Phosphate Analysis: performing a phosphate analysis survey
Phosphate analysis is an efficient, inexpensive and non-invasive technique that can easily be applied to a variety of archaeological sites. Phosphate analysis can be used to locate archaeological sites and determine if there is an archaeological site present in previously unexplored study areas, to identify previously unknown archaeological features on known sites and to differentiate between activity areas in and around know sites and features to examine uses-of-space.
Phosphate analysis surveys are performed by testing small (50mg) soil samples. Phosphate samples are taken with a small gauge auger from 20cm below the ground surface when sites, or areas of sites, have not been excavated. This eliminates the possibility of alterations to the phosphate status of a soil caused by weathering affecting test results and limits the impact of the test on buried archaeological features. Phosphate samples are taken from occupation levels when sites have been excavated. Soil samples can be taken from occupation levels because phosphates released from decomposing organic material fixes immediately to soil particles. Leaching is very rare, and phosphates are generally only lost in soils with very low mineral content or through erosion. Soil samples taken from the topsoil reflect buried anthropogenic phosphate levels because while phosphates do not move horizontally in the soil, they do move vertically through the soil profile. Phosphate are drawn upward in the soil profile by plant roots, but anthropogenic phosphates cannot be used by plants causing them to collect in the topsoil. This anthropogenic phosphate "shadow" will continue to form and reform if topsoil is disturbed, thus creating a permanent indicator of buried archaeological features.
Phosphate analyses are most helpful when soil samples are taken on a consistent grid over a study area. Grid-intervals can range from 10m to 0.5m depending on the specific goals of an archaeological project. The grid interval used when testing for phosphate content can be varied based on the purpose of the testing. Smaller intervals are used if the testing is just being performed within one building, or for an in-depth analysis of intra-site phosphate patterns. A grid interval between 0.5 to 1 m is commonly used over known structures and floor surfaces to identify specific activity areas in detail. Projects that have used an interval of less than two meters for site boundary studies, such as at Cerro El Pajarillo, Portugal, have concluded that it was not necessary and that 2 meters would have sufficed (Sanchez et al. 1999: 50). The basic interval for a site identification survey is between 10 and 5 meters (Craddock et al. 1985: 368); the interval for a site boundary study is 2 meters (Litton and Buck. 1995: 21). A grid interval over 20m is commonly used over fields to locate settlement sites (Orser 1996). An interval between 5 and 10m is commonly used over known sites to identify areas of increased activity (James 1999). The grid intervals used in this study were chosen to test the readability of smaller and larger intervals.
To perform the test 50 mg of soil are placed in the center of a small round of filter paper and treated with two drops of Reagent A, followed by 2 drops of Reagent B exactly 30 seconds later. Ammonium molybdate is used because it allows phosphates to bond easily, and creates a visible blue color when treated with ascorbic acid (Reagent B), rendering the phosphate content visible. The test, therefore, is a measurement of how many active phosphates can be extracted by the HCl and bonded with the ammonium molybdate in 30 seconds by recording visible aspects of the reaction. Resultant color, length of lines radiating from the sample and “percentage of ring preset” around the sample are all measured after 2 minutes. The time it takes for the color blue to appear is also recorded. This is an accurate illustration of the percentage of inorganic active phosphates in the soil (Eidt and Wood 1974: 51-52). Other forms of spot-test extraction uses sodium hydroxide (NaOH) instead of HCl, but this test is used more as a positive/negative, and does not supply specific levels of phosphate content (Holliday and Gartner 2007: 314).
These results are mapped and analyzed (see pattern database) to identify activity areas, create use-of-space models and interpret site and social dynamics.