Chapter 4: Remote Sensing Technologies

Remote sensing provides digital information on land and earth features that can be combined with spectral analysis and GIS modeling to create a powerful screening tool for transportation corridor or regional evaluation. Remote sensing can quickly and cost-effectively categorize and quantify land cover types (wetlands, crop lands, forested lands, etc.). When combined with topographic, environmental constraint, geological, and planimetric information, this data also can be used for quantitative description and evaluation of plan or project alternatives. Combining remote sensing and GIS capabilities offers the ability to present plan or project scenarios in a three-dimensional environment, providing decisionmakers and the public with a clear picture of potential impacts.

Six remote sensing technologies have been selected for evaluation:

• Terrestrial lidar
• Airborne lidar
• Digital aerial photography and photogrammetry
• Radar imaging and mapping
• Ground-penetrating radar
• Multi-spectral and hyper-spectral satellite and airborne imaging

Lidar is a LIght Detection And Ranging system. It is analogous to radar except that light waves (lasers) are used instead of radio waves. Terrestrial lidar units are either tripod or boom-vehicle mounted. They have a limited range (up to 100 meters), but accuracy specifications are very good. The units are used to capture georeferenced x, y, and z coordinate values of thousands of individual points per second. Data captured by this type of system can also be converted to three-dimensional and CAD/CAE model data.

The accuracy of data collected using this technology appears to be better than traditional field survey methods, especially for data points in hard-to-reach places. For this reason, it appears to have good application for bridge or overpass monitoring and retrofitting, construction monitoring, and generating terrain data in dangerous locations.

The speed of acquisition of the data points and thus the turnaround time from field to finish is an order-of-magnitude faster than traditional field survey methods.

This technology is most applicable to the preliminary and final design phases.

Due to its cost, short range of acquisition, and the size of data sets acquired per setup, this technology is most suited to limited-length transportation projects (bridges, overpasses, tunnels, intersections, and shorter stretches of roadway).

Data collected using this technology is suitable for modeling structural features and topography (e.g., bridges, spillways).

The strength of this technology is in its speed and accuracy of data acquisition. What would have taken weeks to capture using terrestrial photogrammetry or conventional field surveys can be captured in a matter of hours. It is especially useful for collection of data in hard-to-reach or dangerous places.

Currently, because of the high cost of equipment and the limited number of vendors available, Terrestrial lidar acquisition and three-dimensional modeling services are being value priced at what vendors believe is the equivalent terrestrial/close-range photogrammetric mapping costs or traditional field survey acquisition costs. Manual intervention is required to ensure accurate, reliable results. The technology has a short data acquisition range, which limits the scale and type of appropriate applications.

Lidar is a LIght Detection And Ranging system. It is analogous to radar except that light waves (lasers) are used instead of radio waves. These systems consist of Kinematic (realtime) GPS receivers, an INU, and a laser measurement device mounted in either a fixed-wing aircraft or a helicopter. The units are used to capture georeferenced x, y, and z coordinate values of thousands of individual points per second. Post-processing of the points is required to reduce the point data to final output and to filter the data to bare earth (last return) or first return (or both) data points. In addition to bare earth, terrain data also can extract elevation data for above-ground features such as vegetation heights, building heights, and transmission towers/lines.

This technology has the potential of greatly decreasing the turnaround time for the production of digital orthophoto base mapping and digital terrain model creation for preliminary highway design and study purposes. Its speed of acquisition and processing to usable terrain data surfaces is an order-of-magnitude faster than currently available by field surveys or photogrammetric mapping methods. It has matured significantly during the past few years.

This technology is applicable to the preliminary design and study phases of transportation projects. It is not currently accurate enough (vertically) to be used for final design purposes.

Airborne lidar is suitable for corridor or smaller scale applications but not at a regional or statewide scale.

This technology provides source data for transportation planning and design applications such as CAD, CAE, and GIS. It also supports visualization "fly-through" applications.

There is currently only limited access to this technology, but its availability is rapidly increasing. Its primary value to users is based on its order-of-magnitude quicker turnaround time and increased accuracy over conventional methods for the production of digital terrain model data, which offsets the high cost when time is the critical factor.

Currently, because of the high cost of equipment and the limited number of vendors, airborne lidar acquisition services are being value priced at what vendors believe is the equivalent photogrammetric mapping costs. Other limitations include the need for manual intervention to ensure accurate, reliable results and limited range of suitable geographic scale and delivery phase uses.

This technology includes the acquisition and manipulation of aerial photography from airborne platforms to produce project displays and base maps to either help conceptualize the project, explain it to the public, or provide base mapping or other data for study and/or design purposes.

Aerial photography and photogrammetry are not new technologies. The new part of these coupled technologies is their relatively recent evolution to "digital" technologies. This evolution, spurred by the rapid evolution of computers—especially their processing power increases and disk storage advances—have provided the tools necessary for the creation and subsequent use of digital aerial photography and photogrammetric products for transportation projects.

Aerial photography (especially higher resolution color aerial photography) is easier for the public and other laymen to visualize than line-drawn maps, and it contains more information. It also serves as a snapshot archive in the history of the project. If designed and produced using photogrammetric mapping techniques, it is also defensible with respect to map accuracy and geometry as well as with respect to real-world coordinate locations, distances, and areas.

In addition to aerial base mapping as a backdrop, these products can be used for multiple purposes such as the creation of realistic "fly-throughs" (with and without future construction features imposed) for public and/or stakeholder presentations. Digital (raster imagery) products are readily displayed by GIS, CAD, and CAE software. Attribute and/or hot-linked information can be attached.

Digital photography and digital photogrammetry are generally conducted during the start-up elements of all delivery phases and then used throughout the process. The determining factors are scale, accuracy, and resolution requirements. Unless the project’s alignment is constrained and the schedule is condensed, multiple scales and multiple acquisition dates will be the norm for a cost-effective project.

This technology provides source data for transportation planning and design applications such as CAD, CAE, and GIS. It also supports visualization "fly-through" applications.

Based on proven technology, digital photogrammetry can be applied to detailed site mapping to regionwide planning activities. The technology is widely accept by planners, engineers, and the public. It is extremely useful for presentation of data to the public because it provides visual landmarks people can recognize. The visual imagery provides documentation of the project area that can be used in response to issues arising during right-of-way acquisition processes. Images can be easily integrated into GIS and CAD/CAE databases.

The cost of production and length of time required to produce digital aerial photography base-mapping products have been substantially reduced over conventional methods, especially when combined with other enabling data collection technologies (GPS and lidar, for example). Because base mapping is a front-end critical-path task, these reductions directly affect the final cost and schedule for the overall project. Cost and schedule will continue to decrease as other enabling technologies, such as digital cameras, INU devices, and lidar mature.

Limitations are similar to conventional photogrammetric applications (scale, accuracy, content), all of which can be overcome with advanced planning.

Radar, which is an acronym for RAdio Detection And Ranging, refers to a specific portion of the electromagnetic spectrum. Advantages of the use of this data source include its ability to penetrate haze, light rain, clouds, and other atmospheric conditions. Radar images of the earth may be obtained at any time, day or night, under most any condition. The images are often very different from those taken in other wavelengths and hence reveal supplementary information about terrain surfaces.

There are many types of radar, both passive and active (passive simply records natural electromagnetic emissions, while active records the response to a burst of energy transmitted by the radar sensor). The active sensors are of more interest in this context.
Radar sensors are carried on both airborne (airplane based) and satellite platforms. The most common types of radar are Side-looking Airborne Radar (SLAR), which pulses a beam at an angle toward the earth and records the time required for that beam to return, and Synthetic Aperture Radar (SAR), which uses the motion of the sensor to observe the same study area from several locations to simulate a very large antenna.

Other types of sophisticated radar are interferometry and multipolarization. Interferometry is basically using multiple images of the same location on the earth to generate terrain information. Multipolarization utilizes the transmission and receiving of horizontally and vertically polarized signals. The technology is particularly useful for vegetation studies, because the data will allow the analyst to observe different types of species and to "see" the volume and distribution of trees under forest canopies.

Radar mapping and analysis should optimally occur in the analysis and assessment phase of project development. It would be expected to provide useful input to the later design phases.

This technology provides source data (digital elevation model and imagery) for planning and preliminary design applications such as CAD, CAE, and GIS.

Radar mapping is valuable for geologic analysis of a study area, either to map terrain that is difficult to visualize by conventional means (continual cloud coverage or heavy vegetation cover) or to generate topographic information. Because of limited access to the technology and high cost, radar mapping technologies are best utilized when conventional means are not successful or feasible.

Limitations are similar to conventional photogrammetric applications (scale, accuracy, content), all of which can be overcome by advanced planning and sensor selection. Achievable levels of accuracy limit the use of this technology to planning and preliminary design applications.

Ground-penetrating radar (GPR) is a noninvasive method of mapping subsurface features and composition of a site. By pumping radar pulses into the ground and creating images of the radar reflections, a detailed picture of potentially important aspects of a site can be obtained before a single shovel of dirt is lifted.

Contrary to popular belief, GPR is not a new technology. GPR systems were originally developed by the military and have been in commercial use for more than 30 years. It is only recently that the environmental, construction, and utility industries have discovered the multiple uses and benefits of performing GPR surveys to gain forehand knowledge of what is underground.

GPR can be thought of as a subsurface imaging system, similar to sonar used for underwater applications. A radar device is like a very sophisticated stopwatch. A transmitter will generate and emit a pulse (wave). If something is in the path of the pulse (a target), it will either deflect, reflect, or absorb the wave. If the wave is reflected, the pulse is picked up by the receiver and processed. The time it takes for the wave to return can be calculated and equated to the distance of the target from the transmitter. In addition, by analyzing some of the characteristic properties of the return pulse, additional information about the target can be obtained. GPR works on the same principle.

During subsurface investigations, especially civil surveys, GPR surveys often are performed near sensitive electronic or tenant-occupied spaces. GPR technology is quite benign. The energy source is a low-power radio frequency, so there are no deleterious effects from destructive radiation and no need remove people from the site to conduct the survey. The antennas used in civil surveys are fully shielded to direct all of the transmitted energy into the ground and to eliminate surface reflection artifacts and radio frequency interference common to an unshielded system.

GPR is an appropriate technology for initial site or alignment selection and, for later in a project, to examine critical areas in a corridor prior to right-of-way acquisition negotiations and permitting activities.

This technology is best applied on small sites.

GPR data are used for evaluation of specific sites by locating archaeological artifacts, geologic features, and subsurface utilities and storage tanks.

As opposed to other locating techniques that are capable of detecting only metallic or conductive utilities and underground targets, GPR can locate and characterize both metallic and nonmetallic subsurface features. It is completely nonintrusive, nondestructive, and safe. Surface conditions are not a major factor. GPR targets can be "seen" beneath reinforced concrete, asphalt, gravel, and most other common surfaces.

GPR is extremely useful for examining possible archeological sites and grave sites without affecting the subsurface material, and it is often used to identify the precise (x, y, and z) location of utilities and hazardous waste receptacles. Determining depth to bedrock (if near surface) or voids can also be accomplished with GPR. The potential for reducing liability, minimizing safety concerns, and widening the time window for data acquisition all have positive effects on a project. The noninvasive aspect of the technology application allows cost-effective data collection when compared to conventional methods requiring open excavation.

The technology is best applied on small sites due to the small footprint of data collected by a GPR device. GPR data are limited to a depth of approximately 10 feet.

Use of qualified technicians is critical. Determination of effective data collection procedures and accurate interpretation of GPR data requires both training and experience.

This technology includes the acquisition and manipulation of multi-band digital imagery whose "bands" are spectrally segmented and unique. This imagery product provides a "picture" of the project area, which includes both visible features and information that is imperceptible to the human eye. This product is beneficial to a project as a base map source but more importantly as a source of spectral analysis that may result in distinguishing and quantifying landcover characteristics.

Imagery can be acquired either via satellite or airplane. Multi-spectral imaging has been a viable satellite technology for more than 20 years, but the spatial resolution needed for most "engineering"-type tasks has not been available from these satellite sources. Although the spatial resolution of available satellite imagery has increased, it is still too coarse for most engineering tasks. The value comes from the associated tasks that support the design effort. Recently, Space Imaging Corporation launched and put into production the IKONOS satellite, which offers 1-meter panchromatic and 4-meter multi-spectral (4 bands) spatial resolution for nearly anywhere in the world. This level of detail may offer information not previously available in a timeframe not previously possible. Otherwise, there are several satellite programs that offer the more common 20- to 30-meter spatial resolution multi-spectral imagery for use in environmental analysis.

The increased spatial resolution provided by the airborne (airplane-based) sensors has only become possible with the advance of computer technologies, such as large data storage arrays and geopositional information capture, in addition to the obviously necessary optical advances.

Every landcover feature, such as open water, has a spectral signature. This spectral signature is the "curve" that represents the character of the electromagnetic energy reflected from the landcover feature. In the case of open water, if there is very little turbidity, then there will be very little reflection in the near and mid-infrared ranges. Water absorbs light in this range. This sort of specific, predictable characteristic is what defines a spectral signature, and thus allows the semi-automated delineation and quantification of a wide range of landcover types.

Hyper-spectral imaging is a newer, emerging technology. It operates in much the same way as multi-spectral imaging, but the spectrum is segmented into many more divisions of finer definition. A hyper-spectral image could have tens to hundreds of spectral divisions (i.e., spectral bands), thus allowing for a much tighter fit between actual spectral signatures and spectral signatures able to be perceived.

All of this means simply that landcover types, vegetation species, and even the presence of specific chemical compounds can be identified from a remote, satellite location. These landcover classifications are also often critical inputs into watershed and hydrologic modeling, although interpretation and finite-data extraction is necessary to provide data to support a model or similar decisionmaking tool.

Similar to the use of traditional photography, these products can be used for multiple purposes such as the creation of realistic "fly-throughs" (with and without future construction features imposed) for public and/or stakeholder presentations.

Optimally, the data acquisition, preparation, and initial spectral analysis would be conducted during the preliminary design phase to maximize input of environmental data to design decisions. Because of differences in their resolution (higher for airborne and lower for satellite images), the two technologies are not applicable to all project delivery phases.

Geographic Scale Applicability

Application of the satellite technology is suitable for all scales except very small geographic-scale projects. Applicability of the airborne technology is limited to local area or smaller geographic scale projects.

This technology provides source data for base mapping and automated classification of land cover features. It also supports both GIS and CAE applications.

Multi-spectral image processing is a proven technology that is suited very well for transportation projects, especially because the corridor nature of most study areas is easily repeatable in the data acquisition process. Satellite-based spectral sensors have the desirable characteristic of having a radiometrically consistent footprint—no vignetting or variation of spectral values across the captured image. As a result, as soon as an image has been acquired and georeferenced, it is immediately useful as both a base map and an analysis source. With little field investigation, spectral analysis can begin. The resulting product is an intelligent snapshot of the landcover-related environmental conditions in the project area. The likeness of the products to traditional photography mapping is easily recognizable to planners, but the additional spectral capabilities of the imagery incredibly expands decisionmaking options.

Limitations are similar to conventional photogrammetric applications (scale, accuracy, content), all of which can be overcome by advanced planning and sensor selection. Hyper-spectral image processing requires extensive training and experience to derive accurate output products. Airborne imagery has limited availability and is provided by only a few vendors. Airborne imagery results in very large data sets that can be difficult to handle. In general, costs increase with increased image resolution.