Field
Methods:
Effects of sampling density
|
|
| The
traverse and sample spacing of measurements
employed in the field determines sampling density, the number of
measurements per unit area. In order to resolve an anomaly
at least one measurement must fall within it, and preferably many more.
A basic rule of thumb is that measurement spacing should be no more than
half the size of the minimum feature one is trying to resolve.
Thus, to resolve an anomaly 0.5 m in diameter measurements should be separated
by no more than 0.25 m. Sampling density is therefore a critical component
that determines the quality of information gained in a geophysical survey.
The figures above compare various sampling densities of magnetic gradiometry measurements obtained over prehistoric burned houses dating to the mid-13th century and located within Menoken Village State Historic Site, ND. At a sampling density of 0.25 x 0.25 m numerous small anomalies less than a meter in diameter (probably exterior food storage pits and small hearths) can be readily distinguished as can the shape of the principal oval-shaped dwelling, including its entryway ramp facing southwest and a centrally placed hearth. At 0.5 x 0.5 m sampling many of the exterior features become uncertain or invisible, although the main characteristics of the house still can be seen. The 1 x 1 m sampling density only indicates the locus of some major anomalies. Few minor ones and little about house form is indicated; its entryway ramp and central hearth are no longer clearly defined, for example.
|
Contribution
by: Kenneth
L. Kvamme, Archeo-Imaging Lab, University of Arkansas
The
effects of sampling density are also illustrated in a typical magnetic
profile, although the same principal applies to any data type. At a
coarse sampling interval entire anomalies can be missed making feature
identification and interpretation more difficult. Increasingly closer
sampling provides better definition and detail about potentially significant
features.