Water Quality Assessment of the Salton Sea

ISE performed a combined air and water toxics assessment of Salton Sea to ascertain general contaminant conditions and odor generation potential to a proposed nearby future development site. This was an interesting project for ISE, not only because it allowed us to break-in our new water quality test lab, but also from a scientific standpoint, because we've always been interested in what is in that water...

For those who don't know, the Salton Sea is an accidentally manmade geologic anomaly. Before 1905 the site was known informally as the Salton Sink, with references to the name dating as far back as 1815. Geologically, the Salton Sea basin is the remains of ancient Lake Cahuilla, which is a ground depression formed by volcanism and has an average elevation roughly 200 feet below sea level (the deepest point is 220 feet below sea level - only slightly higher than the deepest point in Death Valley). A severe rainstorm caused the Imperial Valley dike to be breached in 1905 resulting in the severe erosion of the surrounding earth. As the erosion continued, a waterfall formed, reaching at its maximum flow a drop of nearly 100 feet. The water from this earthen cut, and resulting hydraulic channel, filled the sink forming the sea seen today.

As can be seen in the satellite photograph to the right, this now inland sea encompasses roughly 376 square miles making it the largest lake in California. Originally the Salton Sea (which is technically an inland saline lake) was much larger than it is today, and in the early 1900's threatened to inundate rail lines and towns. In 1907 the flooding was stopped, and over the years the effects of evaporation and agricultural runoff have reached a form of equilibrium at the size the sea is today. Underneath the Salton Sea is an enormous geothermal energy reservoir, due to still active volcanism.

The problem today with the Salton Sea is that with constant evaporation, and no fresh water intake (other than highly nitrified agricultural runoff), the entire lake is not unlike a large watch glass in a laboratory setting, evaporating pure water and leaving ever increasing concentrations of chemical compounds. In terms of a modern-day environmental disaster, this area is certainly one (but that is a subject for another day).

Today the Salton Sea has a salt (NaCl²⁺) content 25% greater than the ocean (roughly 45 parts-per-thousand), in addition to being a repository for a host of other substances including nitrogen compounds, pesticides, and various metallic compounds. Its only appreciable source of water is from the agricultural areas to the southeast shown in the figure.

Here's a shot of the old faithful ISE truck (a 2004 Toyota Tacoma 4WD with locking rear differential) on the main access road. The road is good here, but later on we were driving over small sand dunes. I know that guys here at the office have sunk their Jeeps on project sites before, but we cannot recall ever getting stuck in this truck.

The roads to the Salton Sea, like many periphery roads in the area, are typically lined with agricultural uses. The next photo is of the other side of the road next to the truck. Unfortunately the slobs who also inhabit the world around us, and are too cheap to pay the dump fee, drop litter everywhere.

Remember we said that looks are deceiving. A closer inspection of the 'picturesque' shoreline shows that something is afoot. What is that 20-foot wide swath of brownish material running for miles and miles along the shoreline?

...and what is that smell???

We should zoom in the camera a little more...

The testing was divided into two teams. Dr. Tavares performed ambient air quality monitoring using a Quest AQ5001 Pro air quality meter with various arrays of electrochemical sensors. Of particular interest to the research team is hydrogen sulfide (H₂S) and free ammonia (NH₃).

But before the testing, a few moments to survey the site, let our shoes fully sink into the shoreline goo, and acclimate our noses to the smell (did I also mention that it was 101 degrees F).

Dynamic Soil Testing of Rail Vibration using Modal Analysis

ISE is routinely called upon to perform various types of dynamic monitoring and impact prediction as part of due diligence studies for new projects and infrastructure. In most cases, the applicant is interested in how far a given project needs to be set back from some type of existing vibration source so as not to produce impacts based upon an agreed-upon significance criteria.

For the project shown here, the applicant proposes redevelopment of an unused/agricultural parcel for residential purposes. The project site has prime frontage along an existing rail line that is proposed for a future commuter rail, but more importantly, is currently utilized, and will continue to be utilized, for intermittent freight operations (which would be the worst-case impact condition due to the shear mass of a freight train versus that of a much lighter commuter train).

The typical problem statement would go as follows: What is the ground vibration impact potential from freight train activity to new residential development?

	In order to analyze this problem, ISE established a controlled setting within the proposed project frontage whereby all extraneous variables were removed (such as other ambient vibration sources like cars and busses) and the speed and operational characteristics of the freight activities were known. This only left a couple of independent variables like soil damping and average reference amplitude to be determined in order for a complete picture of the ground motion to be established.
Remember though that vibration is generally defined as any oscillatory motion induced in a structure or mechanical device as a direct result of some type of input excitation. The object (either structure or machine) of interest typically has sufficient inertia (defined as the quantity ‘m’ or mass) so that by Newton’s first law of motion, its rest state is one of zero vibration with a velocity equal to zero. Input excitation, generally in the form of an applied external force or displacement, is the mechanism required to start some type of vibratory response. In order to quantify the problem at hand (which is a statement of Newton's Second Law), testing is required. Thus, we set out a series of Kinemetrics Ranger Model SS-1 moving-coil short period field seismometers. These instruments, which are terrestrial versions of the lunar seismometers developed for NASA, are direct velocity-reading devices capable of measuring inertial changes into the micro-inch-per-second range. This is equivalent to measuring the footfalls from a person one city block away. The seismometers were placed in a linear array at distances from the rail edge coinciding with an increasing geometric distance from the tracks.
	As you can see, it's good to have a really long tape measure when doing this type work and a whole lot of long BNC cables to connect everything up!!! For this test, and for most civil ground vibration testing, we are only interested in the vertical (z-direction) motion. Although the train would produce some sideways (x-direction) motion as shown in this photo and even less y-direction (along the track) motion, the predominant motion (by several orders of magnitude) is the vertical motion. This motion can have various sources and frequencies from low-frequency oscillations due to grade variations in a particular track segment to higher-frequency pulses produced by the steel wheels 'clacking' at the rail joints. Remember, were taking fully loaded freight trains here. Sir Isaac always said that a little bit of mass goes a long way...
Here's a close-up of one of the Ranger SS-1 seismometers. The seismic mass is located in the lower (fatter) cylinder section and has a total mass travel (from end to end) of a mere two millimeters!!!
	Data acquisition for the testing was performed using a pair of Stanford Research Systems Model SR 760 FFT spectrum analyzers for real-time analysis a Tektronix TDS 3014 four-channel digital oscilloscope for waveform storage. The TEK 3014 has a sampling rate of 1.25 GS/s and the ability to store 10,000 data points per sampling interval. This makes for very precise and highly resolved data.
In addition to quantifying the vibration due to existing freight train activities, the second unknown (the amount of damping present in the soil) must be determined. Damping can be thought of as a type of ‘drag force or resistance’ that is always present to some degree in an object and serves to remove energy from the vibrating system. In structures such as soils or rock, damping is generally present within the material itself and hence is called ‘material damping’. The cause of this damping is due to molecular interactions inside the material itself. Damping of surface (or Rayleigh) waves in soils, which is what we're seeking to quantify here, typically occurs as a combination of distance attenuation (radiation damping) and material damping. The latter is commonly approximated using a linear damping model that assumes the overall material damping will increase as a function of distance between the source and receiver (i.e., the more soil between the source and receiver, the greater the material damping level). For the damping test, the previous seismometer array was repositioned at known distances from a large inertial mass (in this case an irrigation weir shown within the chain-linked area in the distance). The purpose of this will be clear in a minute.
	Here's a closer shot of one of the seismometers placed at 20-feet from the base of the weir and an ISE staff member holding a calibrated modal impact sledgehammer which will supply the input excitation to the ground.
ISE's testing for soil damping (which is a modified aerospace modal analysis protocol we developed) requires two channels of simultaneous data acquisition to obtain an estimate of the soil decay rate as a function of time and distance. The input signal is provided from a 12-pound calibrated modal sledgehammer using a soft rubber (low frequency) tip. A single strike of the hammer against the existing concrete irrigation control weir was sufficient to generate the desired soil stress wave and provide the necessary transmissibility between the hammer and the ground. Output vibration response from the hammer pulse was determined using a the Kinemetrics seismometer array. The monitoring was performed at distances of 10, 20, and 40 feet from the impact point. Data was fed from both inputs directly to the same Tektronix TDS 3014 digital oscilloscope with a capture resolution of 200 milliseconds per division.
	The modal sledgehammer is essentially a hammer with a force-calibrated load cell at the tip. The load cell (the silver part of the hammer shown in the photograph) is capable of measuring impulsive forces up to 5,000 pounds. This is the signal that is fed out the end of the hammer's handle via the BNC cable shown in the photo above. The tip of the hammer (the grey part of the hammer between the load cell and the concrete structure) can be composed of different materials ranging from soft to hard. The harder the material, the higher the frequency content of the impulse. For the testing we're performing (civil seismic vibration), we're using a soft rubber hammer tip with a frequency roll-off of approximately 200 Hz.
So, what were the findings... The ambient ground motion at the test site sans any freight train activity (shown as the blue lines in the figure to the right) was 1.04 micro-inches per second (or 1.04 one-millionths of an inch per second). For comparison, structural damage has been shown to occur at levels of 1.0 inch-per-second for older historic structures and up to 2.0 to 3.0 inches-per-second for newer construction. The pass by of the freight train shown at the beginning of this article (the red line in the figure to the right) produced elevated ground motion of 78 micro-inches-per second at 50 feet from the rail edge - still not really a big deal. As can be seen from the plot, most of the energy of the freight train was between 8 and 20 Hz. The ambient baseline was pretty flat with only some really low frequency motion (around 1.5 Hz) and a little blip at 60 Hz (which was caused by some ground-loop feedback from our generator powering the equipment).
	Here's the same data plotted in a linear magnitude versus frequency fashion (compared to the log-magnitude plot above). Same data, but each plot provides different insight into the problem. Here the dominant frequencies are quite obvious along with the small ambient motion at 1.5 Hz. The freight train is seen to produce most of its energy in a fixed band having two dominant spikes (which, based upon the speed of the train is due to the aforementioned 'click-clack' of the wheels along the rail edges). Care to guess what caused the 1.5 Hz ambient motion - it's the Earth !!!
The damping results were equally interesting. The Logarithmic Decrement Method was used to calculate the relative amount of damping present in a known amount of soil mass by examining the rate of decay in amplitude of each successive cycle produced by the hammer blow. The faster the decay, the larger the amount of damping present in the soil. Here we see the hammer strike (the downward pulse on the red line at 0.58 seconds after the data capture was started) and the resulting oscillatory motion of the surface wave within the soil. Great results, and from them we can determine the characteristics of the soil not only as a function of distance from the hammer, but also as a function of depth. We can actually look into the soil (well the Rayleigh Wave does) and see what it is composed of. The larger and longer the seismometer array, the deeper we can peer into the ground. Our results indicated a damping ratio of 0.01918 per foot. This level is higher than what would be expected for a well-packed soil-only system (typically around 10 percent or 0.010 per foot) and can be attributed to the loosely consolidated soil conditions at the project site. No evidence of rock mass was found at any of the depths that we could 'see' into the soil and the ground under examination behaved pretty much as a homogeneous sandy soil mass.
	And the findings are... Modeling of the worst-case freight excitation (basically the measured level with a healthy margin of safety) on a damped soil oscillator system produced expected impact distanced of between 42 to 49 feet from the rail edge depending on the impact standard utilized (i.e., the ISO vibration standard of 0.007 inches-per-second or the 0.004 inches per second standard established by the FTA). In either case, if you are 50-feet or more from the rail edge, vibration impacts are not expected to be an impact. These results were experimentally verifiable and were checked by ISE on subsequent visits to the site.   ISO = International Organization for Standardization FTA = Federal Transit Administration

Impulsive Noise Assessment Calibration of IS³ and IMP Models

ISE performed an impulsive noise assessment for the purposes of further calibration of our Industrial Source (IS³) model for these types of noise events. Extremely short duration noise events associated with blasting, high pressure pipeline blow down, live ordnance fire, etc., involve the rapid compression and rarefaction of the air producing, in most cases, an instantaneous supersonic flow condition. Oftentimes these sources are mischaracterized by the casual acoustical consultant and the physics of the problem is lost.

The direct goal of the testing during this exploratory field exercise was to fine tune the ISE Impulsive Noise Program (known as the IMP) to characterize the extremely short duration supersonic pulses produced by standard rounds under a controlled condition. The IMP program can currently handle all manner of steady state shock flow, and has been used by ISE for years to predict the acoustic pressure produced by supersonic gas flow in pipes. The trick to this testing was to obtain the acoustic pressure signature of various weapon types under a free expansion and unbounded test condition, and to measure the empirical loss due to terrain and manmade barriers (such as berms). These results can then be passed into the IS³ model expanding its capabilities to provide a visualization of these types of sources as well.

The data acquisition for the testing was performed using a Larson Davis Model 2900 ANSI Type 1 Octave Band/FFT Analyzer. This unit is quite an impressive bit of technology in a small notebook size package. Data was taken in full 1/64 Octave, as well as in 400-line FFT monitoring modes.

This photo shows of shot of Fearless Leader (Dr. Tavares) making the final preparations to the equipment prior to testing. Data for the test not only consisted of the raw octave band and FFT spectra, but also the value of these spectra for 1 ms time increments (very important when determining the effective pressure and subsequent supersonic flow at the moment of ignition of the primer within the cartridge).

Data from the source gunshot emissions logged within the LD 2900 analyzer was later used to refine the IMP model for these types of events.

During the testing, numerous types of weapons were fired - each one being cataloged according to their unique signature.

Here's a shot of an ISE team member shooting a SOCOM II. The largest "blunderbuss" being fired that day.

Remote Field Monitoring Example

ISE routinely performs remote field monitoring for all manner of physical phenomenon. Light and radiation variations, ambient noise monitoring, seismic and manmade vibration monitoring (such as that due to blasting or construction), surface traffic classification, etc. You name it, and we've probably constructed some type of remote gadget to monitor it.

Below is an example of the installation of two remote monitoring systems for short-term (roughly two weeks) ambient acoustical monitoring of a proposed development site.

Wearing the obligatory construction hat (you can get a sunburn on a cloudy day), the hydraulic auger is set into motion. This ground is particularly hard and rocky, and it will take the machine over five minutes to cut through roughly 2.5-feet of soil.

... and yes, that thing to the left of the Bobcat auger is a large cactus which attacked everyone on the team that day...

After the support pole is placed and plumbed, weather-proof enclosure boxes and wires up the instrumentation (the upper box) to the battery (the lower box) are installed.

All of ISE's remote monitoring systems are designed to operate off of 12V and are typically powered by 18 Ah wet cell batteries. Given the typical operational current drain of most systems (between 30 to 50 mA), the unit will operate for periods up to one month without the need for downloads or maintenance.