. Particles and radiations can be deflected around the earth by this region. It therefore serves as a type of "shield", protecting us against certain harmful radiations. The Earths magnetic field is flexible, like a bed of springs. It reacts to increased solar wind pressure by compressing inward and reacts to decreased wind pressure by expanding outward. Strong solar wind gusts created by powerful solar flares are capable of compressing the Earths magnetic field to altitudes near where geosynchronous satellites reside. Compressions of this magnitude generate enormous currents in the Earths magnetosphere which in turn spawn powerful geomagnetic storms. These storms are closely monitored around the world. Moreover, fluctuations in the speed and density of the solar wind are also responsible for producing geomagnetic storms. In the following sections, we will discuss the properties of geomagnetic storms, substorms, accompanying auroral activity and the combined effects on radio propagation. 3.1. Geomagnetic Substorms When the conditions and characteristics of the solar wind change or fluctuate rapidly, the geomagnetic field can become disturbed. Instabilities also result when interactions with solar magnetic fields occur. Instabilities in the geomagnetic field often result in the generation of electrical currents in the magnetosphere and ionosphere May 7, 1991 - 23 - which in turn, produce accompanying geomagnetic fluctuations detect- able at ground level. Geomagnetic substorms are relatively short-lived, lasting any- where from less than 30 minutes to as much as several hours. Sub- storms are most prevalent in polar and auroral-zone latitudes (lati- tudes above about 55 to 60 degrees geographic latitude - although the zones are more a function of geomagnetic latitude than geographic latitude). 3.2. Geomagnetic Storms When many substorms occur over a period of a day or two, the entire event as is called a geomagnetic storm. Intense geomagnetic storms may last many days, but most occur over a period of 24 to 48 hours. Geomagnetic storms undergo three basic stages of development. These stages are outlined as follows. First, a shock wave from the sun slams into the earth. This sud- den gust and pressure change in the solar wind produces a magnetic impulse that is detected all around the world in a matter of minutes. This magnetic impulse is called a sudden storm commencement (SSC) or sudden commencement (SC). This marks the initial phase of a magnetic storms' development. During the SSC, the intensity of the horizontal component of the Earths magnetic field increases. This increased intensity is due to the sudden compression of the Earths magnetosphere. During the next several hours, the magnetic field remains fairly steady with only minor fluctuations. Approximately three to six hours after the SSC, the second phase or main phase of the storm begins. At this time, the Earths magnetic field begins to fluctuate wildly. The main phase coincides with the arrival of the main cloud of particles that are ejected from major flares or coronal holes. It may take several days for the earth to pass through this cloud of solar debris. During this period, the earth can experience major magnetic fluctuations (substorms) all around the world. Following the main phase is the recovery phase. This phase may drag on for days following the main phase. It is a period when the geomagnetic field begins to return to normal. When the extreme condi- tions in space abate, the magnetosphere begins to relax. This phase of recovery may still experience some periods of substorm activity, but overall activity noticably decreases. Within several days, the geomagnetic field returns to normal and substorming ceases. Not all geomagnetic storms follow these stages precisely. Many storms, for example, do not begin with a sudden commencement. Many storms simply enter the main phase gradually. These types of storms are called gradual commencement storms and are generally associated May 7, 1991 - 24 - more with coronal holes than with energetic solar flares. They also tend to be less intense than storms which are associated with SSCs. Occasionally, a sudden and short-lived shock impacts with the Earths magnetic field. This sudden impulse (or SI) is usually a pre- cursor to increased geomagnetic activity, although storms preceded by sudden impulses are usually only of minor intensity (there are excep- tions to this, however). Sudden impulses occur as very-short-duration (around four minutes) pulses of increased magnetic intensity. They are easily seen on magnetometer traces as distinct short-lived, rela- tively high amplitude "bumps". 3.3. Ionospheric Effects of Geomagnetic Storms Geomagnetic storms can have a profound effect on the conditions in the ionosphere, particularly over the auroral-zone regions. Geomagnetic storms are usually caused by terrestrial interactions with solar-ejected clouds of particles. These clouds of particles are directed by the magnetic field toward the polar regions, but tend to congregate along an oval-shaped region known as the auroral zone. In this region, where the particle penetration is often the highest, ionospheric properties and characteristics fluctuate most rapidly. Besides being the primary particle penetration boundary, the auroral zone is also associated with the strongest magnetic fluctua- tions and levels of instability in the world. These zones (one in the northern and one in the southern hemisphere), are located at approxi- mately 67 degrees geomagnetic latitude (which corresponds somewhat to geographical latitudes between roughly 55 and 70 degrees). These zones often migrate equatorward as geomagnetic activity increases. Hence, for many middle-latitude locations, strong geomagnetic activity can place them directly in the heart of the auroral zone due to the migration of the auroral zone. The ionospheric properties over the auroral zone can change rapidly. This zone is the home of the auroral electrojet, which is an oval-shaped core of intense electrical current which courses through the ionospheric and magnetospheric regions. This current (along with particle precipitation) can cause large temperature anomalies at ionospheric heights. The magnetic field and ionospheric densities at ionospheric heights sensitive to radio communications likewise, undergo large fluctuations and other anomalies which can affect radio propagation conditions. The ionosphere basically consists of four distinct regions of ionization. These regions, called layers, are defined according to height. The lowest layer, called the D-region, resides at a height of about 70 to 90 km and appears only during the daylight hours when solar radiation is sufficiently high to ionize the ionosphere at these relatively low heights. The E-layer lies above the D-layer at an altitude between 90 and 150 km. This region, like the D-layer is pri- marily ionized during the day, but can remain ionized sufficiently to provide distant radio communications into the evening hours. The region above the E-layer is the F1 layer, which is located at a May 7, 1991 - 25 - distance of between 150 and 250 km. During the day, this region is distinct and separate from the last layer of the ionosphere, the F2 region, which resides at an altitude which varies between about 250 and 400 km. During the late afternoon and evening hours, the F1 and F2 regions merge into a single region of ionization simply called the F-region. Auroral activity and maximum energy deposition occurs in the auroral zone at a height of about 110 km. This coincides with the E- region of the ionosphere. During geomagnetic storms, the amount of ionization in the ionosphere over the auroral zone is often intense enough to absorb all radio signals which pass through that region. These polar blackouts are usually confined to the high latitudes and polar regions, but can slip southward with the expansion of the auroral zone equatorward. One of the most pronounced effects of geomagnetic activity on ionospheric properties is the ability for VHF radio signals to be "bounced" from regions of visual auroral activity. During intense geomagnetic storms, the dip angle of the Earths geomagnetic field at ionospheric heights over the auroral zone can deviate by several degrees. The deviation introduces a curvature in the dip-angle of the magnetic field which serves as an effective medium for bouncing VHF signals. The curvature, combined with the high levels of ionization near E-layer heights permits these high-frequency signals to be scat- terred by the ionospheric anomaly. This process is called auroral backscattering and is a primary source for long-distance VHF communi- cations. Geomagnetic storms also influence the maximum usable frequencies (MUFs) of the various ionospheric layers. The maximum usable fre- quency of the F2 region is affected most profoundly during geomagnetic storms. In most cases, the MUF decreases well below normal values at F2 layer heights. These heights happen to be most sensitive to HF long-distance communications. In some rare instances, however, the MUF of the F2 region actually increases during magnetic storms. The main phase of geomagnetic storms affect the MUF of the E-layer as well. Depressions in the MUF of the E-region are most often associated with large geomagnetic storms. The ionization which occurs at E-layer heights is also responsi- ble for another type of radio phenomenon known as sporadic E. Sporadic E occurs when cloud-like areas of enhanced ionization form at altitudes between about 90 and 150 km. These so-called "clouds" drift with time and are most prevalent during the daylight hours and during periods of geomagnetic storming over the auroral zones. Intense storming often causes abnormal increases in E-layer ionization, which can result in polar blackouts. 4. Radio Signal Propagation Radio science rests on the discoveries of Ampere, Oersted, May 7, 1991 - 26 - Faraday, and Henry, who developed the principles of electric induction and electric and magnetic fields surrounding conductors carrying current. A single unified electromagnetic theory was achieved between 1867 and 1873 by the Scottish physicist James Clerk Maxwell. In 1887, Heinrich Hertz discovered radio waves and showed that they exhibit all of the properties of light waves. In 1896, Guglielmo Marconi assembled various items of equipment developed by Hertz, D.E. Hughes, Edouard Branly, Oliver Lodge, and others, and approached interested British party's with a proposal to use the Hertzian waves for commercial communications. In 1897, the Wireless Telegraph and Signal Company was formed for this purpose, and by the end of that year, messages had been sent over a distance of 18 miles. Between 1897 and 1899, Marconi developed equipment for tuning transmitters and receivers to the same frequency to avoid interference between stations and to conserve the power of the radiated waves. Shortly thereafter, in 1900, Marconi successfully transmitted a tran- satlantic message to anxious ears in North America. Luckily, the ionospheric conditions at that time were favorable for transatlantic communications. It may have been quite a setback if their attempts at transatlantic radio communications had failed due to geomagnetic activity or solar flares. Since that first transatlantic contact, we have signficantly expanded our knowledge of ionospheric radio wave propagation. We have formulated models of ionospheric behavior in propagating radio waves and have learned of the types of solar phenomena which can have impacts on radio propagation. In this section, we will briefly exam- ine some of the more important aspects of radio propagation as it deals with VLF, HF and VHF radio waves. 4.1. Propagation of VLF Signals Very Low Frequencies (VLF) are those frequencies which range from below 1 KHz to approximately 150 KHz. These frequencies are home to the navigational beacons which transmit on these very low frequencies. Radio signals in the VLF range are affected differently than those in the HF range. VLF signals are generally enhanced during solar flare induced SIDs (sudden ionospheric disturbances). Signal strengths of VLF signals have been found to increase, often quite dramatically, during solar flares. They also tend to become enhanced during the initial phase of geomagnetic storms, but may later suffer strong absorption during the main phase. We will restrict our discussion of VLF propagation to the above, due to the inability of most people to use this band of frequencies for any useful communications. The bandwidth of VLF communications is insufficient to permit voice communications. Hence, this band of fre- quencies is not heavily used for day-to-day long-distance radio com- munications and is not of particular importance to us in this docu- ment. May 7, 1991 - 27 - 4.2. HF Signal Propagation The most widely used communications frequencies for long-distance communications are those which span the frequency range between 1.8 MHz and 30 MHz. At these frequencies, the ionosphere is capable of bending radio signals back toward the earth. This makes long-distance communications a viable possibility on the HF bands. Radio propagation on the HF bands is most dependent upon geomag- netic activity, auroral activity (both of which determine the state of the ionosphere and which are most applicable over middle and higher latitude paths), and solar activity. Solar activity can severely affect the propagation of HF radio waves through the ionosphere. Significant solar flaring can produce isolated temporary periods of radio blackout conditions. The intense levels of radiation which accompany strong solar flares ionize the Earths ionosphere over sunlit portions of the earth and produce strong absorption levels capable of completely absorbing radio signals. Flare-related radio blackouts do not occur very frequently, however, and are limited only to the rare occasions when complex solar regions form and spawn flares of unusual severity. When a major solar flare produces radio wave absorption over the sunlit portions of the earth, the phenomena is called a short wave fade (or SWF) and typically lasts between 20 and 50 minutes. Some long-duration events, however, can severely affect radio propagation for extended periods of many hours. These cases, however, are usually reserved for the large rogue flares which occur most frequently during the solar maximum years. It is important to note that SWFs do not occur over the dark side of the earth. Although there is some evidence to suggest some subtle night-time effects, they are generally restricted to the daylight hours. Strong solar proton flares frequently produce accompanying PCA and satellite proton events. As was mentioned earlier, PCA events have powerful effects on polar and high latitude radio communications. They are perhaps the most severe form of radio absorption that can occur over these latitudes. They can last for many days and can cause wide-spread polar blackouts on all radio frequencies. Their effects are not confined to the polar and high latitude regions, however. Strong events can migrate equatorward, and can engulf middle latitudes as well. Low latitudes are generally unaffected directly by PCA events. However, during periods of PCA activity, low latitudes are restricted in the latitudinal range where they can make radio con- tacts. They may be completely unable to establish contact with others at high or middle latitudes. They will almost certainly be completely unable to make contacts at polar latitudes. Likewise, signals which graze the PCA zone may be completely absorbed. HF transmissions dur- ing the daylight hours over low latitudes during PCA activity are gen- erally weaker and less reliable. Higher powers usually do compensate, but may not aid in penetrating to long-distances. Night-time May 7, 1991 - 28 - communications during PCA activity over low latitudes are usually not heavily affected. Therefore, there is a noticable diurnal pattern of increased absorption over latitudes during periods of PCA activity. Basically, all latitudes are affected by PCA activity to some degree, although high latitudes and polar regions are by far affected the most. Geomagnetic storms can be almost as devastating to high latitude and polar latitude radio transmissions as PCAs, although they are almost always less constant when compared to PCAs. That is, during geomagnetic storms, there will usually be periods of time where at least poor communications is possible. During PCAs, however, communi- cations is often completely blacked out with very few (if any) oppor- tunities for any HF propagation of radio signals. During magnetic storms, auroral activity usually abounds in the high and polar latitude regions. Middle latitudes can also experience significant periods of strong auroral activity which can severely impact radio communications. During these periods, HF radio signals can become so garbled as to be completely unintelligable. Rapid fad- ing of HF signals caused by auroral activity is called auroral flutter. Rapid fading and strongly erratic signal strengths over much of the HF spectrum can destroy attempts to communicate during auroral and geomagnetic storms. Low latitudes are again, generally better off than higher lati- tudes during geomagnetic storms. They experience less fading, less absorption and less flutter. However, even low latitudes do not escape all of the effects of geomagnetic storming. Over all lati- tudes, the MUF of the F2 region decreases (often quite dramatically). Likewise, the MUF of the E region also often decreases. Also, the lowest usable frequency (LUF) almost always increases during a geomag- netic storm. The combined effects of decreased MUF and increased LUF effectively narrow the usable HF spectrum. Often, the F layer becomes completely unusable for HF communications, as has been observed many times with ionosonde maps of the ionospheric layers. The F region may completely disappears from such maps during some intense magnetic storms. At other times, there may exist spread-F which can also strongly influence radio communications over all latitudes. Spread F is caused by the scattering of radio signals by anomalies in the F- layer region. Spread F can limit the amount of information that can be transmitted long-distances and can also produce high fading rates, limiting the ability for long-distance radio communications. The use- fulness of packet radio communications can be strongly affected by the occurrence of spread F. Ionospheric conditions during magnetic storms vary considerably over small changes in latitude and longitude. These changes modify the character of radio signals which propagate through the changing layers of the ionosphere. Radio propagation over long-distances is therefore, difficult to accomplish with any reliability or success during magnetic storms. Some very long-distance HF propagation has apparently been May 7, 1991 - 29 - accomplished in the past during storm periods, but such contacts are not very common. HF radio signals are more likely to be severely dis- torted and/or absorbed by the anomalous ionization and magnetic behavior in aurorae than to be reliably propagated to long distances via aurorae. However, for the ambitious soul willing to attempt to establish auroral-contacts, note that your best chances are via CW. Voice communications via aurorae are for the most part, very unreli- able, very unintelligable and suffer severe distortion and fading by the time they reach their destination. As will be seen in the follow- ing section, VHF radio propagation via auroral backscatter is a more reliable method of using aurorae for communications. 4.3. Long-Distance VHF Signal Propagation Under most normal conditions, long-distance VHF signal propaga- tion is next to impossible. Frequencies of 144 MHz are almost always well beyond the critical frequencies for the ionospheric layers. Attempts to transmit VHF signals long distances by the same means used for HF signals will prove fruitless in most cases. Frequencies transmitted to the ionosphere simply pass through it and out into space. Only under special conditions are VHF signals capable of being transmitted long-distances via ionospheric properties. Probably one of the best known methods whereby this is accom- plished is via sporadic-E. As was mentioned in previous sections, sporadic E occurs when isolated areas of enhanced ionization drift into the area. Radio signals of unusually high frequencies are able to be refracted or scattered by these localized "ionization clouds" back to the earth from E-region heights. These clouds are sporadic in nature. Hence any communications accomplished is likewise only tem- porary. There are several other conditions that have yielded fairly good long-distance VHF communications. However, determining when these conditions will occur is almost as difficult as predicting sporadic E. Solar flares which produce SIDs often generate the enhanced ionization levels required for long-distance VHF communications. However, such communications are only possible over locations where SIDs are observed. SIDs occur only over the sunlit areas of the Earth. They also occur with less intensity over higher latitudes where the eleva- tion of the sun makes a shallower angle with the horizon than at lower latitudes. Season therefore, plays an important role in the inten- sity, duration and frequency of SIDs for VHF propagation. Low lati- tudes generally have better luck in propagating VHF signals using the enhanced ionization produced during SIDs than high latitudes. Middle latitudes are also generally good for such types of propagation, but effectiveness decreases during the winter months due to the decreased elevation angle of the sun. High latitudes generally do not experi- ence significant SID-related propagation possibilities on VHF frequen- cies during the winter months. However, the prospects improve dramat- ically during the summer months. The only other major form of potential VHF communications takes place during auroral and geomagnetic storms. Propagation via aurorae May 7, 1991 - 30 - on VHF frequencies is called auroral backscattering if long-distance contacts are made as a result of the radio signal bouncing off of the aurora. Likewise, forward scattering occurs when signals scatter off of the aurorae in a forward direction toward the polar regions. Two- way auroral communications on VHF frequencies is called bistatic auroral backscatter communications. It is important to note that "scattering" does not mean "refrac- tion." It means radio signals are literally scattered off of anomalies in the ionosphere near regions of auroral activity. Some- times signals are scattered backwards. Sometimes they are scattered forwards. In rare cases where auroral geometry is just right, VHF sig- nals can be scattered multiple times off of multiple aurorae to achieve significant long-distance communications. However, in these cases, the quality of the radio signal decreases dramatically with each contact of the scattering source. Scattered VHF signals can be discerned by their very gruff, motoring sounds. These types of signals are affected by very rapid fading which often fade in and out at frequencies as high as 100 Hz. These signals are said to be sputtering or caused by auroral sputter. In order to achieve auroral backscatter communications, auroral activity must be visible low in the horizon. The more intense the activity, the higher the probability for achieving long-distance back- scatter communications. Directional antennas are a definite asset, since most of the power of the transmitter must be directed toward the auroral region. The auroral region must be at a low elevation angle in order to provide the geometry required for backscattering to occur. The distance of transmissions also increases with increasing distance to the aurora. Hence, low transmission angles are required. The prospects for distant bistatic auroral backscatter communications increases if CW communication is used. CW is much more intelligable when distorted by aurorae than is voice and therefore can be under- stood even when severely distorted by auroral activity. The probability of achieving auroral backscatter communications is a function of latitude and geomagnetic activity. Lower latitudes do not experience auroral backscatter communications nearly as often as northerly middle latitudes and high latitudes where auroral activity is more prevalent. However, even at these higher latitudes, such communications depends on the extent of magnetic activity. It has been found that auroral backscatter communications only become widespread during major geomagnetic storms. Minor geomagnetic storms are capable of providing conditions necessary for isolated auroral communications, but generally the best communications possi- bilities occur when geomagnetic conditions reach major storm levels (ex. magnetic K indices of 6 or greater). Backscatter communications have two well defined diurnal peaks. The largest peak typically occurs in the late afternoon/early evening hours. This peak is not quite so dependent on geomagnetic activity, although it does appear to be somewhat sensitive to it. The second May 7, 1991 - 31 - peak occurs near local midnight, which coincides with the peak of auroral activity over most locations. This second peak appears to be heavily dependent on geomagnetic activity. Widespread backscattering has been known to occur during this second peak during periods of major geomagnetic storming. During quiet magnetic periods, the peak is almost non-existant, indicating only very rare and isolated incidents of backscatter communications. From the foregoing, it is clear that long-distance VHF propaga- tion is indeed possible, but requires special conditions before DX communication can occur. The best times for DX are in the late- afternoon and early evenings. The next best opportunities come near local midnight during minor to major geomagnetic storms. Generally, the prospects for DX increase with geomagnetic activity. This is in sharp contrast to HF communication, which is seriously eroded during periods of high geomagnetic activity. 5. Characteristics of Auroral Activity The Northern Lights (aurora borealis) or the Southern Lights (aurora australis) - hereafter referred to as aurorae - are beautiful, shimmering displays of lights in the skies. These lights have been a source of wonderment and awe for centuries. They are without a doubt, one of the most awesome displays of natural beauty known to man. Aurorae are caused by high-speed, high-energy protons and elec- trons which collide with atmospheric atoms of oxygen and nitrogen. These bombardments cause the gas in the ionosphere to become ionized and give off photons of light. The "fluorescing" gas is not unlike the gases in a fluorescent light bulb, which also become ionized and give off light when excited. Aurorae generally form at an altitude of about 100 km, within the E-region of the ionosphere. Occasionally during intense auroral storms, the lower boundary of the visible auroral forms dips down into the D-region heights slightly below the 90 km level. The height at which aurorae occur enables them to be seen for hundreds of kilometers before the curvature of the earth, light pollution, geographical obstructions or atmospheric anomalies blocks their view. The complete morphology of aurorae is complex and beyond the scope of this document. Suffice it to say that the particles which penetrate into the atmosphere are directed by the Earths magnetic field and that the main penetration belt coincides with the auroral zone. For more information, the interested reader is directed to the many available books on aurorae and magnetic storms. 5.1. Auroral Relationship with Geomagnetic Activity Auroral activity is invariably linked with geomagnetic activity. Magnetic storms are always associated with auroral activity. More- over, auroral activity is proportional to the intensity of magnetic storms. Increasingly intense magnetic storms yield increasingly May 7, 1991 - 32 - intense auroral activity. The intensity of an aurora depends on several factors. Auroral brightness, aerial extent, latitudinal extent, frequency of changing forms, pulsations and color changes are all used to determine the relative intensity of auroral activity. We say "relative intensity" because the intensity of an aurora is relative to the observer making the observation, and his or her experience in doing so. Aurorae are most frequently seen at areas that reside in or near the auroral zone, a boundary where aurorae form most frequently. Glo- bal geomagnetic activity is also highest in this zone. The locus of auroral activity has been determined to reside near a geomagnetic latitude of about 67 degrees. Areas between approximately 65 and 70 degrees geomagnetic latitude are generally considered to be within the auroral zone (with some diurnal exceptions which will not be con- sidered here). The auroral zone contains the electrojet, an area within the auroral zone where high electrical currents surge through the ionos- pheric and magnetospheric regions. This electrojet is responsible for the majority of magnetic perturbations which occur in that region. The particularly strong anomalous behavior of the electrojet (as well as other current systems) during magnetic storms is what causes the intense magnetic fluctuations which are observed in and near the auroral zone. Even during periods of quiet magnetic activity, fluc- tuations in the auroral zone can be many times greater than fluctua- tions outside of the zone. It is now clear that the auroral zone carries more meaning than simply the definition of the zone where aurorae occur most frequently. It is also the zone where magnetic activity is highest, where particle penetration into the atmosphere peaks, where anomalies of the iono- sphere are most severe, and where atmospheric electrical induction becomes most pronounced. Auroral activity in the auroral zone does not usually become dis- tinctly visible until the geomagnetic field becomes unsettled. The threshold for observing auroral activity increases with increasing distance equatorward of the auroral zone. For example, middle lati- tudes generally require at least active geomagnetic conditions before any auroral activity can be discerned over the horizon. Minor storm- ing usually provides good opportunities for auroral observations at middle and high latitudes. Low latitudes are generally incapable of viewing the auroral activity until major to severe geomagnetic storms occur. During periods of major geomagnetic storming, the auroral zone migrates equatorward and often resides over the Canada/U.S. border and into the northern U.S.. These periods are usually associated with sustained K-indices of six or more over the middle latitudes. With increasing activity, the visibility of auroral activity becomes possi- ble at progressively lower latitudes. It should be noted that the behavior of the southern auroral zone is no different than the northern auroral zone. Therefore, areas of May 7, 1991 - 33 - Australia, New Zealand, etc., can apply these characteristics equivalently. 5.2. Significance of Aurorae to Astronomers Considering the intrinsic brightness of aurorae, their occurrence can be an annoyance to astronomers. Bright aurorae associated with strong magnetic activity can obscure most of the sky. Moreover, their brightnesses can easily exceed the brightness of most stars. Aurorae therefore, pose as a threat to the observing astronomer. Astronomers usually attempt to get as high above the atmosphere as possible to observe stars. However, even above all of the clouds and major atmospheric constituents, auroral activity can remain an annoying interference since their occurrence in the atmosphere is at an altitude of between 90 km and several hundred km's. Luckily, how- ever, most of the high-altitude observing sites are in the low- latitude regions, where aurorae occur relatively infrequently. Aurorae can, on the other hand, be a real treat for the astrono- mer who searches for them and enjoys observing them. Aurorae can pro- vide a significant amount of excitement. The activity in aurorae is often remarkable. Huge and rapid changes in color, brightness and form can all contribute to the spectacular events which can be observed in aurorae. Activity peaks when aurorae are seen directly overhead. Large, wavelike pulsations of light become easily visible when seen overhead. These flaming aurora are often intensely bright and are constantly in motion. Bursts of auroral activity (associated with magnetospheric substorms) can dramatically increase the bright- ness and intensity of auroral activity within minutes. The combined brightness of auroral activity during intense auroral storms often surpasses the light given off by the full-moon. It is no wonder many astronomers often greet auroral activity with smiles and cheers. 5.3. Auroral Classifications There are several ways of classifying aurorae. They can be clas- sified according to shape, brightness, activity and even color. For most purposes, however, classifications according to shape and activity are enough. Aurorae can occur in a near-infinite number of shapes and sizes. There are, however, forms which are more commonly seen. These forms have been given names to help identify them. The zenith aurorae is best known near and in the auroral zones where aurora are seen throughout the sky, and directly overhead. As it implies, zenith aurorae are aurorae which occur directly overhead. They appear as a closely packed cluster of "beams" or "rays" which often change rapidly in shape, brightness and orientation. They often appear almost three-dimensional and are one of the more active forms of aurorae. The color of zenith aurorae vary considerably with time. Rapid and intense color fluctuations are often associated with these type of aurorae. May 7, 1991 - 34 - A well known auroral form is the curtain aurora. These aurorae are observed away from the zenith (either to the north or the south) and resemble curtains or drapes hung from the sky. They often change in shape moderately quickly. Particularly intense segments of curtain aurorae often drift eastward or westwards. The direction of drift is closely related to the time that the observations are made. Unlike the zenith aurorae, curtain aurorae are a relatively stable form that may persist for hours (although their shapes may change continually throughout their existence). The color of curtain aurorae vary, but are most often seen as greenish-white with occasional tinges of red or pink. Closely related to the curtain aurora is the flaming aurora. Flaming aurora are basically curtain aurora which pulsate rapidly in brightness. The pulsations take on wave-like characteristics which resemble flames of fire. The wavelike pulsations propagate from the curtain aurora upward toward the zenith from all directions. Often, these pulsations converge at the zenith where diffuse aurora of pul- sating shapes become visible. The flaming aurora have been mistaken for huge fires occurring in distant lands by people in the times of the Roman Empire. There was one instance where a Roman Emperor sent out men and equipment to find and extinguish a fire they thought had engulfed a distant castle. Little did they know that the fire was a flaming aurora associated with a strong magnetic storm. The pulsating aurora is a general term applied to auroral shapes which exhibit pulsations. Pulsating aurora do not generally occur until geomagnetic activity reaches minor to major storm levels. They are characteristics of intense ionospheric ionization and tend to coincide closely with magnetospheric substorms (ie. periods of intense magnetic fluctuations and enhanced auroral activity). Diffuse aurorae are most prominent during periods of low to moderate geomagnetic activity. They are usually the first to be seen prior to auroral and magnetospheric storms. During periods of per- sistent magnetic activity, diffuse aurorae may remain visible for days over the horizon. High latitudes are usually able to discern shapes, patterns and or slight pulsations in diffuse aurorae, but such activity is usually of low intensity. These types of aurora are gen- erally inactive and dull forms of auroral activity. Auroral arcs are moderately bright ropes of light that arc across the sky. They can form near the boundary of the auroral zone and the subauroral zone (the region just outside of the auroral zone). Arcs are generally relatively inactive and don't usually exhibit pulsations or rapid color fluctuations. They do, however, undergo occasionally large changes in brightness. The brightness intensifications usually precede periods of enhanced auroral and magnetic activity. The arcs are therefore, often good for indicating when enhanced auroral activity might be expected. The time between an arc brightening and enhanced auroral activity may range from under less than one minute to more than five minutes. Their brightenings are, however, well corre- lated with increased auroral and geomagnetic activity coinciding with magnetic substorms. May 7, 1991 - 35 - These are the major forms of auroral activity which are observed. Although these definitions do not nearly encompass all of the possible forms of auroral activity (each auroral event can differ from others), they encompass most of the major types of common auroral structures. For a definition of the classification of auroral activity, consult the document "Glossary of Solar Terrestrial Terms" available upon request from: oler@hg.uleth.ca. 6. The Impacts of Geomagnetic Storms and Solar Activity Severe geomagnetic storms are relatively rare, occurring most frequently during the maximum of the solar cycle and least frequently during the minimum of solar activity. They are strongly correlated with major solar flares, which explains their solar cycle dependence. Magnetic fluctuations during severe geomagnetic storms often sur- passes 2,000 nanoteslas (gammas), which is the smallest, most commonly used unit of measuring magnetic field strengths. Fluctuations this large over a period of minutes is enough to cause significant effects to terrestrial ground-based systems. Industries which can be hit par- ticularly hard are the electrical generation utilities, communications networks, and companies managing large pipelines or other long conduc- tive objects. Recent research is also revealing a causitive relation- ship between large geomagnetic storms and changes in atmospheric cir- culation. In the following sections, we will attempt to cover some of the relationships between strong geomagnetic storms and impacts with these terrestrial systems. We will also point out some of the more impor- tant research which has been done with regards to solar and geophysi- cal activity on atmospheric circulation. It should be noted that some of the following material may be considered inconclusive and still under research. The reader is warned that the material which follows is of a technical nature and therefore may not be clearly understood. An attempt will be made to pad the discussion with sufficient references to provide a respectable background of information with regards to the following discussions. Please note that the following material is not essential to the understanding of the solar terrestrial reports. It may, therefore, be skipped by those who are not interested in the potential impacts of solar and geophysical activity on terrestrial systems and the environment. The discussion below has been separated into two main sections. The first section discusses the impact of magnetic storms on very long ground-based conductive objects such as electrical powerlines, pipe- lines, railway networks and telecommunications networks. The princi- ples discussed apply to most of these fields. Emphasis is placed on the electrical power generation industry, which can strongly affect the terrestrial community as a whole. The second section discusses the impact of severe magnetic storms and strong solar flares on May 7, 1991 - 36 - atmospheric circulation, which is still in a "gray" area with regards to conclusiveness. 6.1. Magnetic Storm Induction The principle by which intense magnetic fluctuations induce currents into long conductive objects has been extensively studied over the last several decades. The principles are well understood and have been extensively verified by numerous researchers. During major to severe geomagnetic storms, the geomagnetic field exhibits very strong fluctuations in intensity. These fluctuations are caused by strong electrical currents residing in the ionosphere and deep inside the Earth. During these storms, electrical currents are able to flow through the grounded neutral lead of large power transformers and into the power system. These induced currents in the neutral lead causes additional magnetic fields to develope and reside in the core of these large transformers. The presence of these mag- netic fields in the core of the transformer produce spikes in the AC waveform in the transformer (caused by the addition of the normal mag- netic fields with the induced magnetic fields). These spikes produce harmonics which can trip protective relays. They also cause the transformer to operate less efficiently. This lack of efficiency can significantly increase the amount of current drawn by the transformer, effectively placing an additional load on the power system. If the harmonics occur for a sufficiently long period of time, physical dam- age to the transformer can occur. For example, the major magnetic storm of March 13 and 14, 1989 induced electrical currents on many of the electrical power distribu- tion networks in Canada. Induced currents measured by Ontario Hydro during this storm were about 80 amperes/phase. Newfoundland and Labrador Hydro Electric Utilities witnessed geomagnetically induced currents as high as 150 amps/phase. Hydro Quebec experienced magneti- cally induced currents powerful enough to saturate transformers. The transformer saturations generated harmonics which tripped protective relays on static compensators. The loss of power caused by these events (of near 9,450 Megawatts) overloaded the rest of the system within seconds and resulted in a total collapse. The ensuing power blackout lasted about nine hours and affected over six million people in Quebec. This storm had many effects on the electrical power indus- try. Many stations experienced numerous power fluctuations, voltage depressions and surges. The effects of geomagnetic storms on long conductive objects have been studied since the beginning of this century. Since then, many authors have elaborated on the characteristics and principles whereby such phenomena occur. For a good (although technical) discussion of these principles and characteristics, consult the papers by Camp- bell[5] , Watanabe and Shier[6] , Anderson et al.[7] , Lanzerotti and _________________________ .9 [5] (1986) An interpretation of induced electric currents in long pipelines caused by natural geomagnetic sources of the upper atmo- sphere; Surveys in Geophysics, vol. 8, pages 239-259. .9 May 7, 1991 - 37 - Gregori[8] , P.R. Barnes and J.W. Van Dyke[9] , D.H. Boteler[10] , and again by D.H. Boteler.[11] In previous years, telecommunications cables have been damaged by magnetic storms. Damage was reported in 1958 and again in 1972 during severe geomagnetic storms. These lines were made of conductive metal and carried magnetically-induced currents through the lines to equip- ment connected to them. The damage sustained in previous years has been large, despite methods to protect them against induced currents. Recently however, transatlantic telecommunications cable has been replaced with fibre-optic lines, which are not conductive. During the major magnetic storm of March 1989, the fibre-optic cable itself sus- tained no damage and experienced no problems. However, the power- supply lines which accompany the fibre-optic cables and are conduc- tive, sustained damaging voltage surges as high as 700 volts during the March 1989 magnetic storm. Pipelines experience the same kinds of damaging effects as occur on powerlines and telecommunications cables. Protective equipment on pipelines are used to prevent rogue surges from damaging the pipelines through excessive electrolytic corrosion at weak points in the pipe- line coating. During the March 1989 storm, these protective systems were rendered useless on many pipelines due to the excessive currents which were produced during the storm. Some electrolytic corrosion undoubtably occurred on many pipelines as a result. The effects of strong geomagnetic storms on terrestrial systems is well known. The power and magnitude of their influence can, at times, be remarkable (as was manifest by the large power blackout in Quebec during the last severe global geomagnetic storm). Industry is slowly devising ways and equipment to cope with strong magnetic per- turbations, but is still a long ways away from immunity to such natural events. .9 _________________________ [6] (1982) Magnetic storm effects on power transmission systems; Geomagnetic Bulletin, no. 2-82, Earth Physics Branch, Ottawa. .9 [7] (1974) The effects of geomagnetic storms on electrical power systems; IEEE Transactions on Power Apparatus and Systems, vol. PAS- 93, no. 4, pages 1030-1044. .9 [8] (1986) Telluric currents: the natural environment and interac- tions with man-made systems; in The Earths Electrical Environment, U.S. NRC Report. .9 [9] (November 1990) Economic Consequences of Geomagnetic Storms (a summary); IEEE Power Engineering Review, November 1990. .9 [10] (1979) The Problem of Solar Induced Currents; presented at the I.S.T.P. Workshop in Boulder, Colorado in April, 1979. .9 [11] (1991) Predicting Geomagnetic Disturbances on Power Systems; EOS, April 2, 1991. May 7, 1991 - 38 - 6.2. Atmospheric Circulation Modifications For decades, researchers have been attempting to determine whether large solar events and correspondingly large geophysical activity affect the global atmospheric circulation of the earth. A great deal of research has been done in this respect, and further research is still needed in order to qualitatively confirm anomalies produced by any geophysical or solar activity. In this section, we will touch on some of the aspects of geophysical and solar activity which apparently have been well-correlated with changes in atmospheric circulation. The physical mechanisms for such changes are not well known, and certainly in many cases are still heavily disputed. How- ever, the correlations achieved in previous research cannot be easily dismissed. We therefore, expect the reader to understand the nature of this section and treat it as inconclusive, yet correlated evidence. For more information, we trust the interested reader will consult the papers and publications cited herein. 6.2.1. Atmospheric Pressure Responses to Solar Flares A pronounced cellular structure of pressure change was discovered by Schuurmans[12] , who calculated the difference in the 500 mb height before and after a major flare. A total of 53 cases were originally studied, which was later expanded to 81 cases by Schuurmans and Oort.[13] The flare threshold level was chosen to be of optical class 2B or greater. Flares of class 2B or greater were therefore included in this study. Data from 1020 observation locations were used to pro- vide coverage of most of the northern hemisphere. Regions of increased 500 mb height rise were observed near the longitudes 50W, 115W, 150W, 165E, 135E, and 5E. Height decreases were observed near 35W, 175W, 145E, and 85E. The most pronounced changes were areas in the middle latitude zones (40 to 60 degrees) with cellular groupings most apparent near the coastal regions. The height differences were observed to be mostly negative poleward of about 70 degrees latitude. The apparent cellular structure of pressure change following major solar flares was also detected in studies performed by Duell and Duell.[14] Using data collected by Duell and Duell, Schuurmans and Oort performed a critical statistical analysis on the accumulated data and concluded that "the central values in the main areas of height fall and height rise are probably meaningful and thus not due to pure chance." .9 _________________________ .9 [12] (1965) Influence of solar flare particles on the general circu- lations of the atmosphere. Nature, no. 205, beginning on page 167. .9 [13] (1969) A statistical study of pressure changes in the tropo- sphere and lower stratosphere after strong solar flares. Pure Applied Geophysics, no. 75, pages 233-246. .9 [14] (1948) The behavior of atmospheric pressure during and after solar particle invasions and solar ultraviolet invasions. Smithsonian Miscellaneous Collection 110, no. 8. May 7, 1991 - 39 - Schuurmans and Oort continued with an analysis of the pressure changes which occurred in the vertical plane before and after major flares of class 2B or greater. They found that maximum flare response was found to occur at the 300 mb level, at least along the 60 degree north latitude parallel between longitudes of approximately 0 to 70 degrees west. The greatest average change of +4.7 gpdm was found at the 300 mb level over the North Atlantic by a ship positioned at 56.6 N, 51.0 W. At higher elevations, maximum response was noted to occur approximately six hours after flare time. At the Earths surface (approx. 1000 mb), the atmospheric changes lagged the flare time by about two days. Along with the pressure-height changes which were observed over the North Atlantic regions, a fairly significant change in the verti- cal temperature distribution was also observed over these regions. A maximum change of near +1.1 degrees Celcius was observed at the 500 mb level, and a maximum decrease of about -1.8 degrees Celcius was observed at the 200 mb level. The strongest temperature gradients were observed near the 300 mb level where the change in pressure was greatest. The speed of the geostrophic wind flow increased notably at the 500 mb level in latitudes from 55 to 75 degrees north by about 0.5 m/s. Near the 50 degree north latitude zone, a decrease in geos- trophic wind flow by about 0.4 m/s was observed. Seasonally, the cellular structure which was found by Schuurmans and Oort changes very little. However, the largest changes in height were found in the winter and the smallest changes were observed during the summer. Considering the large changes in pressure at the 8 km height level down to the surface over the North Atlantic, formed after major flares, one would expect a mass transport of air downward. In an attempt to determine the validity of this hypothesis, Reiter[15] measured the concentrations of tracer elements Be^7 and P^32 at Zugspitze, which is located at an elevation of 2.96 km. He found significantly increased concentrations of these elements on the second day following major flares of importance 2B or greater. According to Reiter, these two radioactive nuclides are formed in the stratosphere by cosmic ray spallation and their increased concentra- tions at Zugspitze is an indication of a mass transport of stratos- pheric air. Reiter noted that the possibility of increased concentra- tions of the tracer elements at Zugspitze was not likely to have been generated by in situ production by enhanced solar cosmic ray fluxes associated with the flares, because the production rate would be ord- ers of magnitude too small to explain the observed nuclide concentra- tions. Furthermore, he noted that the maximum concentrations _________________________ .9 [15] (1973) Increased influx of stratospheric air into the lower troposphere after solar H-alpha and X-ray flares. Journal of Geophysi- cal Research, #78, page 6167. .9 May 7, 1991 - 40 - coincided with maximums in solar wind velocity and geomagnetic activity following the larger flares. This coincides nicely with the average arrival time of large interplanetary shockwaves for major flares of class 2B or greater. 6.2.2. Atmospheric Electrical Enhancements following Major Flares Observations and measurements of atmospheric electrical proper- ties were made during 70 major flares between 1956 and 1959 by Reiter.[16] Other investigations have been performed by Holzworth and Mozer[17], Bossolasco et al.[18] [19], Markson[20], Herman and Gold- berg[21] [22], Cobb[23], and Reiter.[24] [25] Reiter, at the Zugspitze observatory, found that both the poten- tial gradient and the air-earth current density increased beginning shortly after a major flare. The values peaked between 3 and 4 days after the flare. Measurements conducted by Cobb on Mauna Loa mountain in Hawaii a few years earlier indicated a sharp increase in both the potential gradient and the air-earth current density following solar flares and _________________________ .9 [16] (1969) Solar flares and their impact on potential gradient and air-earth current characteristics at high mountain stations. Pure Ap- plied Geophysics, #72, pages 259-267. .9 [17] (1977) Direct evidence of solar flare effects on weather relat- ed electric fields at balloon altitudes. Eos #58, page 402. .9 [18] (1972) Solar flare control of thunderstorm activity, in Studi in onore di G. Aliverti, Instituto Universitario Navale Di Napoli, page 213. .9 [19] (1973) Thunderstorm activity and interplanetary magnetic field. Revista Italiana di Geofisica #12, page 293. .9 [20] (1971) Considerations regarding solar and lunar modulation of geophysical parameters, atmospheric electricity, and thunderstorms. Pure Applied Geophysics, #84, page 161. .9 [21] (1976) Solar activity and thunderstorm occurrence. Eos #57, page 971. .9 [22] (1978) Initiation of non-tropical thunderstorms by solar ac- tivity. Journal of Atmospheric Terrestrial Physics, #40, page 121. .9 [23] (1967) Evidence of a solar influence on the atmospheric elec- tric elements at Mauna Loa Observatory. Monthly Weather Review, #95, page 12. .9 [24] (1971) Further evidence for impact of solar flares on potential gradient and air-earth current characteristics at high mountain sta- tions. Pure Applied Geophysics, #86, pages 142-158. .9 [25] (1972) Case study concerning the impact of solar activity upon potential gradient and air-earth current in the lower troposphere. Pure Applied Geophysics, #94, pages 218-225 May 7, 1991 - 41 - remained above normal for several days thereafter. Cobb's peak in potential gradient occurred at about the same time as Reiter's, 3 to 4 days after the major flares, but his air-earth current density peaked only one day after the flare. It should be noted that these observations, by Reiter and Cobb, were performed at altitudes above the mixing layer where the potential gradient and air-earth current densities do not undergo any large, localized fluctuations. Therefore, variations in these two parameters should reflect changes on a global scale. The atmospheric electrical changes which appear to occur after solar flares leads to the question of whether the occurrence of lightning frequency increases after a solar flare. With respect to this, Reiter noted a 57% increase in sferics counts maximizing about 4 days after flare-day during the years 1964 to 1967. When compared to Reiters results regarding the potential gradient over these same years, it is found that the magnitude of increases in sferics counts and in the potential gradient are comparable. Markson (1971) analyzed the occurrence frequency of thunderstorms with solar flares in the United States for the sunspot minimum years 1964 to 1965. He found a 63% increase in occurrence frequency maxim- izing about 7 days after flare eruptions. He pointed out that his maximum in the U.S. occurred about 3 days after the maximum in poten- tial gradient found by Reiter at Zugspitze. This long lag time there- fore makes it uncertain (at least, based on these results), whether United States thunderstorm activity is affected by solar activity the same as in the regions observed by Reiter. On a globabl basis, Bossolasco et al. found that thunderstorm activity increased by 50% in solar minimum years and by 70% in solar maximum years about 4 days after flare eruptions. The frequency of lightning strikes in the Mediterranean area was observed to increase markedly about 4 days after the eruption of large solar flares. Through superposed epoch analysis of the data in the foregoing, it has been established that the occurrence frequency begins a notable increase one day after the flare event and achieves a 50% increase on the 4th day. These results are in good agreement with those obtained by Reiter at the Zugspitze observatory. Data analyzed over a full solar cycle (between the years 1961 and 1971) exhibited the same results, as determined by Bossolasco et al. (1973). From these results, it appears that the air-earth current den- sity, ionospheric potential, potential gradient and the frequency of lightning strikes responds to solar flares. Enhancements in these quantities occur between 1 and 4 days after the flare eruption with the increase in lightning frequency responding the slowest. A suggested possible physical mechanism lies in the increased potential gradient around the 20 km altitude level. High energy solar protons ejected from major lares penetrate the atmosphere down to May 7, 1991 - 42 - levels as low as 20 km. The increased ionization at these levels (during intense events) enhance the conductivity above about 20 km. Below 20 km, Forbush decreases in cosmic ray intensity results in decreased conductivity. The potential gradient and ionospheric poten- tial are also alertered and the net result is a possible increase in thunderstorm activity. 6.2.3. Geomagnetic Effects on Atmospheric Pressure Based on an analysis of low-pressure trough development at the 300 mb level in the North Pacific and North America areas for the years 1956-1959, Macdonald and Roberts[26] found that, in the winter seasons, 300 mb troughs entering or forming in the Gulf of Alaska area 2 to 4 days after a major geomagnetic storm are likely to undergo much greater deepening than those entering at other times. Macdonald and Roberts[27] as well as Twitchell[28] verified that these conditions are also manifest at the 500 mb level. Roberts and Olson[29], using a vorticity area index (VAI), extended these earlier results. They defined the VAI as the area of a trough wherein the absolute vorticity was greater than or equal to 20(10^-5)/second summed with the area where it is 24(10^-5)/second. This index removes the subjectiveness from the assessment of the intensity and importance of troughs and the minimum threshold vortici- ties for the definition were selected as such because most wintertime 300 mb troughs exceed a vorticity of 20(10^-5)/second, and large ones have a substantial region exceeding the largest vorticity value. The results obtained by Roberts and Olson confirmed the earlier findings of Macdonald and Roberts. Using data spanning the years 1964 to 1971, Roberts and Olson found that there are two statistically sig- nificant periods of time when key troughs undergo a sharp rise in vor- ticity area index. The first occurs during the first three days of trough lifetime. On the average, this occurs three to five days after the start of a geomagnetic storm. It is important to note that their findings showed that 2 to 4 days must elapse between the beginning of a geomagnetic storm and the appearance of the trough in order for the effect to be observed. On occasions when less than 2 days elapsed, no VAI intensification occurred (as was later discovered by Olson et _________________________ .9 [26] (1960) Further evidence of a solar corpuscular influence on large-scale circulation at 300 mb. Journal of Geophysical Research, #65, pages 529-534. .9 [27] (1961) The effect of solar corpuscular emission on the develop- ment of large troughs in the atmosphere. Journal of Meteorology, #18, pages 116-118. .9 [28] (1963) Geomagnetic storms and 500 mb trough behavior. Bulletin of Geophysics, #13, pages 69-84. .9 [29] (1973) Geomagnetic storms and wintertime 300 mb trough develop- ment in the North Pacific-North America area. Journal of Atmospheric Science, #30, page 135. May 7, 1991 - 43 - al.[30]). The second statistically significant period of time where troughs undergo significant increases in VAI occurs about 10 days after geomagnetic storms. Asakura and Katayama[31] also discovered significant decreases in pressure and increased cyclogenesis over north-eastern coastal regions of North America. Reitan[32] noted, after analyzing data over the 20-year period 1951-1970, that the distribution of cyclonic event occurrence in Janu- ary over the northern hemisphere exhibited a maximum in the areas of the Gulf of Alaska and the northeastern coastal region of the United States. These are also the areas where Roberts and Olson found increases in VAI following geomagnetic storms. A correlation analysis was performed to analyze the association of SSC-related geomagnetic storm occurrences and the number of cyclonic events observed in the United States over the period 1951-1967, by Mayaud[33]. What was discovered was a statistically significant (94% confidence level) correlation coefficient of -.46 between SSC-related geomagnetic storms and the number of cyclonic events observed in the U.S. during the period. These results, combined with those of Roberts and Olson , suggest that, although fewer cyclonic events may occur during the sun- spot maximum years, they are larger and more intense than the more numerous ones that form in the solar minimum years. From the data which has thus far been accumulated, it appears as though the strongest meteorological effects of solar flares and geomagnetic storms occurs during the winter season in the northern hemisphere. Although the data contained in this document just barely scratches the surface of research which has been done over the years, there are still doubts whether a solar or magnetic link to terrestrial atmospheric circulation patterns actually exists. It is our impres- sion that such a link may indeed exist, but additional research is needed in order to determine the areas and physical mechanisms which link solar and/or geomagnetic activity to specific atmospheric events. Nevertheless, the research data which has accumulated over the years cannot be dismissed, for there are a great many relationships between solar activity, geomagnetic activity and atmospheric phenomena which appear to have strong correlations. Those persons with sufficient background who are interested in _________________________ .9 [30] (1975) Short term relationships between solar flares, geomag- netic storms, and tropospheric vorticity patterns. Nature, #257, page 113. .9 [31] (1958) On the relationship between solar activity and general circulation of the atmosphere. Meteorological Geophysics, #9, page 15. .9 [32] (1974) Frequencies of cyclones and cyclogenesis for North Amer- ica, 1951-1970. Monthly Weather Review, #102(12), page 861. .9 [33] (1973) A 100-year series of geomagnetic data: Indices aa, storm sudden commencements. IAGA Bulletin 33, International Union of Geo- detic Geophysics, Paris. May 7, 1991 - 44 - obtaining more information regarding the possible influences of solar activity on terrestrial atmospheric processes, are directed to obtain the book "Sun, Weather, and Climate" by John R. Herman and Richard A. Goldberg (formerly published as NASA SP-426 but recently republished by Dover Publications Inc. in book form). This document nicely sum- marizes most of the research which has been done in this area over the years and provides some convincing evidence between solar, geomagnetic and atmospheric relationships. For more recent information, the interested reader is encouraged to browse through the various journals covering this subject and the published results of numerous solar ter- restrial workshops and symposiums. 7. Concluding Remarks There are many aspects of solar physics and geophysics (not to mention atmospheric physics) which must be understood before a clear knowledge of the interactions between solar activity and terrestrial phenomena can be obtained. This document was prepared to aid in pro- viding the most basic and fundamental characteristics of solar activity and geophysical phenomena required to understand and respect the nature of the solar terrestrial reports which are posted over the networks. This document was intended to be understood by those who are unfamiliar with solar terrestrial physics. The solar terrestrial reports posted over the networks are in as simple a form as is practi- cal without losing any significant resolution of information. They are written in a form that should be easily understood once the basic principles and language become familiar. The preceding presentation was required in order to supply the interested reader with the information and language background to understand the solar terrestrial reports. Only the latter sections were directed towards those with an interest and background in geophy- sics and atmospheric physics. The rest of the material should have been interpretable by those whose backgrounds and/or interests lie in other areas. This document is not intended to be fully understood the first time through. It should be reread and digested as necessary and used (if necessary) as a reference to the solar terrestrial reports. Now that we have the background necessary to understand the solar terrestrial reports, we may begin a systematic analysis of the struc- ture and content of the reports themselves. The accompanying document (part II) will describe the solar terrestrial reports in detail with accompanying hints and procedures that may be used to extract useful and pertinent information. May 7, 1991 - i - Table of Contents Introduction .................................................... 1 Characteristics of the Sun ...................................... 2 Sunspots and the Solar Flux ..................................... 3 The Sunspot Cycle ............................................... 6 The 22 Year Solar Cycle ......................................... 6 The Solar Atmosphere ............................................ 7 The Photosphere ................................................. 7 The Chromosphere and Spicules ................................... 8 The Corona and Coronal Holes .................................... 9 Forms of Solar Activity ......................................... 10 Plages and Faculae .............................................. 11 Prominences and Filaments ....................................... 12 Solar Flares .................................................... 14 Polar Cap Absorption Events ..................................... 16 Significance of Sweep Frequency Events .......................... 17 The Slowly Varying Component .................................... 17 Type I Bursts and Radio Noise Storms ............................ 18 Type III Radio Bursts ........................................... 18 Type V Radio Burst Emissions .................................... 19 Type II Radio Bursts ............................................ 20 Continuum Type IV Radio Emissions ............................... 20 The Earths Magnetic Field ....................................... 21 Geomagnetic Substorms ........................................... 22 May 7, 1991 - ii - Geomagnetic Storms .............................................. 23 Ionospheric Effects of Geomagnetic Storms ....................... 24 Radio Signal Propagation ........................................ 25 Propagation of VLF Signals ...................................... 26 HF Signal Propagation ........................................... 27 Long-Distance VHF Signal Propagation ............................ 29 Characteristics of Auroral Activity ............................. 31 Auroral Relationship with Geomagnetic Activity .................. 31 Significance of Aurorae to Astronomers .......................... 33 Auroral Classifications ......................................... 33 The Impacts of Geomagnetic Storms and Solar Activity ............ 35 Magnetic Storm Induction ........................................ 36 Atmospheric Circulation Modifications ........................... 38 Atmospheric Pressure Responses to Solar Flares .................. 38 Atmospheric Electrical Enhancements following Major Flares ...... 40 Geomagnetic Effects on Atmospheric Pressure ..................... 42 Concluding Remarks .............................................. 44 May 7, 1991