Table 1. A summary of the relative merits in terms of field deployment of optical laser scanning versus photogrammetry.

	Photogrammetry	Optical Laser Scanning
Costs
Hardware	Low field costs since a basic digital camera and memory cards are all that is required. Modest lab costs associated with provision of suitable CPU, dependent on the speed of processing required and software to be run; reducing all the time as standard computational power increases.	High depending on the make and model of the scanner used. Low lab costs since no special computational power is required unless a large number of scanned images are being tessellated.
Software	Zero to modest depending on the software used to generate photogrammetric models. Three-dimensional image software required for post processing and visualisation both commercial and freeware options available.	Variable, most expensive scanners come with basic three-dimensional image software required for post processing and basic visualisation less expensive scanners often don’t.
Deployment
Transport logistics	Easy - photo-scale and camera. In some cases use of tripod mounted arms or A-frames may increase the equipment volume.	Depending on scanner model and the support mechanism – tripod or frame - can be quite bulky. Provision of power supply via a converter and a generator, car battery or lithium ion battery.
Electrical Requirements	Minimal, power is required for camera batteries and photo storage devices such as a laptop or PDA.	Most scanners either require a generator, car battery or lithium ion battery with or without a power inverter, either to power the scanner directly, or to recharge a built in battery. Power is also required for PDA or laptop used to run the scanner.
Data capture time	Approximately 5 minutes per print to take between 20 and 30 photographs per print; quicker times possible when using fixed point frames/tripod requiring a more limited number of images. It is possible to have multiple prints or areas being captured simultaneously with multiple photographers. Photographs can also be collected from Unmanned Aerial Vehicles (UAV) especially where large areas are involved, although this may increase the associated costs and logistics.	Depends on the scanner model and resolution required but usually less than 1 minute per scan. Limited to the number of scanners available to one field project.
Post-processing time	Depends on the software being used and the number of images but post-processing time to generate the model can be up to 12 hours, typically 30 to 45 minutes for a high resolution model.	Depends on the tasks being performed and the degree of data cleansing and optimisation required but can be anything from a few minutes to 30 minutes maximum. Aligning multiple scans, especially from long range scanners with high data throughput can take considerable time (up to 24 hours).
Reconnaissance operation and/or training?	Images can be captured by any operator with a digital camera and basic knowledge of photographs required.	Requires access to equipment and basic training.
Memory Requirements	Can be managed by multiple data cards, field based download to laptop or PDA, or field based upload via internet connection. Data volumes are high depending on the individual pictures resolution; for example, one gigabyte for a trail of 10 prints.	Depends on the make and model of scanner, some scanners can record directly to a data card, most required laptop operation. Typical file sizes are between 1 and 5 megabytes per print, though high resolution scans of large areas (e.g. whole or partial track sites) can be many Gb in size.
Risks to site	Damage can be high from feet of photographer taking multiple images from different angles; damage from the feet of tripods or other fixed arm camera mounts. These can be overcome through the use of UAV’s although their use increases costs and logistics	Damage from tripods or scanner frames can be high.
Accuracy of Outcome
Prohibitive environmental conditions	Sunlight & intense shadow can be problematic and shading may be required for the whole area of the print depending on the colour of the substrate and angle of the sun. Wind-blown dust and rain may hinder operation. Wet rock/sediment surfaces or those with residual water content can limit the accuracy of some models especially where it is variable across a surface.	Most high resolution optical scanners require sunlight shading and protection from wind-blown dust and rain. Scanners can fail to operate in very high ambient temperatures due to sensitive components. Air moisture can also cause interference and laser detection issues.
Accuracy and completeness	Dependent on the quality and number of images obtained and the software used to produce the model. Undercut areas can cause problems as can deep prints causing shade problems at the bottom of the print. For accurate measurements images have to be carefully scaled.	Dependent upon the make and model of the scanner. Difficult to capture undercut or overhanging areas with a vertically mounted scanner; multiple shots may be required and there still may be problems with very deep prints. Scans are scaled accurately as they are captured, provided the scanner is regularly calibrated.
Intra- and inter-site variability	The accuracy of a photo-model is specific to one object and the images taken, there is therefore a strong risk of undetected intra- and inter-site variability in accuracy and reliability of the models. The accuracy of every single model needs to be checked via a reference object in every model.	Provided a scanner is well-maintained and regularly calibrated by the manufacturer its accuracy should be consistent in intra-site setting and inter-site settings subject to a caveat around changing environmental conditions. The accuracy of scanned images needs only to be checked once at a site, or following best practice daily at most.
Edge effects	Taking images close to an excavation wall can be problematic since a full 360° array of images may not be possible.	Depends on tripod or frame configuration, but potentially not a problem especially if oblique scans are also used.
Risks of failure	Data quality - moderate to high, associated with failure to capture sufficient images of good quality and coverage especially when post-processing is being done on return from the field. Equipment - low since cameras are ubiquitous on field expeditions so multiple options are often available when one camera fails assuming flexible camera mounts and tripod connections. Post-processing – moderate to high, failure of the software to produce adequate models.	Data quality - low in terms of failure to capture data since the quality of a model can be instantly verified and checked in the field and scans re-shot if needed. Equipment - moderate to high since scanners are relatively delicate scientific equipment and field failure is usually terminal since few projects have access to multiple scanners. This is low for scanners designed for field use. Post-processing – low focused simply on data quality and enhancement.