Unmanned Aerial System Mission Planning

This assignment was aimed at getting us thinking about the re-planning and preparation that goes in to any type of field work, and especially where very high tech expensive equipment is used. As, any time lost in the field due to lack or preparedness, or failure to assess if the conditions are suitable, can waste significant amounts of money.  

In thinking about the Geographical world today and many of the data collection that is carried out, some astonishing technological advances have been made. Data can be collected, quicker, more often, to a higher standard, with less people, and different data types can be collected. Much of it involves out sourcing to other professions that have a better understanding of the data collection. But these professionals can sometimes have a lack of understanding for the sort o f geospatial thinking that is required by geographers to carry out fieldwork.

In this assignment we were given 5 scenarios in which companies may require unmanned air vehicles in order to collect data in aspects of geographical phenomenon. We had to think about the different things we would have to survey, how we would do that, the inputs that would be needed from technology to a sampling method. In order to come up with some ideas that a company could use in these situations.

Scenario1: 

 

-       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 pooper quality vegetation, and which area was best to harvest then, these trends might apply across a few growing seasons.

I order to highlight the areas where the vegetation is less health than others, we felt that a near infrared sensor could be used. Near 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: An example of a near 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 camera could 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. 

Scenario 2: 

 

-       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 couldn’t 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.

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.

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 2: An Example of the Unmanned Air Vehicle that we feel would be appropriate to use in the data collection.

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, as shown in figure 2 above. 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. The data collected from the fieldwork can be shown in figure 3 below.

 Figure 3: An example of the data collection model that would be created during the U.A.V.  surveillance. 

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 creating a habitat map would be more expensive.

Scenario 4: 

 - 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.

 

 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.

 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.  

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. 4), 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. 5), 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 4: 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 5:  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).

 When fitted with a fuel engine, the Avenger by Leptron (fig. 6) 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 mph 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. 7) 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 6: Image of the Avenger 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 7: Some of examples of 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).

 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).

Scenario 5:  

 

- 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.

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. 8). 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 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 8: 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

  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 9:  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.

 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 10 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 different 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 11 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 cost. 

Figure 10: 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 11: 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

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 12: 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 13: 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).

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.

 

Sources:

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

Clark, R.N., Swayze, G.A., et. al, 2010 , A method for qualitative mapping of thick oil spills using imaging spectroscopy: http://pubs.usgs.gov/of/2010/1101/ (accessed February 2014).

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

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

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

2014).

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

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).

Unknown. Unknown. 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/

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>