 Acoustics and your environment. The basics of sound and highway traffic noise. Hello, I'm here to talk to you about the basics of acoustics with a specific focus on highway traffic noise. I'd like to start by defining acoustics. Acoustics is the science of sound, including its production, transmission and effects. Applications of acoustics include medical ultrasonics, underwater acoustics, architectural acoustics, active or passive noise control, non-destructive evaluation, environmental noise and many, many more. It is the application of environmental noise produced by highway traffic that will be discussed here, starting with the basics of sound. Understanding acoustics requires the understanding of sound. Sound is a vibratory disturbance created by a moving or vibrating source. As the source moves or vibrates, surrounding atoms or molecules are temporarily displaced from their normal configurations, thus forming a disturbance that moves away from the sound source. In order to better understand sound waves, it's first useful to study this animated representation of waves. In this demonstration, the sphere in the middle pulsates in and out at equal intervals, forming waves that propagate away from the sphere or the source. One way to picture this type of wave is by imagining yourself tapping your finger on the surface of water. As you tap, waves propagate away from your finger in a circular pattern. Now we're going to look at a snapshot in time of the wave animation in order to get a better understanding of wave parameters. For simplicity, the outward propagating waves can be approximated with a trigonometric sine function. The pattern of the sine wave repeats itself periodically. The wavelength of these waves, represented by the Greek letter lambda, is defined as the repetition length. As a wave propagates through an unchanging medium, its wavelength remains constant. Another parameter of a wave is its amplitude. The amplitude determines the strength of the wave. Greater disturbances at the source lead to greater strengths during propagation. For sound waves, a higher amplitude equates to a higher volume. Waves also have an associated frequency. Frequency, abbreviated with a lower case F, is defined as the number of cycles of repetition per second. In other words, frequency is the number of wavelengths that have passed by a stationary point in one second's time. The unit of frequency used here is called Hertz, abbreviated HZ. It's defined as the cycles of the wave per second. When the number of Hertz exceeds 1000, it's common to write the amount in units of kilohertz. For example, 1000 Hertz equals 1 kilohertz. Frequency is inversely proportional to wavelength. An equation relating the two parameters is C0 equals F lambda. C0 is the speed of sound in the medium. In air at a temperature of 20 degrees Celsius, the speed of sound is 343 meters per second. This equation implies that longer wavelengths are associated with lower frequencies and shorter wavelengths are associated with higher frequencies. Sound can have a single frequency component, as we earlier illustrated. Or, it can have multiple frequency components at varying amplitudes, thus making each sound distinctive. This type of sound is complex. Most real-life sounds are complex. For convenience, the frequency components of a complex sound source are very often studied in terms of octave or fractional octave bands. These bands each cover a range of frequencies and are referred to by the center frequency of the band. When the sound is complex, it can be called broadband because it encompasses many frequency bands. For octave band analysis, the entire frequency spectrum is divided into one octave bands. A list of these bands is shown in the table. The center frequency of each band is one octave higher than the previous band. Notice the center frequency's associated wavelengths. For the frequency of 31.5 hertz, the wavelength in air is 10.89 meters, or 35.72 feet. And for the frequency of 8,000 hertz, the wavelength is 0.04 meters, or 0.14 feet. We've used animation to help describe various wave parameters. Now that these parameters have been explained, let's look at a demonstration of how sound waves operate. Sound waves must propagate through some medium since it's the medium's particles that support the wave. The example medium here is air. Now a source is introduced on the left side. It is producing a pure tone or single frequency sound wave in the air. The air particles are displaced as the waves propagate away from the source. The particles themselves are just oscillating back and forth, but it can be seen that the sound waves are propagating outward. As the air particles bunch up, this forms a high positive amplitude area. And when they are most spread apart, this forms an area of high negative amplitude. In their undisturbed configurations, the particles represent an area of zero amplitude. The amplitudes correspond to a change in pressure from its ambient value. This is termed the acoustic pressure. A sine wave is again shown to better understand the wave parameters. Notice how the darker areas of the acoustic wave line up with the peaks of the sine wave. We have now defined wave parameters, including frequency, and have illustrated the important characteristics of sound waves. The discussion has focused on the sound source. It's also important to understand the receiver of the sound. As an example, sound waves travel through the air and interact with the human ear. Perfect human hearing lies in the range of approximately 20 to 20,000 hertz. Frequencies below 20 hertz and above 20 kilohertz are heard by other living creatures, but not by humans. 20 to 20,000 hertz is called the audible sound range. In for sound includes frequencies below 20 hertz. And ultrasound includes frequencies above 20 kilohertz. In order to connect frequencies with sounds we recognize, let us first listen to the piano. A piano covers a frequency range of 55 to 8,360 hertz. A middle C is at 262 hertz. Human conversation is also recognizable. Nearly all information in human speech is contained in the frequency range of 200 to 6,000 hertz. It should also be noted that human hearing is most sensitive between about 1 and 6.3 kilohertz. Now that we better understand sound, it's suitable to introduce a more subjective part of sound, noise. Although sound can be measured as a physical quantity, noise requires the judgment of a human listener. Noise is defined as any unwanted sound. Unwanted sound may be defined differently depending on the listener. However, several types of noise are understood to be objectionable. These types include highway traffic. This type of noise will be discussed later in the presentation. Loud machinery or tools. And aircraft. A better understanding of noise involves quantifying its perception. Physical measurements of the strength of the wave are explained first, followed by perceptual measurements. We talked before about the amplitude of a wave. The amplitude of a sound wave can be quantified by measuring the associated pressure disturbance. In other words, we need to measure the change in pressure from its ambient value. Again, this change in pressure is termed the acoustic pressure. Sound can be represented in terms of a physical unit of pressure. One such unit is Pascals, abbreviated PA, a unit widely used in the acoustics community. Acoustic pressure amplitudes can range from the hundred-thousandths to the hundred-thousands in Pascals. Because of the wide range of amplitudes encountered when measuring pressure, it has become convenient and customary to plot the pressure data on the more compact logarithmic scale. On this scale, the unit is the decibel, abbreviated DB. And now instead of plotting the acoustic pressure, you plot the sound pressure level. The decibel values shown here correspond to measurements in the air. The equation for converting pressure to sound pressure level, abbreviated SPL, is SPL equals ten log P over PREF squared. In this equation, P is a time-averaged pressure, and PREF is the reference pressure, a quantity that depends on the medium in which the sound wave is propagating. In water, PREF equals one times ten to the negative six Pascals, which equals one micropascal. In air, PREF equals twenty micropascals. This is the pressure value which represents the threshold of unimpaired human hearing for a one kilohertz tone. In other words, the quietest audible sound at that frequency. To get an idea of how pressure and sound pressure level relate, here is an example calculation where air is the supporting medium. We have an acoustic pressure level of 0.036 Pascals. The sound pressure level is then calculated to be about 65 decibels. This is the approximate sound pressure level for normal conversation. That sounds great. It was fun. It was the first time I've been for a number of years. You've been to England often? Years ago. Family from England. Here are other examples of sound levels starting with a very quiet sound and working our way up. A quiet suburban neighborhood will exhibit an average sound level of about 40 decibels. A gas lawnmower 31 meters or 100 feet away will create a sound level of approximately 75 decibels. A diesel truck 15 meters or 50 feet away traveling at highway speeds will achieve a maximum level of 85 decibels. A jet aircraft at an altitude of 305 meters or 1000 feet will create a maximum level of about 90 decibels. The threshold of pain for humans is between 130 and 140 decibels. The sound pressure level is often represented by a capital L. We will now use L to help explain the combining of sound pressure levels. When you need to add sound pressure levels, it is not a simple matter of algebraically adding the numbers. 80 dB plus 80 dB does not equal 160 dB. As explained earlier in the presentation, the decibel scale is logarithmic. Therefore, to combine decibel values, each must be converted to a linear scale, added, and then converted back to a logarithmic scale. Here, L1, L2, up to Ln represent the n sound levels that are to be combined. For applications requiring only integer decibel accuracy, it is also possible to apply some simple steps to calculate an approximate total sound pressure level. First, find the decibel difference between two sound pressure levels. You then add an adjustment factor to the higher of the two sound pressure levels. If the decibel difference is 0 or 1, then you add 3 dB. 2 or 3, you add 2 dB. And between or equal to 4 and 9, you add 1 dB. If the difference is 10 decibels or greater, you can safely ignore the lower level source. Under these conditions, the higher level source is said to mask the lower one. Here's an example combining three sound pressure levels. The typical procedure is to combine the smallest values first, then work your way up. So we start by combining the 80 decibel levels. Their difference is 0, so we add 3 dB to 80 dB to get 83 decibels. Then the difference of 83 dB and 90 dB is 7, so we add 1 dB to 90 dB for a grand total of 91 decibels. Now that we understand how to quantify sound, we can move on to its perception. A sound's loudness is a subjective rather than an objective description of noise. It all depends on how the sound is perceived for a particular individual. When researchers determined there was a need to find a relationship between the sound pressure level of a noise and its subjective loudness, several experiments were performed. As a result of extensive human testing, objective descriptors of loudness were constructed and applied to human perception. We can take a look at examples of descriptive changes in perception and the corresponding sound level changes. When the sound pressure level increases or decreases, some humans in a normal living environment can detect a loudness change if the sound pressure level is altered by 3 dB or more. A change of 1 or 2 dB will usually go unnoticed. A 5 dB change, on the other hand, can easily be detected by most people. Also, sound will be perceived as being twice as loud if the decibel level increases by 10 dB or will be half as loud if decreased by 10 dB. And it's common to perceive a 20 dB change as 4 times as loud or 1 quarter as loud. Here's an example of a sound becoming twice as loud. We start with the first sound, complex noise. And here's the second sound, the same audio clip, but at a different sound level. The second audio clip should sound twice as loud as the first. There was a 10 dB difference. Listen to the sounds again. We can also look at human perception of sound according to frequency. Remember from before that the audible sound range for humans is from 20 to 20,000 Hz. This is shown plotted on a logarithmic scale. Even though perfect human hearing lies in this range, we do not hear equally well at all frequencies. Earlier it was stated that human hearing is most sensitive in the range of about 1,000 to 6,300 Hz. To describe sound levels in a manner which closely approximates normal human hearing, the actual sound level measurement is modified by applying a weighting. A weighting is a response function that spans the audible frequency range. This weighting assigns to each frequency a weight that is related to the sensitivity of the ear at that frequency. Frequencies to which the human ear is less sensitive are weighted less than those to which the ear is more sensitive. The A weighting curve emphasizes frequencies in the 1,000 to 6,300 Hz range and de-emphasizes frequencies out of that range. This is the A weighting plot in tabular form for octave bands. For each center frequency shown, a dB adjustment value is given. For example, for the 125 Hz octave band, a negative 16.1 dB adjustment should be applied. The 500 Hz band requires a negative 3.2 dB adjustment. The 1,000 Hz band, which can be thought of as the reference frequency for A weighting, has no adjustment. And the 2,000 Hz band requires a positive 1.2 dB adjustment in sound pressure level. A sound pressure level with A weighting applied is often stated in units of DBA. The A weighted sound level is the most widely used measure of environmental noise and is internationally accepted. Noise can be a steady continuous sound or it may tend to fluctuate between intensely loud and quieter periods. An example of the latter is the noise produced by road traffic. It peaks with the passage of a heavy truck and will have quiet intervals when there is little or no traffic. Because of the large array of noises and the need to understand them from different perspectives, there are several ways to describe noise. Many noise descriptors are available where each is appropriate for particular circumstances. A general noise descriptor is one we've already spoken about, the A weighted sound level. The symbol for this descriptor is L sub A. More specific descriptors are the sound exposure level, the maximum sound level, the hourly equivalent sound level, the day-night average sound level, the community noise equivalent level, and the 10 percentile exceeded sound level. While all of these A weighted noise descriptors can be applied to highway traffic noise, the LAE Q1H and L10 are most often used for traffic noise analysis. LAE Q1H, called the hourly equivalent sound level, is essentially the average A weighted sound level occurring during a one hour period. To illustrate the equivalent sound level, we are first showing you the sound pressure level recorded at one location over time. This time history reveals peaks and dips caused by loud and quiet sounds during the time period. Now applying the LAE Q1H noise descriptor, the sound pressure level averages and the plot flattens out. This sound descriptor should be applied to continuous sounds, such as relatively dense highway traffic. The other noise descriptor we will discuss is L10, the A weighted 10 percentile exceeded sound level. It's the sound pressure level exceeded 10 percent of a specific time period. Here's an example. Say 50 samples were taken during a period of time. You extract the highest 10 percent of the sound pressure levels, which in this case would be 5 samples. The fifth highest sound pressure level is the 10 percentile exceeded sound level. In the first part of this video, we discussed the basics of acoustics and reviewed some common noise descriptors which can be applied to highway traffic noise. Now we're going to move on to the acoustics involved in highway traffic noise. This includes the noise source or the vehicles driving on the highway, the receiver of the noise or the people in nearby communities, and the propagation path of the sound as it travels from the source to the receiver. Let's first talk about the sound source. We'll describe two basic sound source types and the relationship to highway traffic noise. A point source is defined as a source that's essentially concentrated at a single point from which noise propagates outward in all directions. In other words, sound radiates spherically from a point source. The sound levels measured from a point source decrease at a rate of 6 dB per doubling of distance. This is a decrease due to spherical spreading. A line source radiates sound cylindrically. It can be composed of multiple point sources arranged on the line located at a perpendicular distance relative to the observer. The sound levels measured from a line source decrease at a rate of 3 dB per doubling of distance. This is a decrease due to cylindrical spreading. Two theoretical sound source types have been described, but what about actual noise generated from a highway? Well, we can consider highway traffic noise as a line source or as point sources depending on the traffic density and the listener's proximity to the highway. Individual vehicles act as point sources. Categorizing vehicles into several different groups is important to predicting noise levels. Vehicles are typically divided into these categories. Automobiles, which includes all vehicles having two axles and four tires, designated primarily for transportation of nine or fewer passengers or for transportation of cargo. Medium trucks, which includes all cargo vehicles with two axles and six tires. Heavy trucks, which includes all cargo vehicles with three or more axles. Buses, which includes all vehicles having two or three axles and are designated for transportation of nine or more passengers. And motorcycles, which includes all vehicles with two or three tires with an open air driver and or passenger compartment. In general, the loudest vehicles are in the heavy truck category. As an example, at a distance of 15 meters or 50 feet, a single heavy truck traveling at normal highway speeds can produce a maximum sound level of around 85 dB a. The quietest vehicles are in the automobile category. For the same conditions as the example heavy truck, the maximum sound level of the automobile is around 75 dB a. The major noise sources in vehicles are for trucks, the exhaust stack, and for all vehicles, the engine and tire-pavement interaction. The sound levels of individual vehicle noise are influenced by how fast the vehicle is traveling. Whereas individual vehicles are here considered as point sources, highway traffic can be considered a line source. It is composed of several vehicles or point sources closely spaced. A community is a sufficient distance from the highway to see highway traffic as a line noise source. This implies that for highway traffic noise, the sound decreases due to cylindrical spreading at a rate of 3 dB per doubling of distance away from the road. Highway noise studies are born out of concern for people. People exposed to highway noise include those who live, go to school, or work in surrounding communities, or people who actually work on the highway itself. The concern lies in their safety and well-being. The most obvious negative effect of noise is physical damage of hearing. Transportation noise levels experienced by communities and the general public, however, are normally not high enough to produce hearing damage. Other effects are the interference of noise with certain activities, such as sleeping, relaxation, study, or conversation. Although most interruptions by highway noise can be considered merely annoying, some may be considered dangerous. An example of this is the inability to hear warning signals or verbal warnings in situations involving workers next to a noisy highway. Less obvious, but nevertheless real, are the stress effects of noise. There is ample evidence that noise can cause stress and thus may be a contributor to stress-related diseases such as anxiety or heart disease. The path the sound takes when traveling from the source to a receiver can be quite complicated when taking into account ground reflections, obstructions, and meteorological effects. The received sound level depends heavily on these parameters. We can start by talking about a simple propagation path, the path going directly from the source to a receiver. Besides geometrical spreading, which causes the sound level to decrease as it propagates toward the receiver, the propagation medium itself can possess inherent losses. Atmospheric absorption is the attenuation of sound during its passage through air. For propagation distances related to highway measurements, this attenuation is only significant at very high frequencies, greater than 5,000 Hz. It is in general small compared to attenuation caused by other propagation mechanisms and can usually be neglected. Meteorological conditions can also affect the direct propagation path. These conditions include vertical temperature and wind gradients. In general, daytime weather conditions permit the ground to be heated by solar radiation. The air nearest to the ground is warmest, becoming progressively cooler with increasing height. This is called a temperature lapse. The speed of sound increases with increasing temperature. This causes sound rays to bend upward away from the earth. In turn, the noise level at the receiver decreases. At night, the ground usually cools by radiation faster than the surrounding atmosphere, thus causing the air temperature to become warmer with increasing height. This is called a temperature inversion. Under these conditions, the sound rays bend downward toward the earth. For this case, the noise level increases at the receiver. Besides the refraction due to temperature gradients, wind can also cause the sound to deviate from a straight path. When sound is propagating upwind, the ray paths curve upward as with the temperature lapse, decreasing the received noise level. And when sound is propagating downwind, the ray paths curve downward as with the temperature inversion, increasing the received noise level. The importance of meteorological effects is that sound may reach the receiver with greater intensity if it is refracted downward toward the earth and with less intensity if the sound is refracted upward away from the earth than in cases where meteorological effects are negligible. It should be noted that while atmosphere conditions can have major effects on propagation of sound over distances greater than about 100 meters or 300 feet, typical highway studies are performed within 100 meters distance from the highway in areas where the meteorological effects would be less severe. As for some other meteorological conditions, effects of fog and precipitation are generally negligible. On the other hand, atmospheric turbulence which can be generated by moving vehicles or heated pavement could potentially cause fluctuations in received sound levels or can reduce soft ground attenuation, the next topic of discussion. In addition to a simple path from the source to a receiver, sound can also reflect off the ground during propagation. The ground reflected wave can interfere with the direct, non-reflected wave to produce a net increase or decrease in sound pressure level. Increases in sound pressure level can be attributed to an acoustically hard ground such as asphalt or water located between the source and the receiver. For practical highway applications, measurements have shown a 1 to 2 dB increase for the first and second row residences adjacent to the highway. It should be noted that theory indicates that in certain situations greater increases are possible. An acoustically soft ground such as grassland, plowed earth or snow can cause a significant broadband attenuation except at low frequencies. As a general rule of thumb, for each doubling of distance, the soft ground effect attenuates the sound pressure level at the receiver by 1.5 dB. This extra attenuation applies only to incident angles of 20 degrees or less. For greater angles, the ground becomes a good reflector and can be considered acoustically hard. Another consideration when discussing ground reflections is the terrain geometry. The sound may encounter terrain that's not flat but rather terrain with small hills. For a downwardly curving ground, the sound rays stray away from the ground's surface. For an upwardly curving ground, the sound rays are forced to encounter the ground's surface. We have so far discussed several aspects of the sound propagation path and have now arrived at the last topic covered in this presentation, obstructions in the propagation path. Natural terrain features such as hills and dense woods as well as man-made features such as buildings and walls obstruct the sound path between source and receiver and in most cases reduce or attenuate noise levels for nearby receivers. The amount of attenuation provided by these objects depends on the size of the object and frequency content of the associated noise source. The term noise barrier refers to any large object that blocks the line of sight between source and receiver including the ground itself if it protrudes upward through the line of sight. A noise barrier is commonly a wall or earth berm specifically constructed for the purpose of noise reduction. This type of obstruction is the most commonly used traffic noise abatement measure. When sound encounters a barrier, there are three possible paths it takes diffracted over or around the barrier, transmitted through the barrier, or reflected by the barrier. The diffracted sound is the most important path that reaches the receiver located on the other side of the noise barrier. This is the sound that bends over the top of the barrier into the barrier's noise shadow. The frequency content of the sound is important. As discussed earlier, there's a direct relationship between the frequency and wavelength of sound. Lower frequencies have longer wavelengths and higher frequencies have shorter ones. As a result, diffraction is not uniform over all frequencies. Longer wavelengths that approach the barrier height easily bend over the top of the barrier right down to the receiver whereas shorter wavelengths bend just slightly over the top, not reaching the receiver. Diffraction also occurs around barrier ends. In these instances, it's usually only important for receivers close to the end of a barrier. The noise shadow behind a barrier is not very well defined, much different than a shadow in optics. When light hits a barrier, there's a well-defined region of darkness or light absence behind the barrier. For sound, the shadow must be considered an area of noise reduction rather than an area of noise absence. The deeper or closer to the base of the barrier a receiver is positioned in the shadow zone, the greater the associated barrier attenuation. Barrier attenuation is usually represented by the more standardized term barrier insertion loss. Barrier insertion loss is defined as the sound level at a given receiver before the construction of a barrier minus the sound level at the same receiver after the construction of the barrier. The construction of a barrier usually results in a partial loss of soft ground attenuation, a loss apparent when compared to the no-barrier case. This is due to the barrier forcing the sound to take a higher path relative to the ground plane. Physically, barrier insertion loss is the net effect of barrier diffraction combined with this partial loss of soft ground attenuation. Another way sound from the source can reach the receiver on the other side of the barrier is by a rather direct although obstructed path. When the incident sound hits the barrier, some of the acoustical energy can be transmitted through the barrier material and continue to the source. The amount of transmission depends upon factors relating to the barrier material such as the weight, stiffness, or loss factors. The angle of incidence of sound and frequency spectrum of sound. The amount of sound reduction during transmission is called the transmission loss. Typically, the transmission loss improves with increasing weight of the material. Most common materials used in barrier construction provide a transmission loss of 20 dB or better. For materials such as concrete or masonry blocks, the transmission loss values are more than sufficient, exceeding 30 dB. In other words, the sound energy transmitted through these barriers is reduced by 30 dB, rendering it effectively negligible when compared with the diffracted sound. As a general rule, the transmitted sound must be at least 10 dB lower as compared to the diffracted sound in order for it to be ignored. The remaining noise is either absorbed by the noise barrier material or reflected. The ability of a barrier surface to absorb incident sound energy is characterized by its noise reduction coefficient, abbreviated NRC. Noise reduction coefficient values are based on an average of absorption coefficients at individual frequencies and theoretically range from zero to one. Zero indicating that the surface is totally reflective or that there's no absorption. And one indicating that the material is totally absorptive. NRC values for an absorptive barrier generally range from 0.6 to 0.9. A reflective barrier surface can cause noise to affect receivers on the opposite side of the highway. It is a common perception among such communities to hear a difference in the sound after the barrier is installed. Although theory indicates greater increases for a single reflection, practical highway measurements commonly show a 1 to 2 dB increase in sound pressure level from the no barrier case due to the sound reflected off the opposing barrier. Such increases can be considered typical for the first and second row residences. While this increase may not be readily perceptible, residents in the opposite side of the highway may perceive a change in the quality of the sound. The signature of the reflected sound may differ from that of the source due to a change in frequency content upon reflection. If we have parallel barriers, instead of just a barrier on one side of the highway, the situation becomes much more complicated. For parallel barriers, sound may reflect back and forth across the roadway many times before ultimately progressing outwards toward nearby receivers. The multiple reflections increase the sound level at nearby receivers and can reduce insertion loss provided by either wall alone by 2 to 6 dB. As an example, a barrier that achieves a 10 dB insertion loss may realize only a 4 to 8 dB insertion loss if a parallel barrier is placed on the opposite side of the highway. The reduction in barrier effectiveness due to multiple reflections from parallel barriers or retaining walls is called parallel barrier degradation. Research has shown that as a general rule, if the ratio of roadway width to average height of the parallel barriers is 10 to 1 or greater, the parallel barrier degradation is less than 3 dB. Remember that decibel changes less than 3 are typically not perceivable. Solutions to parallel barrier degradation include one, applying absorbing material to the face of the barrier and two, tilting the barrier out away from the highway. When tilting a barrier, one must consider tall structures on the opposite side of the highway, so as not to adversely affect them with the reflected sound. Several design considerations need to be addressed when constructing a barrier. Following is a brief discussion of some of the acoustical design parameters. As a general rule, a wall that breaks the line of sight between the source and receiver provides 5 dB attenuation. Thereafter, you can achieve about a half dB attenuation per incremental foot of height or one and a half dB attenuation per meter. Breaking the line of sight with a barrier and getting the 5 dB reduction is simple. With this, you can achieve a 68% reduction in acoustic energy, which equates to a 30% reduction in loudness. The actual insertion loss for such a barrier could be less than 5 dB because of the partial loss of acoustically soft ground attenuation as discussed earlier. It is also usually quite feasible to achieve a 10 dB reduction using walls or berms of reasonable height. Here you would get a 90% reduction in acoustic energy and a loudness reduction of 50%. Greater barrier noise reductions are not so easily attainable. A 15 dB reduction would be very difficult and a 20 dB reduction would be nearly impossible, noise barriers are usually designed with an insertion loss goal of 10 dB in mind. Remember that insertion loss includes the barrier reduction and the partial loss of ground attenuation. Actual barrier insertion losses of between 6 and 8 dB are quite common. Also, keep in mind that a barrier will provide no insertion loss until its attenuation exceeds any loss of excess ground attenuation due to construction of the barrier. Besides obstructions that are specifically constructed for noise abatement, other large objects can interfere with sound propagation. Specifically, we are referring here to rows of buildings or large areas of dense foliage. Buildings refer to houses, offices, apartments, and other similar structures. For rows of buildings, the amount of noise reduction varies with building sizes, their spacing, and site geometry. Typically, 4.5 to 5 dB attenuation is attainable for the first row of buildings and an additional 1.5 dB for each subsequent row up to a maximum of about 10 dB. To get an idea of how gaps or openings between buildings in a row can affect the amount of insertion loss, here are some percentages and the corresponding decibel reduction. If 40 to 65% of the first row is occupied by buildings, the expected insertion loss will be about 3 dB. For 65 to 90%, the insertion loss will be about 5 dB. Lastly, for percentages greater than 90, the buildings will provide attenuation comparable to that of a barrier of similar height. In addition to constructed noise barriers and rows of buildings, areas of dense foliage can provide attenuation. Such attenuation is caused by sound scattering into the sky from trunks and limbs, affecting the middle frequencies, and leaves, affecting very high frequencies. Sound absorption by leaves is generally not substantial. In addition, some low-frequency attenuation results from ground attenuation within the wooded area, where the roots of underbrush produce acoustically absorptive soft ground. To achieve any substantial amount of attenuation due to foliage, such as trees and bushes, foliage must be at least 30 meters or 100 feet deep, in other words, a strip of 30 meter width, and dense enough to block the line of sight between source and receiver. Further, the foliage area should have a height that extends 5 meters or 16 feet above the line of sight, typically as much as 5 dB attenuation is attainable. If the foliage area is doubled in width to 60 meters or 200 feet or extended beyond that, a reduction of 10 dB but no greater is attainable. Theoretically, using foliage to reduce the noise and communities surrounding highways is appealing. However, foliage of sufficient width and density to reduce noise is not usually found along highways. Vegetation planted as part of a highway project will not provide noise abatement. Phew, that was an ear fall. This presentation has highlighted some of the basics of acoustics and highway traffic noise. Of course, many details beyond what was covered here must be taken into account to accurately describe the sound in our communities. There are ongoing studies in this field of acoustics and we can look forward to progress and years to come.