Lab 4: Conducting a Distance Azimuth Survey

                            Group 6: with Carolyn McLeish and Emily Merkel

Introduction

 While today's GPS technology can be used to accurately determine the location of points on a map for a geographical survey, it is important to also become familiar with less advanced survey methods as well. These more basic survey methods include the distance azimuth survey, which is the method that will be examined in this report. By measuring both the horizontal/vertical distance (meters) and azimuth (degrees) of an object, e.g. a tree or bike-rack, from a stationary location with known coordinates, one can determine where to plot that specific point on a map.

The reason for using more traditional survey methods and tools is that complex electronic devices such as GPS are prone to malfunction for a variety of reasons. Furthermore, if one were surveying an extremely remote location, it might be best practice to use both a GPS method and a distance azimuth survey to ensure  all the necessary data were collected accurately and to negate the possibility of an expensive resurveying of the area due to inaccurate data points.

Some of the more basic tools used to conduct a distance-azimuth survey are as follows: a compass and distance finder/tape-measure, or laser a device that can measure both an object's horizontal distance and azimuth, such as the TruPulse laser device (fig. 1). While the TruPulse laser device was convenient for this survey, it too is electronic which means that it could malfunction in the field like a GPS unit. For this reason, our class received additional instruction on how to properly use both a Suunto compass and a sonic distance finder to take the azimuth and x-y distance, respectively. While the compass and distance finder require more work and are more prone to human error than using the all-in-one measurement capabilities of the TruPulse, it is still important to learn how to use them should the need to do so arise.

Once the distance-azimuth survey was completed, the data were entered into Microsoft Excel, and from there converted using the "bearing distance to line" tool in ArcGIS into feature-class points. Finally, these points were superimposed over a Bing imagery map to asses the accuracy of the survey.

Figure 1

Figure 1 shows the TruPulse laser device (yellow) that was used to measure both an object's slope distance and azimuth. For this lab, the laser device was mounted to a tripod to ensure that it would remain in the same location. Using the  single point of origin at the tripod ensured that only one decimal degree location would need to be found. From this decimal degree origin, all other data points measured by the TruPulse throughout the survey could be calculated later on with computer programs when they were entered into ArcGIS.

Methods

Area of Interest

 Given the criteria listed above, the origin of this distance-azimuth survey was located next to the Nursing Education building on the UW-Eau Claire (UWEC) campus. Figure 2 shows aerial image of the AOI; U.S. highway 12/Clairemont Ave. runs along the lower portion the smaller image, while the red X in the larger image approximates the location in which the TruPulse laser device was set up on a tripod (fig. 3). It was approximated from the Bing aerial imagery that the origin was located at 44.797244^o N and 91.502394^o W.

The size of the AOI was decided to be 1 hectare, which was based on the size parameters in the lab instructions. However, although group 6's AOI fit the 1 hectare maximum, no 100 m x 100 m square was actually measured out. Instead, it was decided to simply limit the range of the laser device to less than/equal to 100 meters in any direction. It is now apparent that this was a mistake and the range of the measuring device should have been limited to objects 50 meters in any direction from the point of origin to ensure a 1 hectare AOI. Doing this would have placed the origin in the center of the square. However, even though AOI did not turn out to be a perfect square, the area of the AOI should fall into the overall parameters set by the lab instructions as to the extent of the AOI. This is because the Nursing Education building was behind the origin thus preventing the creation of a 2 hectare AOI.

Figure 2

Figure 2 shows the origin point that was chosen for the distance-azimuth survey. Marked by the red x, the point of origin was approximated at: 44.797244^o N and 91.502394^o W. The locations of the points later surveyed in the AOI were calculated using these coordinates in an ArcMap program.

Data Collection

 It was important to select a point of origin for the survey that was both visible from aerial imagery and one that was relatively permanent. This is because decimal-degree coordinates from Bing Map (in ArcGIS) were needed for the point of origin, which was the location where the tripod was set up with the laser device for the survey. Also, if any of the points needed to be resurveyed for any reason, the permanence of the origin point is essential. For example, if five points needed to be recollected, then the survey tool should be set up in the same location during the resurvey as it was in the initial survey; a failure to do this would mean that the points in the resurvey would be substantially different than those in the initial one.

It was decided that the TruPulse laser device would be used to collect data points for the survey because this tool allowed for both the slope distance and azimuth of an object to be collected with just a push of a button to switch back and forth between the two. The general procedure for collecting the distance/azimuth data is as follows:

•  The laser device was pointed at whatever object was to be measured. In the case of this lab, any stationary object was chosen, such as a lamp post, fire-hydrant, tree, bench, garbage can, etc. because the main purpose of this lab was to be introduced to the methods and tools associated with a distance-azimuth survey.
• Once the desired object was in the cross-hairs of the laser device's viewfinder, the operator pressed the fire button located on the top of the TruPulse (fig. 3). The laser device would then display in the upper portion of the viewfinder either the azimuth (degrees) or the slope distance (meters) for that object; the operator would toggle between these two parameters, depending on whether the distance or azimuth was desired (fig. 3).
• Once either the distance or the azimuth was measured, the person operating the laser device would announce the resulting measurements to another team member who was recording the information. In addition to the azimuth and slope distance, the name of the object being measured was recorded as well.

As mentioned in the AOI method section above, the point of origin for the survey was determined to be at the northeast corner of the Nursing Education building on the UWEC campus. A tripod was used to ensure that the TruPulse laser device (fig. 1) remained in the same location throughout the survey. The use of a tripod was also useful because it enabled each member of group 6 to collect the distance and azimuth data throughout the survey without disturbing the point of origin.

Figure 3


Figure 3 shows the controls on the TruPulse laser device such as the fire button (under Emily's index finger) used to shoot the laser at an object to measure its distance/azimuth. Also, the toggle switches are use to switch back and forth between the distance and azimuth in order to obtain measurements for each via the same device. The thick yellow arrow shows the direction that the laser path takes to the whatever object is in the TruPulse's viewfinder.

Magnetic Declination

Figure 4 shows a general magnetic declination (MD) map of the world, from NOAA. MD is a necessary correction in some regions of the world as the as magnetic north, as measured by a compass, can vary substantially from the geographically determined true north. While the declination map in figure 4 provides a good visual for understanding the concept of MD, it should not be used to determine MD. This is because, in addition to regional variance, MD will change temporally as well; for example, NOAA estimates that the MD for Eau Claire, specifically our point of origin, will change by a rate of 0.1^o  W per year.

From the national oceanic and atmospheric agency (NOAA: http://www.ngdc.noaa.gov/geomag-web/#declination), MD for this survey's point of origin in Eau Claire, WI was estimated at  1.07^o  W and is changing at a rate of 0.1  W per year. The MD for the specific origin point for our survey was done by entering the latitudinal and longitudinal decimal degrees for the point of origin as determined using Bing Map imagery. As mentioned in the AOI section of this lab, the latitude and longitude for the point of origin were 44.797244^o N and 91.502394^o W, respectively. The latitude and longitude were correlated to the point of origin for this survey because it was discovered that the  general MD for Eau Claire, based off the zip code 54701, were different than those at the origin point. For example, at post office the MD was 1.08^o W and while still changing at a rate of approximately 0.1^o  W per year.

While MD for the AOI was low enough to negate its importance in terms of modifying azimuth measurements, this paper will compare map layers generated with the raw azimuth data from the laser device to those created with azimuth data corrected to the MD determined for the region. In order to do this, the 1.07^o  W was added to each raw azimuth value using an excel formula.

Figure 4

Figure 4 shows a modified map that illustrates magnetic declination (MD) from 2010. MD is a phenomenon which occurs because magnetic north, as measured by a compass, is often different than the geographically determined true north. The thick black contour line that bisects the U.S. in this picture represents no variance between magnetic and true north. Also, red and blue contour lines represent areas where magnetic declination is west (negative) or blue (positive). In addition to being a regional phenomenon, MD also varies temporally as well; thus, this 2010 MD map is of little value today in 2014, other than for instructive purposes, as the MD has surely changed since then.

Excel Spreadsheet

Data from the distance-azimuth survey were recorded into Microsoft excel. Information that was included in this spreadsheet is as follows: point number, distance, azimuth, and point data; point data just the name of the point measured. Also, as illustrated in figure 5a and b, both an x and y field were created in the excel spreadsheets as well. These x (-91.502394 ) and y (44.797244) fields respectively correspond to the longitude and latitude values determined for the point of origin from Bing maps, as discussed in detail in the AOI method section.

For the purposes of this lab, two excel spreadsheets were created, as shown in figures 5a and b. In one excel spreadsheet, the raw azimuth data were entered, i.e. the azimuth measurements as they were displayed in the TruPulse ( fig. 5a). A second excel spreadsheet was produced that corrected the azimuth measurements as they were reported by the laser device to better reflect the magnetic declination of the region (fig. 5b), as mentioned in the previous methods section that discussed magnetic declination in greater detail.

Figure 5
a
 b
Figure 5a shows the excel spreadsheet that was created using raw azimuth data from the TruPulse laser survey device. Figure 5b shows the excel spreadsheet that was created to adjust for MD. Note, besides "azimuth," all other fields are identical from the spreadsheet represented by 5a to the spreadsheet represented by 5b.

Map Creation

Two functions needed to be performed on the tables created from the spreadsheets above.First, the Bearing Distance to Line (BDL) tool was used to convert the data in the tables into a feature class of vertices. Next, a second ArcTool, Feature vertices to points (FVP) was used to create points from the vertices. Following are the steps that need to be followed to create each feature class. The specific tools themselves  can be found by using the "search"  interface in ArcMap, or by navigating through the ArcMap toolbox, to data management, to features, and finally to either to the BDL tool or the  FVP tool.

In order to create a map of the data points collected in the survey, it was first necessary to orient them on an x-y axis. To complete this transformation, the BDL tool in ArcMap was used to convert the field data collected during the distance-azimuth survey into a feature class that can be mapped using ArcGIS. Figure 6a give a general idea of how the parameters should appear. For instance, once inside the ArcMap program, the fields in the BLD interface should contain the following information:

•  Input table should have contain the table generated from the excel spreadsheet.
•  Output feature class  should contain the name of the destination file for the resulting feature class.
• X field should contain the corresponding x-field from the table.
• Y field should contain the corresponding y-field from the table.
• Distance field should contain the corresponding distance-field from the table, in this case in meters.
• Bearing field should contain the corresponding azimuth-field from the table, in degrees.
• Click okay to generate the vertices feature class, which should appear like the image in figure 6b.

Figure 6

a 
b
 Figure 6a shows the parameters used by the BLD tool used to generate the feature class below it in 6b.

Once the vertices feature class is created (fig. 6b), it should be converted to a point feature class. To do this, the FVP tool in ArcMap should be utilized. Following is an example of the FVP interface (fig. 7a), as well as the parameters that should be used to create a point feature class in ArcMap:

• Input features should contain the input feature class.
• Output feature class should contain the geodatbase where the output feature class will be located once it is created.
• Click okay on the FVP interface to generate the new point feature class (fig. 7b). 

Figure 7
a
b
Figure 7a shows the input parameters that could be used in the FVP interface to generate a point feature class like the one shown in 7b. Note that the green diamonds are the points associated with the point-feature class and the beige lines are associated with the vertices feature class. The vertices-feature class was left up in ArcMap when the screenshot in 7b was generated to show the relationship between the vertices-feature class and the newly generated point-feature class.

Once the vertices were converted to points using the FVP tool in Arc toolbox, the resulting point-feature class can be superimposed over the Bing imagery base map in ArcMap. Figure 8 shows a map that was created using the point-feature class generated above. The resulting feature class, which represents raw data that were collected from the field during the distance-azimuth survey, that is data that were not corrected for magnetic declination, was then superimposed over Bing Map imagery of the UWEC campus. Similarly, figure 9 shows the map that was generated by superimposing both the raw azimuth point-feature class data (red) and the point-feature class that was generated using azimuths adjusted to the MD of the point of origin; the yellow dot in both figures 8 and 9. Also, the same sized dots were used for both raw azimuth data points (red) and those adjusted to the MD associated with the point of origin (green) for the survey (fig. 9); this was done to ensure that a larger dot representative of a particular feature class would not simply appear to be offset by a smaller one, or vise-versa.

Figure 8

Figure 8 shows point-feature class (red dots) displaying raw azimuth data superimposed over a Bing Map base map in ArcMap. This was done in order to better asses the completed the accuracy of the survey. The point of origin for the distance-azimuth survey is represented by the yellow dot.

Figure 9

Figure 9 shows point-feature classes generated by using both raw azimuth data (red dots) and azimuth data that were adjusted to better reflect the MD at the point of origin (green dots) for the distance-azimuth survey (yellow dot).

Discussion

As expected, the 1.07^o  W MD associated with the MD in the AOI had no profound effect on the PFCs data in terms of their positions relative to one another. This fact supported in figure 9 where both the PFC for the raw azimuth data (red) is displayed in conjunction with the PFC associated with azimuth data adjusted to reflect MD (green dots).

When the point-feature classes generated in ArcMap were superimposed over a Bing imagery base-map it was apparent that some errors were present in some locations. For instance, the red square in figure 10, which displays on raw azimuth data, shows a data point that appears to be inside the nursing education building. According to the attribute table this point represents a tree and is obviously an inaccurate data point.

The next source of error is shown again in figure 10, but this time it is represented by the yellow circle in the parking lot. While a few data points were taken in the parking lot, none of them extended out as far as what is shown in the yellow circle in figure 10. While one could argue that the parking lot was altered during the construction of the new Davies center in 2012 and that this construction was the result of the cluster of points shown in figure 10, it is highly unlikely due to the fact that most of the points were trees.

Some of the data that were generated in the PFC appeared to be correct. This data includes the bike-racks outside directly in front of the Nursing Education building and are highlighted with the green rectangle in figure 10, just northwest of the point of origin.

As far as the accuracy of the remaining data points in figure 10, and in figure 8 and 9 for that matter, it would be interesting to see them superimposed upon imagery that included the newly constructed Davies center. Furthermore, some of the inaccuracies of the data points may be accounted for due to the fact that a major snow-/sleet-storm on had begun Thursday, February 20th when the survey was conducted.

The storm started about mid-way through the survey and it was at times difficult to obtain accurate readings with one shot of the TruPulse laser device. For instance, objects were measured at 5 meters away although they were clearly more than that. In cases like this, the obviously erroneous measurement was not recorded, and instead the laser was shot until a more believable measurement was displayed. Due to the way the TruPulse was behaving during the storm, this would have been a prime time to employ the use of the compass to obtain, or at least check, the accuracy of the azimuth as measured by the laser device; however, our group did not think to do this at the time.

There were other issues with the survey as well, all attributable to human error and a lack of proper planning. This included the determination of the AOI, s discussed in the methods section of this report. While a polygon measurement using tools in ArcMap revealed the total measured area of the AOI to be about 1.1029 hectares, more thoughtful planning could have been done to ensure that the 1 hectare maximum was not exceeded and that the AOI was represented by a perfect square, or more nearly so than in this particular instance.

Figure 10

Figure 10 shows some sources of error associated with the AOI in the distance-azimuth survey. For instance, the red square on the northwest corner of the nursing education building represents a data point the ArcMap attribute table associates with a tree. Also, the data points highlighted by the yellow circle should not be in scattered throughout the parking lot as shown here. In contrast, the bike-racks highlighted by the green rectangle just north and west the point of origin appear to be fairly accurate.

Conclusions

The distance-azimuth survey lab was interesting because it allowed one to gain a better grasp of the basic "map and compass" survey methods. These basic survey methods are important to know, as electronic technology can fail at any time. Our group experienced this failure first-hand when it appeared that snow may have disturbed the TruPulse laser devices collection of distance, and likely the azimuth data.

Mapping the data that were collected during the survey was an important skill to learn as well because it allowed for a more thorough analysis of the data points in ArcMap. For instance, if one were not able to map the points they would not be able to see how much or even if their survey data were inaccurate.

If the lab were to be done over again, it may be a good idea to at least calibrate a few data points using the compass and tape-measure/distance finder to ensure that the laser device was functioning properly.

For future labs, it would also be nice to have more time to complete the survey and analyze those results in ArcMap. This extra time could then be used to resurvey the erroneous data points in AIO in an attempt to correct mistakes made in the first one, if any were present.

 

Mission Planning: Unmanned Aerial System

                            With Carolyn McLeish and Emily Merkel

Scenario 1:

                A power line company spends lots of money on a helicopter company monitoring and fixing problems on their line. One of the biggest costs is the helicopter having to fly up to these things just to see if there is a problem with the tower. Another issue is the cost of just figuring how to get to the things from the closest airport.

Problems:

      The first question for this company would be "How much is 'lots of money?'" While it was difficult to determine the cost to utilize a helicopter from websites of companies that provided such services, those used for the purpose of medical evacuation cost about $6500 per transport in 2010 (Wykes and Sanford, 2013).

       Assuming that a medical transport would last approximately one- to three-hours, one could estimate a cost of about $2170-$6500/per hour of specialized helicopter services. Even so, using the lower end of this estimate, i.e. $2100, a power company would have to spend about $19500 to use the helicopter services, assuming a 9 hour workday.

       The other major issue with using a helicopter is that finding a nearby airport may be difficult in cases where the power lines are located in remote areas. Flying or otherwise transporting  helicopters (e.g. via truck) to such remote areas would only add to the cost of fuel and per-hour cost of the use of the helicopter.

       Another problem with using full-size helicopters to monitor power lines is that flights would be weather-dependent. For instance, if the power lines are located in a region plagued with inclement weather, how likely is it that a cancelled flight would be able to resume ASAP once the weather improved? Probably not too likely considering that the helicopter company would probably have other appointments scheduled with other clients.

Solution:

Rotary Wing Camera Platform

        The most effective solution to the problems presented by full-size helicopter inspection of power lines mentioned above would be to employ an unmanned aerial vehicle to inspect the power lines for damages. However, a fixed-wing UAS platform (FWP) would not be recommended in the case of power line inspections due to: 1) the vehicle's inability to hover and take the pictures/video necessary to asses damage, if any and 2) the danger that power lines pose to the FWP should it become entangled in them. The risk of entanglement in power lines also rules out other, even cheaper, UAS platforms such as kites and balloons for the inspection of power lines and towers.

         The most practical solution to the problems inherited by inspecting power lines and towers would be to use rotary wing platforms (RWPs).  Following are two examples of RWP systems on opposite ends of the price spectrum.  

Cheap
          The cheapest resolution that would allow the utility company to effectively monitor its lines and towers would be to deploy a relatively cheap RC RWP unit to the areas where the towers are located. For instance, the Align RC 600 Nitro (fig. 1), which comes as a kit and costs approximately $700, could be retrofitted with a durable camera on its underside that would allow for the video inspection of power lines and towers .

         The waterproof Ion-Air Pro 2 helmet camera (fig. 2), for instance, weighs only 4.6 ounces, is small in dimension (1.4  x 1.4 x 4.5 in.), and has 2.5 hours of battery life. Costing roughly $250 apiece, several of these cameras could be bought and attached to the Align throughout the workday as the battery fails in each.

          Although the Align comes as a kit, it would likely be no problem for one of the power company's maintenance workers to assemble it on site. Replacement parts for the Align, such as rotary shafts, blades, and fuselages are also available on the NitroPlanes web page (http://www.nitroplanes.com/15h-kx0160npc.html).

          Furthermore, the relatively cheap cost of the Align RWP would enable more than one copter to be purchased, thus cutting down substantially on the time it takes to inspect the towers and lines. For instance, ArcGIS could be used to establish inspection zones and use a feature class layer to represent the towers. Each Align operator could carry a GPS unit that was programed with the coordinates of each tower and geographically "check off" each tower that was inspected in their respective zone. Towers could also get identifying placards installed on them so that their unique identifier could be synchronized to specific coordinates in ArcGIS and the GPS device.

         Mobility is another pleasing aspect to the RWP solution. For example, operators could take the small  (approx. 7.1 pound) Align model with them in their company/all terrain vehicles (ATVs) to the locations where the inspections would take place. Once there, the Align could be deployed and the applicable data collected.

         Weather would not affect Align missions as much as those conducted by companies with full-sized copters because missions could simply be postponed until weather permitted their re-initiation. Also, since all the Align operators would be in-house (i.e. linemen trained to operate the RWP) rescheduling missions would not be as daunting as compared to doing so for independent helicopter companies. Furthermore, the low cost of the Align would ensure that if one of the RWPs did happen to become lost or damaged, a replacement, although not ideal, would be doable in terms of cost.

         One downside to this particular RWP model (i.e. the Align 600 Nitro)  is that its 440 cc fuel tank only allows for 10 minutes of flight time, assuming no payload and ideal conditions. However, the problem of limited flight time could be solved by simply replenishing the fuel supply periodically throughout the workday. Also, the Nitromethane fuel that this RWP uses is relatively cheap costing about $25 per gallon, according to some internet sources (http://www.ultimaterc.com/forums/showthread.php?t=176431) and would allow for 84 minutes of continuous flight time, assuming about 3700cc per gallon.

Figure 1

       The Align RC 600 Nitro is a nitromethane powered, remotely controlled helicopter. With a camera attachment, such as the Ion Air 2 in figure 2 below, this device would be an ideal platform from which to monitor power lines and towers more cost effectively than current full-sized helicopter services allow (http://www.nitroplanes.com/15h-kx0160npc.html).

Figure 2
 
      The cheap, sturdy, waterproof Ion Air 2 helmet camera could be retrofitted to the underside of the Align RC helicopter (or similar RWP system) in order to visually inspect power lines and towers for damage. Multiple Ion Airs could be purchased in order to compensate for the devices 2.5 battery life.   (http://www.bestbuy.com/site/ion-air-pro-2-wi-fi-hd-camcorder-blue-black/2174008.p?id=1219070712053&skuId=2174008&ref=06&loc=01&ci_src=14110944&ci_sku=2174008&extensionType={adtype}:{network}&s_kwcid=PTC!pla!{keyword}!{matchtype}!{adwords_producttargetid}!{network}!{ifmobile:M}!{creative}&kpid=2174008&k_clickid=02b4ced1-feed-2e49-2a09-00003036eaf6#tab=overview).


     

 Expensive:

         When fitted with a fuel engine, the Avenger by Leptron (fig. 3) gets about 2 hours of flight time. The biggest draw-back for the Avenger is its price tag which equates to about  $100,000 apiece (Joyce). The reason for the high cost of the Avenger compared to the Align 600 is because, in addition to increased flight time; durability; and performance (i.e. its ability to operate in 40 m.p,h winds), the avenger is much more versatile in terms of operability. For instance, the Avenger can be manually controlled by an operator through either a laptop Windows interface, or via a controller.

        Also, a more sophisticated RWP such as the Avenger can also be flown by using GPS way-points to guide its flight path (autopilot). This option would be useful as the RWP could be flown to previously geocoded towers before the operator switches over to RC mode in order to perform a more precise inspection of the tower. Once each geocoded tower was inspected, it could be "checked off" the list if the inspection was a part of routine, preventative maintenance (PM).

        Also, the ability of the Avenger to switch between RC and auto pilot mode is good since remotely located power lines might be miles from the road. In this case, the 11-pound Avenger could be transported via ATV  or company vehicle to the area of interest and operated by remote control in order to inspect power lines and towers.

           Another attractive aspect of the Avenger is that Leptron sells specialty cameras that can be fitted onto the Avenger. These turret-mounted cameras (fig. 4)  have geo-locator capabilities, are stabilized, and can be operated from the Avenger's remote control as opposed to commercially available cameras that could be mounted to the avenger in order to cut costs.

         One problem that the power company may have with the Avenger is that its price may limit the utility company to only one unit, and thus less area covered over a given time as compared to multiple cheaper units being operated simultaneously, as given in the Align example. 

 .

Figure 3

  
        Image of the Avengenr by Leptron in flight.  Although far more expensive than the Align RWP, the durable Avenger integrates all its geospatial technology, such as geocoding, geo-locating,  and GPS way-points, into one unit so that data relating to power line and tower inspection can be easily classified (http://www.leptron.com/corporate/products/avenger/specs.php).
 

 

 

 Figure 4

 

Some of examples of the the more sophisticated, turret-mounted, remotely operated cameras that can be used fitted onto the Avenger RWP system (https://www.leptron.com/corporate/products/avenger/camera.php).

Conclusions:

       While the Align and Avenger RWP options above both solve the fiscal problems associated with of utility line inspection via full-sized helicopters, each does so in a different way. For instance, while the Align option is much cheaper than the Avenger option, the Align would be much more cumbersome in terms of operation, mobility, flight time, convenience and data accuracy. That being said, all the problems associated with the Align option could be solved, but it would require unconventional synchronization of many different systems such as cameras, GIS, GPS, and flight operation; whereas with the Avenger option, all these systems would come already integrated with one another.

        However, with the convenience of the integrated flight, GPS, and GIS systems, as well as other luxuries such as improved quality and performance in addition to high-tech camera systems, the Avenger by Leptron comes at a price. While the price of the Avenger may limit the utility company's ability to purchase more than one unit, the overall price of the system would still save the company money in the long run with the unit paying for itself after five or so uses (assuming $19500/nine-hour day for a conventional helicopter service).

 

Sources:

Sanford, J., and Wykes, S, 2013, Study examines cost-effectiveness of helicopter transport of trauma

       victims: http://med.stanford.edu/ism/2013/april/helicopter.html (accessed February 2014).

 

 

                                                                                                                                              

Scenario 2:

        An oil pipeline running through the Niger River delta is showing some signs of leaking. This is impacting both agriculture and loss of revenue to the company.

Problem:

           According to the scenario above, the main problem is that the oil company does not know where the source of the leak is located on the Niger River Delta (fig. 1). Following is one method in which the leak could be determined in a cost effective and efficient manner from an unmanned aerial platform in order to prevent further damage to the delta environment as well as to the oil company's revenue. 

     
       The proposed system will not only make locating the leak easy, but will also allow for the data obtained from from the proposed monitoring system to be easily synchronized with geospatial systems. For instance, taking advantage of such geospatial programs such as GPS and ArcGIS in order to locate the leaky pipeline.

        However, it should be noted that the following idea involving the use of tethered balloons to locate the source of the leaking oil on the Niger River are based solely upon the small amount of information provided by the oil company thus far. It may be determined that other, more effective unmanned aerial systems may be better suited to locate the oil leak after the following important questions are answered by the oil company:

1) How was the leak discovered?

2) What, if any, is the estimated cost of the leak in terms of its impact on the delta region and in terms of revenue lost to the oil company?

3) What is the estimated area of interest (AOI) of the leak in both terms of size and geographic location?

4) What measures, if any, have already been undertaken to locate and stop the leak by the oil company?

5) What, if any, has been the involvement of Nigerian government regarding the matter of the contamination of the Niger River delta?

6) Has the oil company consulted with other authorities, such as environmental consulting firms, on the matter of contamination due to oil leaking into the Niger River delta?

       

Figure 1

       This Google image shows the general area of interest in the Niger River delta on the west coast of the African continent. More information from the company whose oil pipeline is leaking will be needed in order to pinpoint the exact AOI in this region.

 Solutions:

Locating the Leak

       The first problem is that the location of the leaking pipe is unknown. In this case, an aerial surveillance system consisting of near infrared (NIR) cameras suspended from tethered balloons will be placed at various, predetermined locations on banks of the Niger River in order to take aerial photographs of the river's surface water. NIR cameras attached to each balloon platform would be periodically retrieved so that the spectral data collected by them could be computationally analyzed. From this spectral data it could then be determined whether or not a specific section of the Niger River corresponding to a particular balloon was contaminated with oil. Once a non-contaminated portion of the river was found, ground crews could then search between the balloon that exhibited no signs of an oil leak and the nearest balloon that did downstream of it; this is the area where the leak should be.

       In order to illustrate this procedure more clearly, figure 2 shows a series of balloons along the banks of a model river; numbers on the right-hand side of the image correspond to arbitrarily determined river-miles. Blue and grey shading corresponds to water that is uncontaminated and contaminated by oil, respectively, while arrows indicate the direction of water flow. So, for example, if the sensor on the balloon at river-mile 9 detects no contamination, but the balloon at river mile 7 does, then it could be reasoned that the leak in the pipeline is between river-miles 7 and 9 and ground crews could be dispatched to this area in an attempt to locate the leak.

Figure 2

       Figure 2 illustrates how the source of oil contamination could be determined using a system of tethered balloons to monitor contamination in the AOI. Balloons in this diagram correspond to odd-numbered river miles. Each balloon will take a series of aerial photographs in NIR to locate surficial oil contamination on the river (grey areas). Once an area the river is found to be free of contamination (blue) using aerial surveillance, ground crews  need only to search between that balloon and the nearest one exhibiting contamination downstream of it to find the source of the leak; in this example, between river miles 7 and 9.

Sensors

         In order to determine whether or not the water in the Niger River is contaminated,  the correct sensors must be attached to the tethered balloons. Figure 3 shows some of the spectra associated with oil slicks on water, as determined by the USGS during the 2010 Deepwater Horizon (DWH) oil spill in the Gulf of Mexico.
    
         While the DWH spill was likely more massive than the one being examined in this article, the USGS found that  when viewed in infrared wavelengths, different thicknesses of oil slicks displayed differnt spectral signatures. Computational analysis could then be performed on the images collected by the sensors to determine whether or not the portion of the river corresponding to that particular sensor was contaminated or not.
    
          Using the spectral information provided by the USGS, NIR cameras would likely be the best photographic method for determining whether or not the surface waters on the Niger River are contaminated with oil or not.

         Figure 4 shows an example of  a near-infrared camera, from Edmund Optics, that could be suspended from a balloon platform in order to locate surficial oil contamination on the Niger River. While far from cheap at nearly $2000 apiece, this price likely pales in comparison to what the oil company is losing in revenue and mounting cleanup costs.

 

 Figure 3

 
       Figure 3 shows an example of the spectra measured by the USGS during the Deepwater Horizon oil spill in 2010. It was found that when using NIR sensors, thin layers of oil could be spotted on the surface of the water; i.e. those less than 0.5 mm thick (blue line).

Figure 4

       Figure 4 shows one of the cheaper NIR cameras offered by Edmund Optics. This device, which weighs about 90 g and costs about $2000, could be suspended from the tethered balloon platforms in order to detect thin layers of surficial oil contamination on the Niger River delta (http://www.edmundoptics.com/imaging/cameras/near-ir-nir-ultraviolet-uv-cameras/1460-1600nm-near-infrared-camera/2384). 

 

Monitoring Platforms

           According to precipitation graphs for Lagos, Nigeria, which is approximately 200-300 miles away from the AOI on the Atlantic coast, weather should not inhibit the deployment of balloons except, maybe, in the months of May, June, and July, when rainfall exceeds 200 mm per month (fig. 5). However, if inclement weather were to occur on a day when the balloons were scheduled to collect data, their deployment could be easily rescheduled until a more meteorologically favorable day.

         The cost of the balloons themselves is very minimal when compared to overall cost of the spill in terms of ecological damage and revenue lost. Offered by Balloons Direct, figure 6 shows an example of a weather balloon that could be used in this project. Each balloon costs about $35 and has a payload capacity of 3 pounds, which is more than enough to lift the 90 gram infrared sensor mentioned above in figure 4.

         In order to deter theft of the expensive NIR cameras, it would be beneficial to outfit each camera with a harness system that was easy to detach from its balloon monitoring platform. This detachable harness system would also be beneficial as the NIR cameras would need to be removed periodically anyway in order to download their images onto a computer for spectral analysis.

        Balloons could be tethered to the ground using a rope or cable attached to either a hand operated or motorized winch. However, the balloons would likely not be very high off the ground (<20ft.) and a more expensive, motorized winch system would be more of a luxury than a necessity.

Figure 5

      Figure 5 shows the average precipitation for each month in Lagos, Nigeria, located approximately 200-300 miles up the Atlantic coast from the Niger River delta. Based on rainfall averages projected here, the only problematic months for a balloon launch somewhere in the Niger River delta would be May, June, and July of any given year; that is, when the precipitation is greater than 200 mm per each  month (http://www.eldoradocountyweather.com/climate/africa/nigeria/Lagos.html).
 

 

Figure 6

       This figure shows the cost-effective ($35) "Cloud Buster" weather balloon offered by balloons direct. Its 3-pound payload capacity would be more than adequate to lift the 90 g NIR camera shown in figure 4 (http://www.balloonsdirect.com/products/55-foot-cloudbuster-weather-balloon-orange).

 

 

Conclusions:

         While locating the oil leak on the Niger River Delta is no easy task, regardless of what method is used to determine its source, the use of tethered balloons outfitted with NIR-sensors would provide an efficient and cost effective manner of doing so given the information that was made available by the oil company thus far.

          Once the questions presented in the beginning of the article are answered, other, more effective measures may be recommended based on that information. For instance, if the size of the leak is large enough, the Nigerian government or an environmental consulting firm may be able to offer further assistance to the oil company in addition to our services.

 Works Cited:

Clark, R.N., Swayze, G.A., et. al, 2010 , A method for qualitative mapping of thick oil spills using

      imagingspectroscopy: http://pubs.usgs.gov/of/2010/1101/ (accessed February 2014).

El Dorado Weather, 2014, Lagos, nigeria, africa average annual temperatures:

      http://www.eldoradocountyweather.com/climate/africa/nigeria/Lagos.html (accessed February 2014).

                                                                                                                                 

Scenario 3:

A military testing range is having problems engaging in conducting its training exercises due to the presence of desert tortoises. They currently spend millions of dollars doing ground based surveys to find their burrows.

Questions / Concerns:

·         What is the scope of the area to be surveyed?

·         Would it be preferable to only find the burrows or to find the suitable training ground?

Analysis

     The Desert Tortoise is an endangered species that lives in the Mojave and Sonoran desert of southern California, Nevada, and Utah. They prefer semi-arid grasslands, desert washes, and sandy canyon bottoms that are below 3,500ft elevation. They live in burrows that are 3-6ft deep. They are most active in the Spring and least active from November through February, when they hibernate in burrows. Desert Tortoises depend upon vegetation such as new cacti growth for food and water; they also consume calcium-rich soil for digestion, and prefer to burrow in sandy loam soils (ardisols) with varying amounts of gravel or clay. When rain is anticipated, the tortoise will dig basins to collect the rainwater. Tortoises also prefer south facing slopes. A recent study performed by the Department of Defense, states that tortoises prefer to build burrows under a vegetation canopy near to a desert wash. (Grandmaison 2010). All of these factors can be used to aid in locating the tortoise habitats.

 

 Figure 1



Figure 1http://www.spatialsource.com.au/wp-content/uploads/2013/09/Aibot-3-280x203.jpg

  The use of unmanned aerial systems (UAS) can aid surveyors in determining where Desert Tortoise burrows are located. There are several options available for UAS, including a fixed wing UAS, or rotary wing UAS. A fixed wing UAS is more suitable for covering large areas, and can travel in a preplanned grid flight path. A rotary UAS is more versatile and can be used for small, but hard to reach areas. A gas powered fixed wing UAS can have up to 10 hours of flight time, allowing your organization to cover large areas in one survey. A multi-spectral camera can be attached to this UAS to survey the area and determine the soil type, vegetation and moisture of the ground below. Since Desert Tortoises dig their burrows or basins the freshly dug soil may have a different spectral signature than the ground; a simple remote sensing analysis of the collected image would be required. The same multi-spectral sensor can be used to create a false color image that will aid in visualizing areas of high vegetation and moisture content, which tortoises prefer.

Figure 2


Figure 2 http://img.directindustry.com/images_di/photo-mg/fixed-wing-civilian-mini-uavs-101645-2972005.jpg

         Other sensors could be used to create a point cloud which would be used to create a digital elevation model through photogrammetry. This model would be used to determine elevation and slope. Combining the vegetation, elevation, slope and soil type information, a habitat map could be created which highlights key areas that Desert Tortoises prefer indicating areas also that would be better suited for training exercises. This survey could be completed in early spring or during the months of November through February.

         Using a simple camera at low altitude and analyzing the photography would be a low cost option to detecting the burrows, other options such as using a multispectral camera and perhaps

Sources:

 

Grandmaison, David D., 2010. Landscape-Level Habitat Associations and. (n.d.). Department of Defense Legacy Program. Retrieved February 16, 2014, from http://www.denix.osd.mil/nr/upload/08-385-Technical-Report-Landscape-Level-Habitat-Associations-and-Phylogenetics-of-Desert-Tortoises.pdf

UAV: fixed wing or rotary?. (n.d.). sUAS News. Retrieved February 16, 2014, from http://www.suasnews.com/2013/09/25214/uav-fixed-wing-or-rotary/creating a habitat map would be more expensive.

                                                                                                                                                                    

Scenario 4:

          A pineapple plantation has about 8000 acres, and they want you to give them an idea of where they have vegetation that is not healthy, as well as help them out with when might be a good time to harvest.

      Firstly, the area is fairly large so if the methods we come up with prove to be too time consuming or expensive, a sampling method could always be employed in order to get a general idea of the vegetation on the land.

       Also, we thought that maybe the type of data that we would be collecting could maybe be used over several seasons, so may prove to be worth the money and time. We thought that once it had been highlighted which areas had the proper quality vegetation, and which area was best to harvest then, these trends might apply across a few growing seasons.

       In order to highlight the areas where the vegetation is less health than others, we felt that a thermal infrared sensor could be used. Thermal infrared sensors detect electromagnetic waves of wavelength 3.5 to 20 micrometers, as this is the wavelength of moisture particles. As water is one of the input elements of the process of photosynthesis, we thought that areas where we could detect higher moisture levels would be areas where the vegetation would be healthier. Figure 1 below shows an example of this type of sensor being put in to practice in relation to vegetation health. You can see how it clearly shows the different areas of soil health, and how you could determine which areas were less healthy.

Figure 1

     Figure 1: An example of a thermal infrared sensor image being used to measure the health of  vegetation in Colorado. (© Federation of American Scientists)

       However it must be taken in to consideration that these sensors can be expensive, and they must be kept very cold when used, as the radiation that is being sensed is so weak. In order to acquire knowledge on areas where the plants are ready to harvest, we thought perhaps a digital cameracould be used in order to see the colours of the landscape and wee where the crop is ripe enough to pick. This would be a fairly cheaper part of data collection as cameras are cheaper than remote sensors, but it would have to be a high resolution camera, in order to detect the image from a significant height above the fields.

        These sensors would then have to be attached to an unmanned aerial system, in order to acquire the data from above and keep costs to a minimum. We would recommend using an Aerosonde, as theses are commonly used for collecting weather data. It is gasoline powered which we felt was necessary as we are covering such a large area, and many of the battery operated ones do not last very long. It has the ability to hold sensor equipment as many of them come with different sensors, and it can last about 38 hours in the field in one flight.

        We propose that a flight course would be pre-made to fly the vehicle back and forth across lengths of the fields taking images along the way until the whole area had been covered. You may want to carry this fieldwork out just before the 18-20 month growing period is over. Also, seeing as many of the regions where pineapples are cultivated have cold very hot days but cold nights, it might be best to fly the unmanned aerial vehicles at night, so as to keep the sensors clod, but there would still be moisture content in the air. The data on harvesting times would have to be done during the day, so that the hues of the vegetation could be seen, and the camera does not require a special temperature.

 

Sources:

 

AAI. 2014. Unmanned Systems. AAI Corporation. Available at: https://www.aaicorp.com/products/unmanned-systems <Accessed on: Friday 14th February 2014>

Dole. Unknown. Pineapple Cultivation. Dole-Plantation. Available at: http://www.dole-plantation.com/Pineapple-Cultivation <Accessed on: Friday 14th February 2014>

Unknown. Unknown. Introduction to Thermal Infrared Remote Sensing.University of California - Santa Barbra, Department of Geography. Available at: http://www.geog.ucsb.edu/~jeff/115a/remote_sensing/thermal/thermalirinfo.html <Accessed On: Friday 14th February 2013>

                                                                                                                                                                                                                         

 

 

Scenario 5:

        A mining company wants to get a better idea of the volume they remove each week. They don’t have the money for LiDAR, but want to engage in 3D analysis (Hint: look up point cloud).

      We are presuming that since the company the company wanted to use LiDar but could not afford it, then the type of mining they are engaged in is open pit and not underground. We came up with two possible ways for monitoring the amount removed from the mine each week. One approach would be to use digital imagery to detect the slag heaps, where measurements could be carried out from the data collected to detect the volume of matter removed. The other would be to use a cloud point method where laser detectors are used to detect and later recreate an area, these detectors would be flown over the actual mining pit in order to construct the space, and then once this is done over time we could see how the size changes and therefore the volume removed could be calculated.

          For the first method the data collection should be fairly quick so that type of unmanned air vehicle we would recommend would be perhaps a slightly cheaper unmanned aerial vehicle can be used, so as to save money. The flight time does not need to be that long so maybe even a battery operated one would be sufficient, a quad copter may be a good choice as it could fly straight up and over the slag heap and remain fairly steady and balanced for the image taking.

        This data collection would have to be carried out during the day so that the heap could be seen, and the time of year would only be an issue if the mine was located in a region that experiences winters with high precipitation rates, that might obstruct the camera’s view. As the company wants to know how much it removes each week, the image could be taken once a week. Then using computer software like ArcMap, the image could be downloaded a scale applied, and then volume calculations could be made from measurements made on the computer.

         3D scanning can be performed using a regular camera attached to a UAV and entered into the appropriate modelling software. A steady camera would be required, such as rotary wing copter with the ability to hover. A rotary copter is most suitable for hard to reach locations, which may include some mines. Using this technology an open pit mine can be visualized; from this visualization it may be possible to determine the volume of the mine. Another option may be to attach a specific 3D scanning camera to the UAV creating a point cloud mesh, this option is very similar to LiDar; but would increase the cost of the survey. Use of 3D sensor camera would yield a more detailed report of the mine and would be very similar to overhead aerial LiDar and ground LiDar surveys.

         The data collection for the cloud point method will be quite extensive, as the unmanned aerial vehicle that the sensor will be attached to will have to cover all of the exposed mine, and may at some points need to go down in to the pit. So definitely a gasoline powered on would be appropriate and with a long flight time, so we would recommend a General Atomics GNAT. Also, as the data needs to be recorded each week, the cost of the vehicle should probably be kept fairly low.

Sources:

http://www.geog.ucsb.edu/~jeff/115a/remote_sensing/thermal/thermalirinfo.html

http://www.dole-plantation.com/Pineapple-Cultivation

https://www.aaicorp.com/products/unmanned/shadow_m2