Preparing a map
Using transformed images in a GIS or CAD gives an interpreter the facility to zoom in and out as required. This allows an overview to be seen as well as detail and it is easy to switch between images to verify whether a feature occurs on more than one, or to find the clearest view. Zooming in too close can result in a loss of the sense of the overall picture and there is a danger of beginning to draw the edges of individual pixels.
Use of original photographs
If you have copies of the original photographs and are working at a computer it is useful – if not essential – to constantly refer to them to check, clarify and reinforce what is being seen on screen. Using stereo pairs will give confidence to interpretation and can help eliminate processing faults that may occur on one image and, by allowing reconstruction of the 3D view, help place the archaeological features in their topographical context. Some books on photo interpretation in other disciplines go as far as saying that full interpretation is not possible without use of stereo pairs.
Colours, conventions and metadata
It is usual to show different kinds of features – for example: ditch, bank, wall, palaeochannels, backfilled quarry – using a range of colours and/or on different layers. Use of standardised colours or conventions makes it easier for others to understand what has been mapped. It is expected that, at any scale, finished maps will include a layer to show, for example 'possible archaeological ditch' to indicate features of which the interpreter was uncertain but which may be archaeological in origin.
It is important to record which photographs have been used to make interpretations. This can be done definitively in GIS or CAD by linking each drawn line with a small metadata file that will identify the main photograph(s) used for its interpretation. These files can also include notes about what the feature was thought to represent and its degree of certainty. This information becomes valuable if an interpretation is queried because the only reliable way to check what has been mapped is to use the same photographs. Less definitive, but just as important, is to list all photographs that were examined and the date at which the interpretation was completed since additions to collections of photographs can include 'new' old images as well as those newly taken.
Levels of interpretation
It is usual to decide the scale at which a map is to be compiled before starting an interpretation. This affects not only accuracy and the level of detail that can be shown but is also a basis for making decisions about what types of information are to be sought and mapped.
Use of 'levels' is to explain the different ways we think as we interpret images rather than to propose a set of standards for our work. Interpretations made at 1:10,000 level are likely to be undertaken to build up a map of an area that will show different types of archaeological features in their topographical contexts and indicate, if possible, the relationships between them. Some generalisation or 'smoothing' may be necessary, especially of very small details that cannot be shown at small scale. Work at 1:2500 level will be expected to interpret and map considerably more information. Archaeological features are likely to be drawn to show as much detail as can be seen on the images and to include, for example, changes in ditch width, accurate sizes of breaks in ditches, all pits and postholes, buttresses on walls...and so on. Interpretations made at this level are likely to be used as the basis for further investigation which may include field walking, geophysical survey and targeted evaluation trenches, or which may be necessary for management purposes.
A finished interpretation may be filed away as part of a larger project but at some date, publication may be needed. As with interpretation, there are a number of reasons for publishing and these will affect the final presentation.
There are a few generalities that can be noted:
- the chosen scale should be the best to satisfy the reasons for making the interpretation,
- it has become common in the UK, for example, to publish areas of landscape at 1:25,000 and detail at either 1:5000 or 1:2500,
- finished drawings should be related to a national grid or some other means through which others can identify the locations of features shown,
- completed interpretations should be accompanied by a list of all the photographs consulted in their preparation and a date of completion.