Producing An Archaeological Geophysics Report

Page Index

[ Archaeological Geophysics Page ]

Introduction

So you have finished the survey and downloaded the data to your computer, what do you do with it now? After you have a nice image, how do you interpret and present it? How is the report and your data to be stored for future use? These are the questions answered in this section.

Processing The Data

We will assume here that you have downloaded your data into the software you are using the instructions applicable to that software. We will also assume that you have taken the individual grids and applied the layout of those grids in the software to produce a composite image. Now comes the fun part, applying the various filters to produce a nice image and interpreting the results, assuming the software you are using comes with these filters.

Now you have a basic image to work with, you need to look at it to see what filters will help. I will cover here the most basic filters, which you will use most. Your software may have other filters which may be of limited use, but you will need to consult the documentation that comes with your software for those. Not all filters are useful in all situations, and it is worth remembering that filters should be used in a certain order for the best results.

Most images will have some 'noise' which is not applicable to the features you are trying to identify. In resistivity, you may have hit a stone and recorded a very high resistance reading. In magnetometry, there may have been a part off a tractor buried in the field causing a magnetic spike. By noise, we specifically mean a single high or low spike within an area of otherwise average readings. Whatever the reason for this noise, it can skew the results of other filters because of the extreme readings involved. So before you apply any other filters, it is worth using the Despike filter to remove these.

Before Despike
Before Despike
After Despike
After Despike

Next for resistivity, after the Despike, you should now be looking at the edges of the individual grid squares. If you can see where one grid finishes and another starts by a distinct difference in the readings between the two squares, this needs correcting. In resistivity, this is usually caused by the fixed probes being moved. Your software will have a tool to match these grids so one should flow seamlessly into the other. In Geoplot, this is called 'Zero Mean Grid', in Snuffler, it is called 'Edge Correction'.

Before Edge Correction
Before Edge Correction
After Edge Correction
After Edge Correction

Continuing with resistivity, you should now have an image which is free of artifacts caused by the survey process. You may be able to see some features, but there are ways of making them clearer. The local geology may play a significant role in your image. If these are small features such as solution hollows or cracks in the bedrock, i.e. high frequency features, then you are stuck with these as the features you are looking for are also high frequency. If on the other hand, you have a slow change in the depth of the topsoil across the site, causing a slow but visually significant change in the readings, then there is a filter called a 'high pass filter', or 'Remove Geology' in Snuffler which can help. This filter removes these slow changes and flattens out the site, making the 'high frequency' archaeological features easier to see. You are usually asked to supply a sample size. The filter works by taking an average of the readings around each reading to calculate a background. If you take a large number of readings for the average, then the background will change less and there will be less flattening of the image. If you take a small number of readings, the background will change more and you may start to loose the features you are looking for. You should try several sample sizes to get the right one.

Before High Pass Filtering
Before High Pass Filtering
After High Pass Filtering
After High Pass Filtering

Moving on to filters specific to magnetometry. The setup of fluxgate gradiometers, and thei directional nature, will produce a stripy effect across your surveys. This can be removed with a destripe feature. Particularly strong signals may cause this filter to leave, or even produce stripes in places. Destriping per sensor/direction may help here, which takes a wider average rather than just dealing with each line one at a time.
Before Destriping
Before Destriping
After Destriping
After Destriping

The next useful magnetometry filter is destagger. The machine operator may consistantly misjudge the position on the grid by a small margin, or walking on a slope might make it difficult to walk at a steady pace. The result can be a staggered effect on linear features. The destagger filter can correct these sort of errors.
Before Destaggering
Before Destaggering
After Destaggering
After Destaggering

The last feature to be used, for both resistivity and megnetometry, and it must always be the last, is the interpolation filter. Interpolation will increase the number of readings and guess what is in between, thereby smoothing out the image and making it easier on the eye. Whilst it doesn't provide any more literal information in the image, it takes advantage of the way the eye and brain work in processing information to help you see the features instead of concentrating on the differences between individual readings.

Before Interpolation
Before Interpolation
After Interpolation
After Interpolation

Now you have a nice image, you can try and work out what archaeological features are visible. They may not be obvious straight away, prolonged consideration of the image may be needed, especially if the area covered is quite large. Comparison with an aerial photograph is useful as the two can provide complementary information. The process is best done by viewing the image on a screen rather than a sheet of paper as the colour reproduction is usually better on a screen. Also, get many people to look at the image, as other people can often spot things that you miss. Other than that, recognising features is not something you can explain on a web page, you just have to get experience by looking at many images to be able to distinguish between the archaeology and your local geology.

The Report

The final report can be summed up in a list of guidelines for information that should be included. If you are just doing a quick survey on a day to satisfy curiosity, and the results are nothing spectacular, then doing a full report is not worthwhile. A large survey or one with good results deserves a full report.

You should have recorded various pieces of information such as the dates of the survey, the location, survey area, the specification of your machine, the weather, the local geology and land use. This information should be given along with a note on any effect it had on your survey. The reason for doing the survey in the first place is also important, with a description of known features or other information that has caused this survey to take place, such as field walking or aerial photographs.

A diagram to show the grid layout and measurements relative to known features or landmarks is helpful. It will allow someone to return to the site and mark out the features you have found for excavation. Then of course the images you have produces using your geophysics software, with appropriate text, north arrow and and scales. The final diagram is an interpretation of the results, showing the survey grids with a diagram of the features found. If the period of the features are known, then they can be marked in different colours. Different colours should also be chosen for high and low resistance features. If colour is unavailable, then different types of shading can be used.

Finally, a text discussion of the results and their meaning along with acknowledgements and bibliographical references. Getting someone to proof read it is helpful if you are going to be sending your report to government departments so that you don't look like a fool. You can see an example of a report here, being A Resistivity Survey of a Neolithic Causewayed Enclosure and associated monuments.

The Archive

Even after the report is finished, your work is not over. A report is no good if there is no-one to read it. First of all, you have to make sure your report reaches the right people. Apart from all the people who have helped you out, there may well be several organisations who need a copy of your report, and here I give examples based upon England, where I live.

Firstly, a copy should be sent to the Historic Environment Record, which is held by the County Archaeologist. If you don't already know who this is, then you can contact the County Council to find out. If you have had to get Scheduled Monument Consent, then English Heritage will need a copy as part of that consent. Your local library may well be interested in a copy, as well as the libraries of any local or county archaeological societies. You will of course have received permission from the landowner to do your survey, so do send them a copy as well.

A digital archive is also desirable. Both of the report, and also the raw data from which you have produced your results. The raw data is best stored in the XYZ format, which is a standard text file that records the data along with its position. Your geophysics software should be able to produce this for you. Other information is best stored in Open File Formats, by that, I mean file formats which are not exclusively owned by one company and for which the details of the format are available for anyone to use. As far as image files are concerned, PNG files are best, both because it is an Open File Format, and also because it is not a lossy format, i.e. the compression in the file will not change the image. Any decent paint program should be able to convert your images to the PNG format for you. Documents are best stored in PDF format, MS Word is nice but Proprietory and constantly changing. HTML is open but you can't control the page breaks. If you can't afford Adobe Acrobat to convert your documents to PDF, then your can get them converted on-line if your want. After you have your digital archive ready, burn it to a CD and give copies to whoever you think needs it. Publishing your files on the internet is also desirable, but not if the site is of a period that gold-digging metal detectorists might be interested in. They may use the information to loot the site.

[ Archaeological Geophysics Page ]