5 Sensing the Past and the Remoteness of the Future: A Soil Resistivity Survey at the Native Alaskan Village Site ANDRE P. TSCHAN HE FOLLOWING CHAPER consists of a short description of subsurface investigations, a condensed summary of one geophysical technique used in archaeol- ogy, and an analysis of the soil resistivity survey con- ducted at the Native Alaskan Village Site. SUBSURFACE GEOPHYSICAL SURVEYS The continuing encroachment of outlying properties by urban development initiates many rescue or "salvage" excavations. This unfortunate, but modem reality often involves large areas that need to be surveyed within a short timespan in order to successfully retrieve and conserve the most important traces of ancient life (Clark 1990:12). There is a demand for a technology, such as remote sensing, that can quickly produce detailed information on buried material contexts. In response to this demand, efficient, cost-effective, and feature-specific sampling strategies for excavations under time con- straints have been developed. Archaeological investiga- tions can incorporate various geophysical tools (e.g., soil resistivity, magnetic and ground probing radar testing) to facilitate the delineation of the spatial organization of site areas. Intrasite maps can be created from the data collected by applied remote sensing techniques to further determine detailed layouts of particular features and their boundaries. As a final point, it is important to understand that there are two main categories of subsurface detection methods according to Weymouth (1986:313): Passive = measuring gravitational/magnetic fields produced by buried features Active = measuring the return signal from a device's originally emitted electric/electromagnetic signal. The following summary highlights one of the most effective tools available. The focus of this report, however, lies in the soil resistivity case study of the Russian-American venture at Fort Ross and not in a specific analysis of the history and background of geophysical remote sensing equipment. Hence, a great many aspects relating to the usage and development of this methodology, as well as any description of other potentially rewarding techniques, have been deliberately omitted. SOIL REsISTIVITY SuRvEYs The important difference between the terms resis- tance and resistivity is that the former relates to a measurement of current flow while the latter indicates material properties (Sears, Zemansky, Young 1983:539). The underlying physical principle that forms the basis for this active geophysical remote sensing method relies on an electric current caused by the movement of charged particles. This net flow of charge determines the resis- tance of a conductor between any two points when applying a potential difference (V) between these points and measuring the current (I) that results (Halliday, Resnick, Walker 1993:771). The formula for electrical resistance is expressed by R = V/I in which: V = symbol for potential difference, or Volts I = symbol for current, in Amperes R = symbol for resistance, in Ohms (Q) Soil resistivity surveys have established that very little current is carried by earth, rocks, and pure water sources, since they actually exhibit insulating properties. However, rainfall that contains dissolved carbon dioxide and carbonic acid from the atmosphere forms conducting electrolytes that react with the minerals in the soil. Together with weakly conductive organic acids, these break down into negative and positive ions which are the actual carriers or conductors of electricity (Clark 108 The Native Alaskan Neighborhood 1990:27). In addition, moderni agricultural fertilizers and humic compounds, some of which have anthropogenic origins (pits, etc.), will increase the chances for a successful survey measuring the electrical resistivity (or conductivity) of the soil in a restricted volume near the surface. Potentially, any human alterations of the natural stratigraphy can be detected using this technique. Major intrusions from building activities for shelter and storage, as well as utilitarian activities, cause a distinguishable disruption of the geomorphological distribution in a site by exchanging materials from normally discrete soil layers and importing or exporting them into upper or lower levels. Thus, man-made scars in the landscape useful to resistivity research include areas of humus buildup (e.g., refuse pits), soil compaction (e.g., path- ways, house floors), and aggregation or loosening of soils (e.g., ditches, midden) (Weymouth 1986:320). Soil resistance data can be quickly recorded at a site by inserting two electrodes or probes deeply enough into the ground to make adequate contact. The electronic collection device will, upon achieving a closed circuit, store the resistance reading. The probes, nowadays mounted on a portable frame together with the recorder itself, can then be moved along a traverse to the next location. Assuming that the earth is a "semi-infinite medium," the hemispherically shaped research area is sliced in half by its upper and only boundary, the ground surface. Because the electrodes have a very small contact area compared to the volume of earth that is crossed by the current, there is a much higher resistance immediately around the probes than deeper in the ground (figure 5.1); the soil at the top also tends to be dry and to have a high resistivity. This problem is resolved by applying a four- terminal measurement known as the "Wenner configura- tion" (figures 5.2a and 5.2b). ARl in all, the function of the current electrodes is to set up a field of potential gradient in the ground that is sampled by the potential probes. In a homogeneous soil, half the current spreads to a depth of half the spacing between the current probes (figures 5.3a, 5.3b). As a result, buried archaeological features are likely to cause anomalous readings because they force currents to flow around them in an effort to find longer and easier paths (figures 5.4a, 5.4b, 5.4c). Thus, any past human intrusion clearly shows up against the general and most likely equipotential background of the untouched surrounding soil. THE SOIL RESISTIVITY SURVEY AT FORT Ross-A CASE STUDY IN COMPUTER-AIDED ARCHAEOLOGY The current analysis of the two main resistance data collections-the first collected in the summer of 1992, the second in December 1993-serves as a prime from a geophysical survey for archaeological research. Specifically, the unknown locations of house structures, middens, and activity areas in the Native Alaskan Village Site (NAVS) have become a great deal more decipherable due to the electronic resistivity maps generated. Hope- fully, the insights gained through this work will guide future research designs and will help to facilitate contin- ued excavational projects at Fort Ross. SENSING THE PAST AT FORT Ross The fundamental objective of the survey was to establish a detailed, electronically processed map depicting the resistivity of the soil within a depth of half a meter (0.5 m) (figure 5.5). The desired goal, in conjunction with excavation, was to arrive at a data collection set that would pinpoint the specific location, the spatial boundaries, and the intrasite organization of the Native Alaskan Village. Obviously, the area encompassed in the site grid is indicative of the potential time and resources required for a broad-scale investigation of this site. The initial research objective for the 1992 survey alone encom- passed some 7,600 square meters (figure 3.1). One of the main reasons, therefore, to conduct a resistivity survey on the site was to provide large-scale spatial information. Preliminary work undertaken by Dr. Lewis Somers (Geoscan Research USA), using a gradiometer, provided evidence of clear delineation and concentrations of features buried in the soil south of the Russian Stockade. Given the site's size and the exploratory nature of the initial magnetometer survey, further geophysical surveys were deemed necessary. With the assistance of Geoscan Research USA-providers of the equipment for the investigations in 1992 and 1993-the goal set by the researchers was to obtain, by testing the soil resistivity, a second and independent subsurface data set that would provide more infonnation to authenticate the original magnetic results while providing complete site coverage. MEFHODS Geophysical surveys are a collaborative effort involving remote sensing specialists, usually geophysi- cists, and archaeologists. In order to bridge the gap between the two disciplines, frequent and repeated interaction must occur to ensure a rewarding end-result. Any conclusions are the product of combining the available information of the two disciplines in an attempt to arrive at a single interpretive picture. This is, in my opinion, the only justifiable approach to archaeological remote sensing, for neither of the two scientific fields can be treated as mutually exclusive when it comes to analyzing subsurface data. STAFF The staff for the 1992 resistivity survey included example of the successful results that can be gathered John Anderson and myself as the principal investigators', Resistivity Survey at NAVS 109 Figure 5.1 Contact Resistance and Polarization Problems When Taking Ground Resistance Measurements (TextlGraph: Clark 1990:28) [' ~~- 1 ['' 1 + g C t ( _i Figure 5.2a A Resistivity Measuring Circuit (used in the Martin-Clark meters) I R I't Depicted is a Wenner electrode setup indicating equal spacing (a) between the probes. At this point, contact resistance is overcomc by applying separate potential/current electrodes while polarization is avoided using an AC supply. (Text/Graph: Clark 1990:28) Figure 5.2b The Detection Hemisphere Resulting from a Wenner Configuration (TextlGraph: Clark 1990:28) 110 The Native Alaskan Neighborhood Figure 5.3a Simulhiting a Wenner Configuration Cross Section of the Current and Potential Distribution P2 C2 The percentage of the total potential difference is indicated for each equipotential line while the current flow is represented by the crossing dotted lines. Half the current from electrodes Cl and C2 flows above the maximum depth of half their separation (indicated by the dark dotted semi-circle. This is the most sensitive area for the geophysical investigation). (Text/Graph: Clark 1990:29) Figure 5.3b The Potential Gradient betwveen Cl and C2 Represented as a Plot (Text/Graph: Clark 1990:29) Pi C2 as well as Tina Choy, Robert LaDue, Lisa Holm, and Laura Willman as assistants on separate days. The data was collected during a period of eight working days in June, 1992. Due to time limitations combined with some inefficiencies in sampling strategy and instrument settings, this initial resistance data set was as incomplete as the previous investigative magnetic survey by Dr. Somers. Thus, in December of 1992, John Anderson and I retumed to Fort Ross to collect more readings. The final and complete collection was not obtained, however, until December of 1993, with Lisa Holm and myself as the investigators. The decision for redoing the entire site arose after a series of discussions throughout the year with Professor Lightfoot and Dr. Somers. These talks prompted a new research strategy which was developed based on earlier survey results. Also influen- tial was another resistivity study on a nearby site (CA- SON-177), conducted by John Anderson and myself in February and May of 1993. Because the most compre- hensive principal survey was produced in December of 1993, any reference to resistivity data contained herein relates to that particular set, unless there is a specific statement to the contrary. EQUIPMENT AND INSTRUMENT SETrINGS The machinery used for this research included a RM15 Resistance Meter bolted to a half meter (0.5 m) array and Geoplot version 2.0 as the computer imaging I I II I Resistivity Survey at NAVS 111 Figure 5.4a A Centrally Located High Resistivity Feature Causes Marginal Interference of the Current Flow (Text/Graph: Clark 1990:28) Figure 5.4b The Percentage Difference in Potential between P1 and P2 Increases Due to the Feature Being Located Directly beneath P2 and C2 (Text/Graph: Clark 1990:28) Figure 5.4c A Centrally Located Conducting Feature Interferes to a Greater Extent Than its Counterpart Depicted in Figure 5.4a (Text/Graph: Clark 1990:28) 112 The Native Alaskan Neighborhood Figtire 5.5 Geoplot Resistivity Map of the Native Alaskan Village Site The Fort Ross Cove Road is visible on the left side of the image (linear scar in the landscape). Resistivity Survey at NAVS 113 program. Both the hardware and software were loaned by Geoscan Research USA which, in collaboration with its British partners, is responsible for building a variety of geophysical tools as well as writing the necessary software to translate the data into visual images for the computer screen or printer. The two fixed and frame-mounted spikes of the array, defined as the "mobile probes" (MP) and spaced at a half meter (0.5 m) distance, were connected to the RM15 Data Logger by an electrical cable; the digital collection unit was mounted on top between the handle- bars. In order for the injected current to behave in the desired fashion-namely to be completely hemispherical in its distribution through the ground-it was also necessary to have a set of permanent or "remote probes" located no closer than 30 times the sample interval spacing from the point where a reading was being taken (i.e., 0.5 m x 30 = 15 m minimum distance). Because the area of NAVS investigated was quite extensive (160 m x 100 m, maximum dimensions) and the electric cable hooking the remote probes to the RMl5 was only 150 m, three stations were established so as to ensure that none of the data would be jeopardized by too close a proximity between the remote and the mobile probes. Finally, the computers used for the processing work included an IBM clone with a 486/50 MHz processor, 16 MB RAM, and SVGA display as well as a Compaq Contura 4/25CX laptop with a 486/25 MHz chip, 4 MB RAM, and SVGA display. The main data presentation and manipulation software, as mentioned above, came in the form of Geoplot version 2.0, kindly loaned by Geoscan Research USA. This program is a DOS application and therefore manipulations of base memory were required for some data processing and in particular, the Interpolation utility. Any open terminate, stay resident (TSRs) programs reduce the capacity to perfonn this operation successfully. Overall, Geoplot Version 2.0 is a highly sophisticated tool for processing and visualiz- ing the collected readings using high resolution graphical formats. Thus, coupled with any of the Geoscan tools for subsurface investigation, one has a powerful, reliable, and cost-effective package to retrieve buried traces from the past Additionally, an Autodesk software program, AutoSketch for Windows Release 2.0, served as a precision drawing tool for numerous maps presented throughout this paper. Lastly, a series of Fort Ross archival photographs and map images were captured on 35mm film by Nick Jones upon our visit to the storage facility of the Depart- ment of Parks and Recreation in Sacramento, California. Each historical reference was recorded experimentally at four different light settings as well as in monochrome and color. Eventually all 104 images were developed on CD- ROM as Kodak Photo image files (PCD). Further computer processing took place using the Media Cyber- PhotoStyler, version 2.0. These software programs were able to correct the deficiencies that were inherent in the old maps, which were mostly enlarged, fairly poor copies of obscure originals. DATA COuECrION AND SAMPLING STRATEGIES In order to guarantee data congruency, fixed control locations were established and were monitored three times a day. At these "reference stations" we checked our readings each day to make certain that no overall changes occurred in the surveyed area during this investigation, either as the result of climatic influences or equipment problems (figure 5.6). For example, if the data had shown a significantly different value from the previous day's last reading, then the remote probe spacing of the spikes could have been adjusted to be greater or smaller than the original half meter (0.5 m). As a corrective measure, this procedure would have lasted until the resistance number was close enough to, or ideally matched, the desired output. This monitoring system was very important, since the absence of a comparable reference station setup dramatically impaired both of the 1992 sample sets. Weather conditions remained constant throughout the collection period, and no precipitation beyond morning dew was registered. Interestingly, but without consequence, the last day proved to be rather warm for that time of year, and we encountered increasingly high resistance readings. In retrospect, this "warming up" of the site was only a circumstantial phenomenon, greatly outweighed by factors such as the ground's general aridness as well as the complex archaeological composi- tion that prevailed among the grids done last according to our sampling strategy. The recurring dilemma of having to redo the site's grid layout because of park visitors repeatedly removing the wooden stakes necessitated that we locate a geo- graphic reference point that would be seen clearly on the electronic maps. The obvious choice, based on the previous surveys in 1992, was the Fort Ross Cove Road that leads down the cliff terrace to the coastal cove. This feature, however, was extremely impenetrable due to a cementing compound that makes up the basic com- pressed gravel track. Fortunately, the morning dew added natural moisture to the roadbed, allowing us to take readings with greater ease. This otherwise impass- able stone obstacle is apparently rich in minerals that readily dissolve in water, possibly salt crystals or other super conductors are broken up by passing motor traffic and disintegrate when exposed to rain or mist. Hence, we made effective use of the night's residual moisture on the ground, and the road became the focus of most of our pre-noon data collection. The attempt to achieve our goal of surveying the entire expanded 1993 site (figure 5.7) within three netics HALO Desktop Imager, version 2.00 and Aldus working days was somewhat ambitious. Because the cliff 114 The Native Alaskan Neighborhood Figure 5.6 Survey Reference Stations MAP LEGEND A Datum Point (OS, OE) C Grass Field g Fort Ross Cove Road Cliff vwth Rock Outcroppings L] Beach Sand and Rocks Survea Reference Stations RM1 5 Remote Probe Locations = 89.5S, 34.5W =92.5S,8E = 60S, 20E RM15 Mobile Probe Locations 120S, 40W t=4OS, 27W '405, 40E G = General/Main Ref. Station 0 10 20 30 40 m 0 2 4cm Os - 1os - 20S - 30 S - 50s s - 60S - 70S - SoS - 6 .~~~~~~~~~~~~~~~~~~~L 60W SOW 4OW 30 W 20 W IOW O E iOE 20 E I - I I I I I I I I I Modifiedim 0i Co Deasiby Edwwd C Mxwk . Resistivity Survey at NAVS 115 Figure 5.7 Survey Grid Sizes MAP LEGEND A Datum Point (OS, OE) g Grass Field E Fort Ross Cove Road 2 Cliff with Rock Outcroppings [] Beach Sand and Rocks 0 1o 20 30 40 m 0 2 4cmn CuafmDeipby EdwidC Mad A B C D E S - Wb - TUInJF - @-AUg - S s O S - 20 x 20x20m 20x20m ios- 7 2 - j / 20 1I >- - -- + - -2 7 20 S t>--*-,lL-jv | < 40S- 3: z SL 20 20x20m 20x20m 20x20m sS - 20 H--- ~~~~~~~~~~~~~~~~~605- 4 4 20 I, 20 4 7x2C2m|2x 20x20m 20x20m 3 20 v e _ w 7 ff - V 's-viQ. 80S - s s V_l - 6 20x 20 x 20 m 20x20m - 9 1 20 V.' v 20~~~~~~~~~~~~~~~~~2 2 0 20x20m 20x20m J. : : s 20 '_ _ 1___ 1__ -J ~~120 S - 20 ' 0 vV V .V 60 W sow 40W w 20W ow O E VE 20E 1 40 E I I I I I . I v I 116 The Native Alaskan Neighborhood to the east is subject to strong erosional forces, many of the outer traverses had to be done from rather uncomfort- able positions-sometimes actually hanging off the steep cliff wall. This particular area represented the most arduous and laborious part of the data collection process, and our sampling strategy was swiftly determined by focusing on the hardest work first. The developed approach allowed grids on flatter terrain to be surveyed towards the end of the day. The cliff section was an obvious first choice, since we had two comparative resistance data sets from the earlier work at NAVS. This collection plan allowed us to collect the first day's worth of readings and, based upon that night's analysis, we could have adjusted and refined our research strategy if necessary after comparing the data to our earlier findings. The sampling interval was changed from the 1992 surveys, where 1600 recordings were taken in each 20 x 20 m grid, to the new approach which called for a more efficient 800 values to be collected for each of these 400- square-meter areas. The main problem was an obvious loss of resolution due to the smaller data set density, but the dramatic results obtained from the first day negated any such worry. Some areas did not yield complete 20 x 20 m grids containing the total number of actual collected readings as depicted in figure 5.7. The so-called "Dummy Log" entry composed of an extremely high value (2758) was needed to complement some of the eastemmost cliff grids in places where no further survey results could be obtained. The widths (E-W) were supplemented on three of these units (B8, C7, D5) and, in one instance, even the partial length (N-S) of a grid (C7) was substituted using these off-scale entries. In the western grid units, time constraints resulted in two grids where the first 13 m (E-W) were operator-entered values (AS, B2) as part of an effort to collect only the essential minimum extension of the road. All other parts of the expanded 16,000-square-meter site area were also treated as dummy entries in the final production of the electronic maps. In all, Lisa Holm and I collected 20,800 readings in three days which amounted to nearly nine grids a day. RESULTS Undoubtedly, any analytical results from this survey can only be evaluated through archaeological excavation. The resistivity study serves its main purpose by directing attention to specific anomalies present in the ground. Therefore, the geophysical data collected becomes an additional means to develop, support, or adjust a research design and field strategy. At the same time, it allows one to make inferences regarding the overall spatial organiza- tion of the Fort Ross coastal terrace, including the Village to the extent that it was sampled. The following analysis is based on the electronic resistivity maps of NAVS generated using Geoplot version 2.0. Aside from the main composite image combining all forty 20 x 20 m units into a single site display, the NAVS data was also divided into five segments encompassing 20 grids each, 100 m in an east- west direction (width) and 80 m in a north-south align- ment (length). These mini-site composites or restricted- view maps are each advanced from their previous section by 20 m along the longitudinal axis. The reduced sectional displays better exhibited high and low end values, fluctuating less and resulting in often improved overall feature definition and resolution. All the de- scribed data sets were then processed using the same criteria for each. Table 5.1 explains the post-collection digital enhancements and correctional utilities that were applied using Geoplot 2.0. In conclusion, all the electronically available information was evaluated in tandem with earlier archaeological investigations (i.e., subsurface and surface collection data, excavation reports, etc.) to the fullest extent available. Most post-excavational analysis of the Table 5.1 Geoplot Processing of the Native Alaskan Village Site Dataset Geoplot File Extensions -(None) File Description Raw Data: Original collection readings Processing No processing Despiked: Processing method to repair erroneous and extreme data samples (high/low) Low Pass Filter: Processing method to visually enhance and emphasize the gradual changes represented in the data collection High Pass Filter: Processing method to visually enhance and emphasize the rapid changes represented in the data collection Interpolated: Introduction of additional data points between readings in order to artificially increase the overall number of samples thus improving the resolution of the image X-Radius = 2, Y-Radius = 1, Thresh- old = 2, Spike Replacement = SD (Standard Deviation) X-Radius = 2, Y= Radius = 1, Weighting = Gaussian X-Radius = 10, Y-Radius = 5, Weighting = Gaussian Direction = Y, Scaling Factor = 2, Expand -K -L -H -I . Resistivity Survey at NAVS 117 NAVS material, however, is concufrent with the writing of this report and, due to the massive resource base generated by four field seasons, some restrictions applied when trying to adhere to the scope desired. Future projects include computer-aided graphical overlays of the NAVS geophysical survey and historical maps using GIS technology, including visually represented excavation data generated in CorelDraw 3.0. Similarly, any further accessible ethnographic and historical evidence (e.g., texts, photographs) will be used to determine the nature and origin of the recorded resistivity anomalies. A NORTH, A MIDDLE, AND A SOurH SECrION: A JUSTIABLE DELINEATION? When examining the general overall patterns evident in the soil resistivity maps of NAVS, it soon becomes clear that there is a striking separation between the northern grid areas and their southern counterparts. This delineation is largely the result of the data in the southern and northem grids exhibiting rapid changes or strong anomalies, whereas a substantial middle section of the survey site displays a mostly uniform and tight value range of collected readings. The reason for such clear boundaries is not deternined. Without the archaeologist to refute or support this boundary separation into three distinct "zones", the data will remain at best a series of strong interpretational arguments for the presence or absence of occupational taes. In particular, the detailed information retrievable from artifactual evidence is crucial in further substantiat- ing any broad site delineation into a north, a middle, and a south area. (See chapter 3.) GENFJRAL OBECT-SPECIFIC DATA INTERPRETATION The road is probably the most dominant anomaly of the entire survey and serves its desired purpose in orienting the map reader. At the same time, it is a vivid testimony to the extent of human landscape modification. This subsurface record of a presently existing utilitarian feature proves the effectiveness of geophysical surveys as a valuable tool to be applied in archaeology. Not only did the road itself register well, but the extensive histori- cal disturbance to each side of it clearly delineates this feature. The quality of this distinction was an added indicator that we would be able to obtain good resistivity results. Another general anomaly visible throughout large stretches of the site included a fence scar running along the eastern cliff terrace, pictured clearly in an aerial photograph taken in 1964 just before the site was cleaned up for reconstruction (Tschan 1994: 44, figure 24). This feature was among the strongest detected by the magne- tometer, and it also appears dominantly in the resistivity maps. The discovery of numerous wood posts and their fragments in the East Central Trench, as well as some barbed wire pieces in the same location, correspond very well with the resistance readings depicting a linear arrangement through the easternmost extension of this excavaton area. THE 1817 MAP VERSUS THE PRESENT CENTER OF THE SITE By correlating the survey data with historical references, some interesting anomalies became apparent For example, the earliest documented NAVS location based on a Russian map produced in 1817 for the Spanish colonial government (figure 5.8) overlaps to a large degree with an area of very low resistance readings. In the electronic map, this part of the site is of uniform context, and faunal remains in richer (darker) midden soils were evident along the terrace edge. The "empty" area where the earliest map of Fort Ross puts NAVS may be explained in several ways: 1. The relative precision of the 1817 map in regard to the Russian structures can not by default be applied to the native residences surrounding the palisade. The artist may not have deemed them critically important for the Spanish request. 2. There was a deliberate attempt to deceive the Spanish authorities by limiting the exact spatial extent of the Russian colony. 3. The instrumentation used for mapmaking did not allow the degree of accuracy that can be obtained in surveys today. 4. There were shifts in the actual location over time; thus, the original site was later abandoned, and the new buildings were located elsewhere on the cliff. 5. A lack of consideration of erosional forces may distort the view presented. In other words, the location of the Village midway between the Fort and the cliff as depicted in the 1817 map would place the Village further west and north of the midway point in the 1990s, after 150 years of cliff erosion. It also appears that the Russians generally had little regard for the cultural traditions of their fur-hunting laborers, as is evident in their severely limited accounts of the Unangas and Alutiiq peoples at Fort Ross. This might be an important factor. If their survey instruments really lacked the capacity to accurately and effectively register vertical, horizontal, and especially distance dimensions, it would seem equally impossible for the Russian authority to have engaged in construction work, successful navigation of the world's oceans, or any meaningful cartographic production. On the other hand, the argument that coastal erosion contributed to the disappearance of large tracts of the earliest Village may be partially valid, but basically the area designated in the 1817 map as belonging to NAVS is still readily accessible (figure 5.9). Regardless of the above and other possible scenarios, 118 The Native Alaskan Neighborhood irrefutable evidence (ethnogriphic and historical refer- ences [chapter 1]) indicates the presence of a Village site within this area. The residents must, therefore, have left visible scars in the landscape by erecting structures which would have left the area in a greatly disturbed condiion. Consequently, it is likely that an early NAVS may have been located somewhere in the general area of the cliff, but without direct correspondence to the first available map evidence. Using this probable conclusion allows one to engage in a less restricted search for NAVS by solely interpreting the resistivity results. Unfortu- nately, the few traces discemible in this middle area of the site are very linear and seem to resemble more the layout of corrals used during the later American Period when intensified animal husbandry became popular (figure 5.10). It therefore seems unlikely that a large- scale settlement could still underlie the extensive "empty" area of the middle section. OBJECT-SPECIFIC DATA INTERPRETATION (MIDDLE SECTION) The middle section, as discussed above, is rather limited in historical resources that portray past occupa- tional site layouts. The resistivity survey did however retrieve some long, linear features as well as one or two small squares containing further internal division (B, figure 5.10). The most dominant "line" in this estimated NAVS section is a long transect (A, figure 5.10) coming across from the westem section of the road into the eastern part where it connects to a possible pathway or fence scar that runs mostly parallel to the previously mentioned site-transcending fence. One interpretation may also conclude that there is a rectangular spatial definition within this middle section suggesting that the pathway actually originates here. A number of small, dispersed anomalies which seemed to have some interior definition were detected. One is found at the intersection of the suggested pathway and an odd-shaped traverse. This particular feature exhibits circular characteristics (C, figure 5.10) while another intemally defined object (D, figure 5.10) shows up well within the enclosed space formed by the remain- ing dominant spatial delineation (interpreted as fences). Interestingly, the former seems to be connected to the traverse anomaly (E, figure 5.10) that mostly runs in a northeastern direction and eventually seems to lead to the "entrance" of a huge northern barn-like structure. Similarly, at least one of these square objects-a Figure 5.8 1817 Russian Map of Fort Ross Russian Stockade Compound Native Alaskan Village Site Resistivity Survey at NA VS 119 Figure 5.9 Site Reconstruction of Native Alaskan Village Site Based on the 1817 Map MAP LEGEND A Datum Point (OS, OE) Grass Field g Fort Ross Cove Road 2] Cliff with Rock Outcroppings 3 Beach Sand and Rocks NAVS According to 1817 Map Excavation Units Trench Reference Points Represent Extension Corners East Central Trench Area NW=71S,2W NE=71S,5E SW = 75S, 2W SE=75S, 5E South Trench Area NW=119S,26W NE=119S,17W SW- 126S, 26W SE = 126S, 17W O 10 20 30 40 m 0 2 4cm Os - 1os - 20S - 2 k itL --- -- ,. -- t / 30S - I I~~~~~~~~~~~~~~~~~~~~s Lt I- .L.- .. t 40S - 3 1 * ^ S -- - 50S- .AL AL. 7 --- -v, '_v 1.Li.Lz~~~~~~ V V _._ VV 7S 7 130S -~~~~~~~~~~~77:: ~~~~~~ +~~~~~~~7 / - - - -z -' 16+ 80 S - I0 v ow 40 30 20 -o OE ..E 2.. ..E .. .4v.0E '" c I I s I .. .. . . 706~~~~~~~~~~~~~~~~~~~0 s 15 7 / 7 v v V-- - 2 . -. 130 S- y v:"::v ':'Pv:"vv' '1 - ' 1 ' : 8 ,; Z , , :4 15 1sS _^ ,_ __-l_ _ _ i- -- - J 160 S- 60 W sow 40W 30W 20W iOW OE iOE 20 E 30 E 40 E I I - -1 I I I I I I I Motfied t 0 C odr Cs t7 Edward C. Mack . . . . . . . 120 The Native Alaskan Neighborhood MAP LEGEN; A& Datum Pc Kj Grass Fie , Fort Ross 2] Cliff wIt [jBeach Sa Excavation Ur Trench Refere Represent Ext East Central I Figure 5.10 Middle Section Map Indicating the Dominant Resistivity Survey Features ID A B C D E pint (OS, QE) 1 -Id s Cove Road 2 'Rock Outcroppings L. rd and Rocks 3 i1its 4 nice Points 4u LA ension Corners rrench Area 5 -'a-1ZS S S S | g | | _ F - ~ - V\o NW=71S,2W NE=71S,5E SW= 75S, 2W SE = 75S, 5E South Trench Area NW = 119S, 26W NE = 119S, 17W SW = 126S, 26W SE = 126S, 17W 0 10 20 30 40 m 0 2 4 cm 6 1 v V 7 I I~~ _ i I.. ': ~~~~~~~..V. s. ..v t - i _ i v l/.,-.jz b tv . 7O 5o 4O 3w O iow OE io 20C O s- los - 20S - 30S - 40-S 50 S - 60 S - .v 70 S - -d 80 S- * I 90 s - *1. I: It/A I - lOO -jIu- 10 . iii 12 ! 130 S A140 S 1505s _. - - - 160 S- 3 0E 3 40E I I North Section [Z Middle Section I South Section Modified from Origa1 Compass Design by Edward C. Mack -7 Resistivity Survey at NAVS 121 plausible interpretation identifies these as troughs- appears on the electronic maps in what I consider to be the south section, possibly indicating shared feature occurrences while generally maintaining the site's three- part separation scheme. Alternatively, the actual occupa- tional chronology for Fort Ross supports different scenarios for the use life of the three main areas. Future archaeological investigation may provide clues to distinguish more clearly any such multi-cultural and long-term activity boundaries. MULTI-USE DISORDER IN THE NORTH The anticipated delineation of the site into three main sectors is further reinforced by the distinct differences between the northem and southem grids. The former are riddled by extremely rapid changes, while the latter exhibit much more gradual, though equally numerous, resistance anomalies. This striking and heavy distur- bance in the area immediately in front of the Stockade also shows up in the later historical maps that have been assembled as part of the continuing Fort Ross Archaeo- logical Project. None show any buildings further from the southeastern tower than 50 to 60 m (figure 5.11). Photographs of the same area also depict a series of classic American barns (Tschan 1994:40; figure 21). A slaughterhouse, a butcher shop, a post office, a gas station, and a convenience store (Murley, personal communication 1993) all were once present within the same area and, in most instances, they used the same two foundational layouts and superstructural designs, though chronologically most are distinct. The only marked exception is the transition of an eastern wooden barn from a definite square base, which persisted at least until the 1906 earthquake (Tschan 1994:40; figure 22), to a later north-south aligned rectangular structure. This northem section, with its close proximity to the Fort, is also the most puzzling due to its lengthy use-life. Until 1964, when the entire cliff terrace south of the compound's walls was cleared, there were remnants (heavily decayed) of buildings and fences (Tschan 1994:40; figure 23). When looking at the overall NAVS resistance readings, one is immediately drawn to a roughly square 30 x 30 m feature (B, figure 5.12) that is in stark contrast to the shell midden bordering it immediately to the south. It becomes particularly visible using Geoplot color graphs. Since the midden and feature are adjacent areas, the data from either one would suffer from resolution and detail loss when overemphasizing the opposite value range. Further investigation, especially chemical soil analysis, will have to suggest a cause for the distinctive boundary between the two. Although the Geoplot Relief utility treats the data with much greater detail by representing the survey results as surface features, these high reading patches gathering methods. Nonetheless, depending on the Scaling Factor, the Sun Direction Degree, or the Sun Elevation Degree in Geoplot, it is possible to distinguish other structural features. For example, there are two long, large linear "pipelines" or "electric cables" which may represent Pacific Gas and Electric power lines. These seem to be connected to what may be the residual shadow of the former gas station's fuel tanks (presumably removed) (A, figure 5.12). Other features with high resistance values include the square feature mentioned above. Within it, one may discern a "four-room structure with a center hearth." Coinciding with these dramatic top end values is a surface area that is almost barren. This is in marked contrast to the surrounding area which needs repeated mowing throughout the year. Perhaps the safest interpretation of the circular features calls for either a natural outcropping of the local rock or the filling in of an area with high resistance materials after more fertile matter had been removed, as in the case of a dump. This should not completely discredit the possibility of archaeologically significant architectural structures being present, but when extremes in resistance values are combined with a lengthy use period for the site and heavy disturbance, it is better to engage in test excavation before claiming fanciful discoveries. These strong and extensive anomalies are among the most intriguing. Particularly rewarding might be the geophysical investigation into the potential effects of penned-up herd animals (i.e., uric acid generation) or other waste products of a dairy and meat operation, in conjunction with an open dump site. The question to be answered for geophysical testing relates to why former animal compounds or butcher areas would show up as high resistivity features, for most of the organic waste products should assist, rather than hinder, in the ground's conductivity, as in the site's middle section. Regardless of the factors contributing to the lack of vegetation in the proximity of the eastern barn (slaughterhouse), two corrals were located there, and these structures overlap or make up to a large extent the areas with the top range resistance values (B, figure 5.12). This north sector also has geometric structural features that are remarkable, since some do not corre- spond very well with the historical evidence of buildings situated close to the Fort within the last 100 years. As a final point, the dirt track road is surely nothing more than a modern construct. It therefore must be considered an arbitrary divider in relation to the archaeology of the Russian Period, with their own coastal pathway lying further east. Rapid resistivity changes clearly continue, particularly to the southwest of the Fort Ross Cove Road. OBJEcT-SPECIFIC DATA INTERPRETATION (NORTH SECTION) Although the most obvious and dominant anomalies could not be further interpreted using current information have been highlighted above, the massive rectangular and 122 The Native Alaskan Neighborhood Figure 5.11 611164 Department of Water Resources Map of Fort Ross Copied at the Department of Parks and Recreations, Sacramento, CA. Scale unknown. square outlines of what must be interpreted as barn-like structures caused some considerable debate (C, D, figure 5.12). None of the historical maps show a north-south oriented elongated building in the immediate vicinity of the road. This peculiar feature seems to be further divided into three segments, with the middle being the largest and showing an extensive opening in the center of its eastern boundary. Also, a major linear object originat- ing all the way from the middle section of the site and best described as a pathway leads up to this possible gate or doorway. The smaller, yet still impressively sized, "rooms" to each side (north and south) of the middle interior space each contain a large circular feature in their centers. All in all, this structural complex may encom- pass an area of up to 250 square meters measuring 25 m in length and 10 m in width. Partially overlaying this huge spatial anomaly, which seems to end in a semicircle at its north end, is another possible foundation layout. This one is fairly square and includes less distinguishable interior detail because of its positional placement over most of the top resistance "mystery square." This second likely architectural remnant also refuses to line up directly with historical maps, yet there is close resemblance to a square building described in the Veasey map (figure 5.13). Its proximity would make it the "butcher shop" rather than the "store." There is a lack of documented evidence for either of these two well-defined and enclosed spaces. Although Resistivity Survey at NAVS Figure 5.12 North Section Map Indicating the Dominant Resistivity Survey Features MAP LEGEND A Datum Point (OS, OE) ] Grass Field E Fort Ross Cove Road ] Cliff with Rock Outcroppings [ Beach Sand and Rocks Excavation Units Trench Reference Points Represent Extension Comers East Central Trench Area NW=71S,2W NE=71S,5E SW = 75S, 2W SE= 75S, 5E South Trench Area NW= 119S, 26W NE = 119S, 17W SW = 126S, 26W SE = 126S, 17W O 1o 20 30 40 m 0 2 4cm A B I * j1 j A IL _ L LLL .wo aLLu 4 j L1J ,.xuw C D E Os - 1os - 20S - 30S - 40S - 50S - 60S - 70S - soS - jt_~~~~~~~~ H - - -'ffl2Oi 1S - ................. - ..... t l I ,'_,~~~~V''. v' v -V ~ ~ 155 0 ;:w 4W3W 0Wiw OE lO 20E -.E 40Eh .I Z a I _ . . -- ' V, II, 7 F t- ;/'-'.130 S 8 ISO 1Nsll_ II ,:- 1,-- li I I 'o t I J' I ? I I I I I I D- North Section [I] Middle Section Modified from OrigaI South Section Compas Design by Edwad C. Mack I I I I I 123 124 The Native Alaskan Neighborhood Figure 5.13 1892 Veasey Map of Fort Ross Musset Nl Copied at the Department of Parks and Recreation, Sacramento, CA. Scale unknown. they probably reflect European and American building traditions, one must place these structural remnants as early as the Russian occupation, yet no later than the 1906 cartographic evidence. The previously mentioned photographic data shows a change in the alignment of the "butcher shop" from its original, mostly square layout to an undoubtedly rectangular and north-south oriented design built after or right around 1906. WAs NAVS FURTHER TO THE SourHwEsr? Without a doubt, the most insightful data was collected from the southern grids, and it becomes difficult to discuss general trends for an area that produced such varied results. A clear overall observation, however, is that most of the structured data is grouped into what looks like an outer layer of distinct higher readings extending broadly across the road in a westward direc- tion. This information could possibly resolve the question regarding the spatial ethnographic descriptions for the sizable interethnic Alaskan households on the cliff terrace (Lightfoot, Wake, Schiff 1991:23). Hence, a series of clear linear and circular features that registered prominendy in the survey fall within a restricted area which could indicate the presence of a closely knit community (A, figure 5.14). Unfortunately, only the area to the east of the road has been investigated, either archaeologically or by the soil resistivity survey. Be- cause the analysis has concentrated on the original predicted location of the NAVS, future research should include the areas lying to the west of the beach road in an attempt to validate the geophysical results. The southernmost units recorded (160S, 60W and 160S, 40W) have an equally massive cluster of high resistivity results (B, figure 5.14) much like the north section's "mystery square." Yet, unlike its northern counterpart, this area also contains a clearly exposed and seemingly "stacked" rock pile on the surface, situated within the space immediately south and outside of these end grids and the road's sharp hairpin trn. By virtue of its proximity to what might well be a deliberate assem- blage, the aforementioned buried resistivity feature could, by direct association, be interpreted as evidence of some Figure 5.14 South Section Map Indicating the Dominant Resistivity Features MAP LEGEND A B C D A Datum Point (OS, OE) 1 Grass Field Fort Ross Cove Road 2 - - Cliff with Rock Outcroppmgs -- - Beach Sand and Rocks 3' Excavation Units Trench Reference Points Represent Extension Comers East Central Trench Area NW=71S,2W NE=71S,SE SW=75S,2W SE=75S,5E South Trench Area NW= 119S,26W NE= 119S, 17W SW = 126S, 26W SE = 126S, 17W O 1o 20 30 40m 0 2 4crn 4i. -- svl 30 Xl aL j- 1 -6 'V vvt 6 f't t ) B o .5..v : .::-.-: IIOS - 6< v ' +q120S 7 , 7 , ; o' ib ,'V. V. 130 S y \1l ; v v -v+-r - - 6-- -.H ~~~~~~140S- 8 ISO. s g ,.8 V s C , l {! ' V2 1 _ _ _ _ 1 _ _-- ~~~~~~~~160S- 60W 50W 40W 30W 20W iow OE IOE 20E 30E 40E L I I I . 1 1 _ i I I M & E North Section E Middle Section Modified frxm Orignial ] South Section Compass Design by Edward C. Mack Resistivity Survey at NAVS 125 Os - 1os - 20 - 30S - 40S- 50 S - 60S - . . . . . . . . 126 The Native Alaskan Neighborhood utilitarian formation process. ' In order to establish any direct correlation with this subsurface anomaly, the exact nature of the visible assemblage outside the survey area must be determined. This large rock pile, which marks the 180 degree turn in the road, may have been placed deliberately as part of modem use of the site, or it may merely represent natural, geological processes. If their placement is the result of human activity, then the origin of the boulders is impor- tant, since deliberate deposition generates suspicion about geological explanations for the geophysical anomalies found in the surrounding survey area. Heavy subsurface disturbances could be the consequence of raw material procurement and its transport. A fairly well- structured and spider-shaped high resistivity feature clearly dominates the two southemmost grids. Were these features a solely geological contribution to the data, one normally would assume greater irregularity in design. During the 1992 field season, the discovery of rock rubble in the South Trench and a similar drainage system that assisted construction inside the Fort (Farris 1981:17) may have constituted a unique resistivity signature across the area. This human introduction of fist-sized stone fragments never exceeded in depth the maximum subsurface penetration of one half meter (0.5 m) reached by the current of the RM15. Unfortunately, the 1992 geophysical survey which detected the anomalies in the excavation units prior to the actual foundations being unearthed has the stigma of severe unreliability in terms of data interpretation. The subsequent resistivity testing conducted in 1993 could therefore only register, albeit vigorously, the plastic-lined and stream-gravel-filled trenches due to the post-excavational clean-up procedures on the site. The resulting high readings also persisted in the immediate vicinity of the excavation units-the consequence of filler material spilled around the edges. At this point, the principal data set will serve as a reference to which all upcoming excavational discoveries can be correlated. If these foundations are spatially restricted structures, specific to the dimensions of single dwellings, then further households may be located by focusing on similar anomalies in the electronic maps. OBJECT-SPECIFIC DATA INTERPRETATION (SouTH SECTION) There are enough anomalies discernible in this last section to hint at a one-time only occupation of the cliff terrace by fur-hunting laborers. As the most potentially insightful research target, it provides spatial clues about what may have been the Native Alaskan Village Site. The obvious structural features that dominate the area are two very long parallel lines reaching a maximum extension of perhaps 30 m aligned in a northwest- southeast direction (C, figure 5.14). Four round "corner posts" and two additional dividers perpendicular to these baselines partition the entire feature into three equal segments. In addition, the whole center seems to be composed of a large circular depression. All these components together form a slim, elongated barn-like structure (plank house) possibly extending across the road in a westward direction. Grouped around this European or American design pattern are at least three, perhaps four, circular anomalies with a diameter of about 5 m. The furthest westward extension of the South Area Excavation seems to have penetrated the outer perimeter of one of these structures. This coincides with a stone "foundation" and the remains of a pit excavated at this exact location. If the electronic data and the actual archaeological evidence retrieved support each other, then these features may indicate the presence of a traditional Alaskan semi-subterranean structure. At the least, the deliberate layout of these archaeological remains could be used to identify any additional, similar structures within the resistivity map. Lastly, the spider-like anomaly, some 10 to 15 m in diameter at the southernmost tip of the survey area also seems to be connected to what might be a pathway coming from across the road. This could indicate the presence of a now long-gone residential or utilitarian object (B, figure 5.14). It exhibits a clear central area with "legs" extending around it in all directions. As mentioned before, whether or not a natural rock forma- tion produced these high resistance readings remains to be seen through future investigations. Any geometric pattern in a geophysical survey, however, should be treated with suspicion and, in this particular instance, it may well be a pile of deliberately deposited rocks assembled for a currently unknown purpose. ADDITIONAL RELATED RESULTS Additional information was obtained in regard to how weather affects the data collected in resistivity surveys. Even without a controlled measurement scheme lasting for at least a year with readings collected in monthly intervals, the resistivity set obtained during the wet December of 1993 starkly differs from its counterpart recorded during the hot June of the previous year. Consequently, a survey scheme on a Native Californian site undertaken in February and May of 1993 was geared at specifically investigating climactic influences for remote sensing along this region's system of ridge top sites in an attempt to determine the range of productive months versus the less rewarding calendric choices for a resistivity survey. Within this short time period, the weather pattern generally shifted from very wet to less frequent but persistent precipitation. As a result, there was a considerable difference in the data sets, and the closer it got to summer, the fainter the features registered in the resistivity maps. A study into the Californian north coast's geophysical potential has yet to achieve the nccessary scientific recognition by use of a rigorous Resistivity Survey at NAVS 127 control system, however. FCiture research into this area could fill a dire need for remote sensing parameters to govern this otherwise geophysically neglected, but archaeologically most intriguing, region of the Pacific Rim. CONCLUSION The spatial distribution of the Village established by the Native Alaskans was the primary focus of the resistivity survey at this site. The particular research Figure 5.15 Entire Site Map Indicatiy A MAP LEGEND A A DatumPoint (OS,OE) 1 j Grass Field - - _ 7 Fort Ross Cove Road 2 K - 6 Cliff with Rock Outcroppimgs K-,- - E] Beach Sand and Rocks 3 L NAVS According to 18 1 7 Map H ,, - - Excavation Units 4 Trench Reference Points Represent Extension Corners _ East Central Trench Area 5 pu J NW=71S,2W NE=71S,5E -; SW = 75S, 2W SE = 75S, 5E 6 South Trench Area 7 NW=119S,26W NE=119S,17W SW= 126S, 26W SE= 126S, 17W 8 C 0 lo 20 30 40 M m 60 6W 50ow 0 2 4 CM I I objective consisted of a search for distinct boundary lines and architectural demarcations. Interpretation of the collected remote sensing data is a puzzling, challenging, and complicated affair (figure 5.15). The vast amount of information that has been obtained at NAVS, however, may answer some of the most pressing questions regard- ing the spatial layout and organization of the site. As such, this resistivity survey has allowed the separation of NAVS into three characteristically distinct areas: 1. North = The most actively used area in the past ng the Dominant Resistivity Survey Features os 10 S 20S - 30 S - 40S - 50 S- 60S - 70 S - 8SO - 90S - I@S - 110 S- 120S - 130 S- 140S - 150 S- 160S - Modified from Original Compass Design by Edward C. Mack I - . 128 The Native Alaskan Neighborhood as well as the most complex area of the resistivity survey. Modem features may dominate this section, yet Russian or Alaskan materials likely lay adjacent to or beneath these later occupational remains; 2. Middle = An "open" tract of land devoid of extensive resistivity features, aside from occasional fence scars. Nonetheless, there may be a chance of earlier materials underlying this area, including settlement remains 3. South = This section is the most likely candidate for an occupational site established by Native Alaskan hunters due to the numerous structural pattems noted in the resistivity data. Unfortunately, the vast quantity of data presented in the Geoplot maps of the Fort Ross cliff terrace make it virtually impossible to address additional, often merely pixel-sized, anomalies. My aim, therefore, was to provide some analytical conclusions regarding the major resistivity features and to avoid undue geophysical analysis and hedging. ACKNOWLEDGEMENTS Throughout this work, the assistance of the principal Fort Ross research staff, including Professor Lightfoot, Staff Archaeologist Ann Schiff, and graduate students Nette Martinez, Peter Mills, and Tom Wake, proved instrumental. Their interpretational input when examin- ing the electronic maps was essential to the success of this project. Furthermore, the staff of the California Department of State Parks and Recreation, including Glenn Farris, Dan Murley, and Bill Walton, were of great help in providing written, photographic, and personal - documentation. Nick Jones was responsible for capturing the available maps of Fort Ross on film which then were used in subsequent, computer-related image processing. As a professional photographer he worked miracles with the often dismal cartographic documents, and it was thanks to his talents that such a high standard was achieved. David Wheatley, lecturer at the University of Southampton, and Johan Wikman at Autodesk also assisted in some tricky computer aspects of graphical software programs and output interfaces. Both are appreciatively mentioned as helpful contributors. My research partners John Anderson and Lisa Holm, how- ever, undoubtedly deserve most of the credit, for without their personal sacrifices and their unconditional enthusi- asm, this work could not have been completed success- fully. Any complex computer-related problems were resolved mainly by my wife Colleen. She was ingenious at arriving at solutions to software and hardware difficul- ties. Her patience kept me sane during the most hectic of times; as a companion and helpmate, she was and continues to be my most invaluable asset. Finally, the individuals without whom this project would never have become reality should be mentioned: Dr. Lewis Somers of Geoscan Research USA-our educational source for remote sensing techniques, instrument handling, and data analysis, as well as the generous provider of the neces- sary equipment-has been the main guiding force behind all of the work undertaken. Dr. Somers and his wife Elizabeth have tolerated the repeated and lengthy invasion of their private residence by myself and my partners. The unconditional hospitality, magnificent knowledge base, and extensive training they bestowed have been appreciated on a personal and professional level by all parties far beyond what these few lines may credit. Therefore my special thanks go to Geoscan Research USA and all its affiliates. REFERENCES Clark, Anthony 1990 Seeing Beneath the Soil-Prospecting Methods in Archaeology. B. T. Batsford Ltd., London. Farris, Glenn J. 1981 Preliminary Report of the 1981 Excavations of the Fort Ross Fur Warehouse. Cultural Resources Manage- ment Unit, Resource Protection Division. California Departnent of Parks and Recreation, Sacramento. Halliday, David, Robert Resnick, and Jearl Walker 1993 Fundamentals of Physics - Fourth Edition. John Wiley & Sons, Inc. New York. Lightfoot, Kent G., Thomas A. Wake, and Ann M. Schiff 1991 The Archaeology and Ethnohistory of Fort Ross, California: Vol.1, Introduction. Contributions of the University of California Archaeological Research Facility no. 49, Berkeley. Sears, Francis W., Hugh D. Young, and Mark W. Zemansky 1983 University Physics Part II. Addison-Wesley Publishing Co., Chicago. Tschan, Andre 1994 Sensing the Past and the Remoteness of the Future. Unpublished honor's thesis. Copy at the Anthropology Library, University of Califomia, Berkeley. Weymouth, John 1986 Geophysical Methods of Archaeological Site Surveying. In Advances in Archaeological Method and Theory, vol. 9. Academic Press, Inc., New York. Personal interviews were also conducted throughout this project. The main contributors included: Dan Murley, Park Ranger, Fort Ross, California; Dr. Lewis Somers, Geoscan Research USA, Sea Ranch, California; and Bill Walton, Park Ranger, Fort Ross, Califomia.