Wall Drawer: A Simple Solution to Traditional Drawing and Photographic Recording of Archaeological Features

The search for novel methods to record excavated features in archaeological sites has led to the invention or adaptation of many technological tools, ranging from photogrammetry to 3D reconstruction. Unfortunately, many of these methods require more training, more preparation and more expense than the traditional method of handing a measuring tape, plumb bob and graph paper to a volunteer. While the traditional method is the most convenient for many field directors, it is often subject to inaccuracy and can be very time consuming. In contrast to both the traditional and the more expensive methods, we have developed a simple protocol using a computer program which can process digital photographs to remove parallax problems and define the edges (e.g. boundaries between blocks in a wall) in the resulting image. The end result is a properly scaled, un-warped image, with or without highlighted edges, which can be printed or stored electronically. This method requires the time of only one operator with minimal training in digital photography and a brief introduction to the program. It has been successfully employed at the Brown University Excavations of the Petra Great Temple, Jordan. Introduction: The Recording of Archaeological Features The accurate recording of excavated features has been a problem for archaeologists since they have striven to define archaeology as a science. Respected guides on the methods of field archaeology (Joukowsky 1980; Sharer and Ashmore 1979:251-252; Fagan 1988:273) describe the importance of recording features, especially wall surfaces and balks, with both scale drawings and photographs. The most typical technique involves the use of pegs to secure a level line, from which a tape and plumb bob are hung in order to find coordinates of points on major features. These points are plotted onto graph paper and then the smaller formations between these major features are sketched in by hand. While simple and relatively technology-independent, this method suffers from many drawbacks. Manual drawing can be extremely time consuming and sometimes dangerous to the archaeologist, especially when dealing with large features, like the 6 m high walls uncovered at the Petra Great Temple. The use of pegs can sometimes damage delicate features, or the pegs can shift, causing the line on which all measurements are based to move, distorting the drawing. The act of drawing itself is subjective since, consciously or not, we human beings focus on things which we deem to be important and ignore the rest. This subjectivity brings certain features into focus and makes them more accessible, but if an important detail is not drawn, the information is lost and cannot be replaced. Human beings also tend to prefer aesthetic drawings to ones which honestly record the features, and so may alter the image to make it look better. Manual drawings are inaccurate, since the thickness of a pencil line on a 1:20 scale drawing can be scaled up to 2 cm (Avern 2001:4), and since many of the lines are drawn in free-hand. Also, drawings go through many iterations, from re-mapping to inking, which increase the error margins. The end product of the drawing effort can be poor, resulting from a sketcher with limited artistic abilities working on paper which is often damaged or dirtied in the field. Alternatives to manual wall drawing are becoming more popular and more widely-used, in the hopes of producing a system that overcomes the limitations of traditional data-collection. While most of these systems solve the problems of accuracy and time, they tend to come with their own problems including expense, inconvenience and technical complexity. Photography is one of the easiest supplements to drawing, since it solves the problems of subjectivity, time and permanence. Unfortunately, because photographs must be taken from one point, parallax distortions are almost universal. Various styles of bipods, tripods, quadrupods and booms have been used to help minimize the distortion (Poulter and Kerslake 1997), but it cannot be entirely removed from any normal photograph. Because of the