Avifaunal extinctions, vegetation change, and Polynesian impacts in prehistoric Hawai`i.
Pre-contact avifaunal extinctions in Hawai`i generally have been attributed to human predation and/or landscape alteration by colonizing Polynesians. However, until recently there have been insufficient data for evaluating most of the important variables involved in this issue. This situation has changed with recent archaeological, paleontological, and wetland coring research conducted on O`ahu's `Ewa Plain, a hot, dry emerged limestone reef characterized by numerous sinkholes. The main evidence obtained from this research includes (1) wetland coring data that stratigraphically demonstrate forest decline before any burning, (2) radiocarbon dating of bones of rats and extinct birds that provides a time frame for their occurrence unavailable from stratified deposits, and (3) the radiocarbon-based history of human settlement of the `Ewa Plain.
Based on this evidence the argument is made that (1) at least some major avian extinctions occurred within the period immediately following Polynesian colonization, (2) these extinctions were due primarily to the rapid decline of their native lowland forest habitat, (3) human settlement of the `Ewa Plain occurred after native forest collapse, not coincident with it, and (4) the main source of destruction of the native forests was the introduced Polynesian rat, Rattus exulans, not Hawaiian agricultural clearing and burning. This model also explains the absence of large quantities of bird bone in early sites (in contrast to other places in Polynesia and Micronesia), and the absence in early middens of many plants (notably Kanaloa kahoolawensis) that were common in the native forest.
The cause of Holocene avifaunal extinctions in Hawai`i has been a topic of intense interest since the first detailed investigation of subfossil paleontological remains almost two decades ago demonstrated that Hawai`i once had a much richer species inventory of birds than previously suspected (Olson and James, 1982, 1991; James and Olson, 1991). Recent interdisciplinary investigations on the `Ewa Plain of southwest O`ahu (Fig. 1) provide new insights into the process of extinction and environmental change in the Hawaiian lowlands (Athens et al., 1999). The new data 1) document the unexpected rapidity of vegetation change, 2) provide detailed information on the chronology of human settlement on the `Ewa Plain, and 3) provide detailed paleontological evidence, particularly as concerns the chronology of avian remains and vegetation change. These data indicate that avian extinctions and extirpations were the result of habitat destruction in the form of lowland forest decline. However, there is strong evidence that forest decline followed Polynesian colonization of Hawai`i but preceded dispersal, suggesting that direct human activity such as burning or gardening had nothing to do with the general decline. We propose that the main cause of forest decline was the Polynesian rat, Rattus exulans, brought by early colonists.
[FIGURE 1 OMITTED]
Problem of avifaunal extinctions and extirpations
Some 35 to 57 extinct avian species have been identified from subfossil remains in Hawai`i, the number depending upon classification (James, 1995). Considering that historically known avifauna numbered from 40 to 55 endemic species, then about 100 avian species were present in Hawai`i prior to the archipelago's settlement by Polynesians (James and Olson, 1991), which occurred between about AD 700-800. (1) Chronological data from a stratified assemblage of avian subfossils on Maui (James et al., 1987) indicate a strong association of a major extinction event with the period of Polynesian settlement. Other studies of subfossil birds in the tropical Pacific follow a similar pattern, indicating that long periods passed with few extinctions despite major climatic oscillations with the last ice age (Steadman, 1995). However, with the spread of humans throughout tropical Oceania beyond Melanesia during the late Holocene, wholesale avifaunal extinctions and extirpations frequently occurred, and the decline to extinction is often evident in early archaeological middens (Dye and Steadman, 1990; Steadman, 1995).
For Hawai`i, habitat alteration by humans is regarded as the leading candidate for extinction of at least the smaller forest avian species, but other factors have been considered, including human predation, predation by the prehistorically introduced Polynesian rat (R. exulans), and introduced diseases (James and Olson, 1991). Research has provided little evidence to support any significant role in avian extinction by disease or by human or rat predation. Archaeological studies have shown there was major habitat alteration from human action, and that by AD 1600 eighty percent of lands below about 460 m elevation had been extensively changed as a result of agricultural development, with fire thought to have been the chief agent of destruction (Kirch, 1982a, 1982b). It had been reasonably argued that Hawaiian agricultural expansion was the primary source of native forest destruction (Kirch, 1982a:8 and 1985:291; Dye, 1994a:9). However, the `Ewa Plain research indicates that agricultural modification of the landscape (including the use of fire) occurred following an initial early post-settlement forest change. The significant alteration of the natural landscape immediately after Polynesian colonization has been demonstrated by a number of wetland coring studies on O`ahu during the last decade that provide detailed information on the lowland vegetation both before and after Polynesian arrival (Athens and Ward, 1993; Athens et al., 1992, Athens, 1997).
There is little doubt that the advent of human settlement in Hawai`i and the loss of the native forests are inextricably linked with avian extinctions and extirpations. The problem, however, is to better specify exactly what these linkages are. Our recent investigations on the `Ewa Plain provide data for a model of these linkages.
The `Ewa Plain: environment and human occupation
The arid `Ewa Plain is situated on an emerged limestone reef west of Pearl Harbor (Fig. 1--environment discussed in Athens et al., 1999; Davis, 1990; Tuggle, 1997). It is roughly 5 x 15 km (75 [km.sup.2]) and rises from sea level to about 15 m elevation. Thick primary and secondary alluvium/colluvium eroded from the Wai`anae Mountains has been deposited on the inland part of the plain, where commercial sugarcane was cultivated until recent years. The limestone exposed in the more shoreward locations is punctuated by numerous sinkholes. Annual rainfall is quite limited, averaging 508 mm and mostly falling during winter. Actual rainfall, however, frequently deviates significantly from monthly, seasonal, and annual averages; often most of what falls during the year occurs as a result of two or three downpours from winter cyclonic storms. Surface water is absent on the `Ewa Plain, though prior to the advent of sugarcane plantations in the latter part of the 19th century, some sink holes may have retained fresh water. For colonizing Polynesians, the `Ewa Plain must have been a marginal and unreliable agricultural environment, a point underscored by observations of the earliest western visitors (e.g., Vancouver [1798-II:217] called it "one very barren waste" when he saw it in 1792).
Modern vegetation on the Plain reflects its generally hot and dry conditions. Dominant post-contact introductions include Prosopis pallida (kiawe) and Leucaena leucocephala (koa haole) along with a variety of shrubs, grasses, and vines. Native species are represented only by dry-adapted Erythrina sandwicensis (wiliwili) trees (in certain areas these may be relatively common) and occasional isolated plants of Cordyline fruticosa (ti) and Morinda citrifolia (noni).
Numerous archaeological studies have been conducted on the `Ewa Plain during the last several decades (for review and synthesis see Tuggle, 1997; Tuggle and Tomonari-Tuggle, 1997a). As of 1995 most of the only 6 to 10 percent of the `Ewa Plain (4.5 to 7.5 sq. km) that was undisturbed by modern development had been archaeologically surveyed (for documentation see Tuggle & Tomonari-Tuggle, 1997a:45). Archaeological findings indicate that Hawaiian settlement consisted of small, scattered communities that relied heavily on fishing, but also carried out limited cultivation. Settlements were concentrated around limestone sinkhole clusters, and included C-shaped structures, thick-walled rectangular houses, small high platforms, low walls or alignments enclosing sinkholes, cobble mounds, and piles of fire-cracked limestone (Tuggle, 1997; Tuggle and Tomonari-Tuggle, 1997a).
Archaeological research on the `Ewa Plain has long involved a substantial paleoenvironemental component. The focus was initially on the collection and study of extinct and extirpated subfossil avian remains in sinkhole deposits (Sinoto, 1976, 1978; Olson and James, 1982; Davis, 1990), but subsequently included non-marine snails (Christensen and Kirch, 1986). More recently, however, paleoenvironmental studies have also included the coring of wetland deposits with associated palynological investigations and related studies. The following sections summarize the new data and analyses from our research (Athens et al., 1999). The findings indicate a rapid decline of the native lowland forest almost immediately following Polynesian colonization, the relatively late human settlement of the environmentally marginal `Ewa Plain, and the chronology of avian extinctions. The evidence suggests that both many land birds and the dry-land forest had disappeared by the time humans arrived on the `Ewa Plain.
Wetland cores recovered and analyzed from the `Ewa Plain (Athens et al., 1999) include two from an evaporite pan and one from a large sinkhole basin known as Ordy Pond. The evaporite pan is located about 500 m from the coast at an elevation of 4.6 m. Ordy Pond is about 700 m from the coast and has an elevation of about 1.5 m. It is 70 m across, with 5.5 m of standing, slightly brackish, water. Ordy Pond is just over 1 km east of the evaporite pan.
The two evaporite pan cores document a maximum of 3.76 m of mostly finely laminated sediments on top of limestone bedrock. Two radiocarbon determinations show that most of the stratified sediments encompass all but the earliest part of the Holocene, extending back to around 7200 cal BP. (2) However, there is an undated lower sedimentary unit in one of the cores. Because of the abundance of terrestrial snails in the sediments (presumably indicative of a time of lowered sea levels), this layer is thought to pertain to the Late Pleistocene. (3)
The single Ordy Pond core record, comprising a mostly laminated sediment column of 8.7 m, dates back from the present to about 1500 cal BP (AD 450); the sediments extend between 4.0 and 12.7 m below sea level (Fig. 2). (4) Because of its relatively high resolution record for the period of interest (i.e., the period immediately prior to Polynesian colonization through to the end of the prehistoric Polynesian period), analyses were primarily concentrated on the Ordy Pond core.
[FIGURE 2 OMITTED]
Although the organic laminations of Ordy Pond were not analyzed, similar laminations in the evaporite pan consisted of mats of two types of freshwater filamentous algae (possibly either Mougeotia or Spirogyra and Vaucheria). The inorganic carbonate laminations of Ordy Pond consist mostly of fine-grained calcite and/or aragonite of apparent authigenic origin (Tribble et al., 1999). It is believed that the organic laminae provide a record of major rainfall events prior to about 590 cal BP (AD 1360). Clusters of algal mats suggest a tendency during which either seasonally increased rainfall or isolated (but repeated) major rainfall events occurred over a period of several decades or more (allowing the algae to grow in the freshwater lens at the surface). These alternated with clusters of non-organic laminae, presumably indicative much drier spells of both longer and shorter duration (during which there was no freshwater lens, and saline conditions of the pond water became increasingly concentrated). Detailed analyses of these laminae and their implications for past climatic patterns is currently ongoing by Jane Tribble and colleagues at the University of Hawai`i.
The chronology of the Ordy Pond sequence is based on three AMS radiocarbon determinations derived from terrestrial plant material (a piece of wood and two seeds) found in the sediments. Use of terrestrial plant material for dating was critical to avoid an old carbon effect from life forms that metabolized carbon derived from the dissolved limestone of the sink (the dating of algal mat samples was attempted, but the results were obviously invalid). In addition, the advent of pollen from historically introduced species in the sediment column provides an important chronological marker for more recent sediments. Based on historical information concerning the introduction and spread of particular plants (e.g., Wagner et al., 1990), the earliest historical pollen is estimated to have begun appearing in significant quantities about AD 1830, corresponding to 120 BP using 1950 as a reference. A depth-age curve based on the dating information is illustrated in Figure 3. This curve allows the derivation of specific dates by linear interpolation for any interval of the core (see Athens et al., 1999:71 for dates of all analyzed intervals); all interpolated ages must be understood to have an unspecified error range similar to the radiocarbon determinations on which they are based.
[FIGURE 3 OMITTED]
Thirty-two pollen samples were processed from the Ordy Pond core in accordance with standard protocols for Pacific lowland sediments (for details see Athens et al., 1999:70, Moore et al., 1991). A pollen percentage diagram with graphs for pollen and charcoal concentrations and pollen sums for each interval is presented in Figure 4. Charcoal particles were counted using an eye-piece graticle with a 10 x 10 grid square pattern (Patterson et al., 1987); exotic marker Lycopodium spores were counted along with them. Charcoal was counted in grid square size classes and the total area in [mm.sup.2] was determined, from which concentrations could be calculated.
[FIGURE 4 OMITTED]
The Figure 4 pollen diagram illustrates only the most common pollen and spores along with charcoal particle counts. In interpreting this diagram, it is important to note the special nature of this record, including the fact that most pollen, other microfossils, and sediments derive from the immediate vicinity of Ordy Pond (i.e., Ordy Pond does not receive water-borne sediments from distant sources within a fluvial watershed). Thus, Ordy Pond contains a largely local paleoenvironmental record (though the presence of extra-regional wind transported pollen is clearly evident from the very sparse presence of a highly diverse assemblage of mesic-wet forest types in the pre-Polynesian intervals--these are mostly derived from the higher elevation areas of the Wai`anae and Ko`olau Mountains). Secondly, the laminae tend to insure the stratigraphic integrity of the analyzed samples, minimizing chances for any mixing of sediments as a result of biotic activity. What this means is that any "blurring" of the pollen results (or other microfossil evidence) as a result of the mixing of sediments from closely sampled intervals (i.e., time periods) is minimized.
An assumption critically important for our interpretation of the Ordy Pond pollen sequence is that the sedimentary record is intact and that there are no periods during which sediment did not accumulate. If there were an unconformity (or multiple unconformities), we would potentially expect to see a gradual pollen transition fore-shortened into what could appear in the pollen record as an abrupt change. Although we do see an abrupt transition between Zones A and B (as will be discussed below), there are reasons to believe it is not the result of a sedimentary unconformity. First, the diatom record (Blinn, 1999), which indicates an alkaline to brackish water environment of moderate to high salinity, shows that a water column was continuously present in the Ordy sinkhole. Thus, erosion or lack of sediment accumulation for a protracted period seems unlikely. This finding also appears logical in view of the nearness of the sinkhole to the coast (700 m) and the depth of the sediments below mean sea level (4 to 12.7 m), suggesting the likely infiltration of seawater into the sinkhole. Secondly, the sedimentary and isotopic evidence do not suggest that the pond ever dried following the postglacial rise in sea level (J. Tribble, pers. comm., 2001). Third, although the pollen diagram shows widely fluctuating values over time for some of the more common pollen types, which might suggest preservation problems due to drying of the sinkhole, a close inspection of the pollen diagram indicates that the timing of the fluctuations is not precisely the same for the different taxa (see Figs. 4 and 5). Thus, Pritchardia values seem to change out of phase with most of the other types, possibly suggesting short-term climate fluctuations rather than a drying out of the sinkhole.
[FIGURE 5 OMITTED]
For purposes of this study, one of the most interesting features of the pollen diagram concerns the very abrupt transition in pollen values separating the lower pre-Polynesian Zone A from the later Zone B, which pertains to the prehistoric Polynesian period (Fig. 4).
The pollen of Zone A is indicative of a mixed dryland forest. Pritchardia palms and Dodonaea viscosa (a small tree or shrub) are apparently common among the larger taxa, and there are more limited representations of Nestegis sandwicensis, Pouteria, Colubrina, and other types whose counts are too limited to represent in the diagram (Diospyros, Elaeocarpus, Erythrina sandwicensis, Nothocestrum, and others). The understory seems to have been dominated by Kanaloa kahoolawensis, chenoams (probably Chenopodium oahuensis), and Chamaesyce along with other less numerous pollen types (Achyranthes splendens-type, Asteraceae (high-spined), Sesbania tomentosa, Sicyos, Solanum sandwicense, grass, possibly a non-aquatic sedge, and others). The analysis of pollen from the earlier evaporite pan produced similar results for the pre-Polynesian middle to late Holocene, though there were some interesting differences in the early Holocene (see Athens et al., 1999).
The transition from Zone A to Zone B is interpreted as marking the appearance of humans on O`ahu. Charcoal particles and Cocos nucifera (coconut), both anthropogenic indicators, occur in the lowermost interval of Zone B. However, what is striking is the abruptness of the floristic change that occurs at the end of Zone A and the start of Zone B. The dryland forest seems to disappear almost at once and there is a concurrent surge in the cheno-am pollen type (i.e., Chenopodiaceae and/or Amaranthaceae pollen, which cannot be distunguished palynologically) and a more limited surge in grass pollen (Poaceae). The latter pollen types are considered diagnostic of an open canopy or disturbance. There can be no doubt that we are witnessing the wholesale and rapid disappearance of the native forest on the `Ewa Plain.
The interpolated depth-age dates and the stratigraphic data suggest that the landscape underwent a transformation of its vegetation within a period of perhaps no more than 50 to 100 years starting at about 930 cal BP (AD 1020; interpolated date). Such a rapid change is not a peculiarity of the Ordy Pond core. Subsequent coring investigations at Weli Pond, close to Ke`ehi Lagoon in Honolulu, also documented the very rapid replacement of the native forest by disturbance or open canopy flora (Athens and Ward, 2000). However, in this case the timing of this change appears to have been slightly earlier, at about 1074 cal BP (AD 876; interpolated).
Microscopic charcoal particles in the Ordy Pond sediments display two interesting features. First, their earliest occurrence is one or two sampling intervals after the beginning of precipitous declines in Pritchardia, Kanaloa, and Chamaesyce, and after the start of the surge of cheno-ams (Fig. 5). On the basis of the Ordy Pond stratigraphic data alone, therefore, forest loss must have begun prior to any direct evidence for humans on the `Ewa Plain. Humans clearly could not have been directly responsible for the forest decline through burning or agricultural activities.
Second, the particle size distribution graph indicates that there is a prevalence of small-size charcoal particles in the earliest sampling intervals in which they appear (Fig. 6). As Morrison (1994:675) observes, "charcoal particles in different size categories do not always co-vary in the core, reflecting differences between more local and more regional fire histories." This is because the smallest particles are more likely to be suspended by wind and transported long distances (Clark, 1988). Thus, a prevalence of the smallest particles in the earliest intervals where charcoal particles occur in the Ordy Pond core suggests a relatively distant origin of anthropogenic fires and hence settlement (this supposition is consistent with the direction of the prevailing northeast trade winds in Hawai`i). The opposite, of course, would indicate the occurrence of local fires and proximal settlement, as seems to have first occurred on the `Ewa Plain around 700 cal BP or shortly thereafter (see Fig. 6).
[FIGURE 6 OMITTED]
In sum, both the stratigraphic position of the earliest charcoal particles with respect to pollen changes, and also the particle size data imply that forest decline on the `Ewa Plain preceded human presence. In light of this evidence, it is worth exploring further the question of the chronology of human settlement on the `Ewa Plain.
Chronology of human settlement
The Ordy Pond charcoal particle size data (see Fig. 6) do not indicate settlement and use of the `Ewa Plain until after about 700 cal BP (AD 1250; interpolated), with the first large particle peak indicative of significant occupation or land use about 606 cal BP (AD 1344; interpolated). This conclusion is consistent with the chronology of human settlement based on archaeological radiocarbon dates from the `Ewa Plain.
There are 194 charcoal (or material attributed to be charcoal) radiocarbon age determinations from cultural contexts on the `Ewa Plain, and these are displayed in a histogram in Figure 7 (data summarized in Athens et al., 1999:260-265). (5) Analysis of these dates indicates that some very limited human activity may have occurred on the `Ewa Plain in the AD 1000-1250 period, (6) but that activity (and inferred population growth) began to dramatically increase in the AD 1250-1300 period (Athens et al., 1999; Tuggle, 1997; Tuggle and Tomonari-Tuggle, 1997a). Another useful way to view the aggregate radio-carbon data is in the form of a cumulative probability curve in Figure 8, which can be interpreted as a population growth curve (see Dye and Komori, 1992a; based on method described by Dye and Komori, 1992b). This pattern is consistent with the population curve and history of Hawaiian cultural growth based on radiocarbon dates developed by Dye and Komori (1992b) and Dye (1994a), in which the "growth phase" begins about AD 1150, or some 100 years before the start of the `Ewa Plain "growth phase."
[FIGURES 7-8 OMITTED]
The combined paleontological and archaeological data indicate that the collapse of the native forest on the `Ewa Plain began after human colonization of the Hawaiian Islands, but prior to rapid population growth and expansion in Hawai`i in general (around AD 1150), and certainly before significant settlement of the `Ewa Plain.
`Ewa Plain sinkholes: avian remains and chronology
Limestone sinks are a common natural feature of the `Ewa Plain. They range in diameter from less than a meter to 400 m. The largest are known primarily from early maps. These natural features probably numbered in the thousands before historical filling for agriculture, military training, and development. Most remaining today are between 50 cm and 2 m in width, and contain deposits whose surfaces (the floors of the sinkhole) are from 50 cm to 2 m below the top of the limestone (Fig. 9). Excavation indicates that most deposits are 1 to 3 m in depth before the limestone base is encountered. Many sinkholes have been culturally modified, with such features as a ring of stones placed around the opening (usually indicating agricultural use of the sinkhole), attached platforms or pavings, and rock fill. Deposits often have some cultural debris, either as primary or secondary deposition. Sinkholes generally have no evidence of activity within the sinks themselves, except that a few were used for human burial. Most sinkholes are repositories of paleontological remains (especially subfossil avian bones and terrestrial snails--for discussion of latter, see Dye and Tuggle, 1998, 2001; Athens et al., 1999:155-166, 209-211). The sinks served as natural traps for these remains, resulting primarily from deposition of surface sediments carried in by sheet wash from occasional heavy rain storms. As expected from this depositional process, skeletal elements of the subfossil remains rarely show any evidence of articulation.
[FIGURE 9 OMITTED]
A summary of the types, numbers, and survival status of avian remains recovered from sinkholes excavated during a recent project is presented in Table 1. These figures are representative of the `Ewa Plain as a whole (see Table 2). The assemblages are dominated by sea birds (74 to 91 percent), with passeriforms and anseriforms making up the only other significant percentages (5 to 23 percent and 1 to 9 percent, respectively).
An example of the stratigraphic distribution of the types of paleontological and archaeological remains found in sinkholes is provided by the results of a sink excavated at Site 5108, Feature 1 (or-F1; Fig. 9; Table 3--see Athens et al., 1999:275-276 for formal sediment descriptions and other details). This was a relatively small but rich paleontological repository. The drilled cowrie (Cypraea sp.) shell octopus lure found at the base of Layer II (see Fig. 9) is a clear indicator that sedimentary deposition of this strata began during the traditional Hawaiian time period. However, this does not mean that Layer II is a primary cultural deposit or that its deposits are stratigraphically intact. There are no animal domesticates (pig, dog, or chicken), and the relatively small fish bones were probably introduced by birds. Thus, Layer II seems to be a paleontological deposit probably dating to the prehistoric Hawaiian period despite the presence of historically-introduced Mus musculus (mouse) bones (see below). Its significance is that it fails to provide evidence for human contemporaneity with any of the extinct land birds, though procellariids are present (see Table 4).
The relative densities of total bird bone (much of it unidentifiable), procellariids, all native land birds, R. exulans, lizards, and mice in the sink at Site 5108-F1 are shown graphically (Fig. 10). It is perhaps significant that the density of R exulans bones in the sinkhole deposit makes up such a large proportion of the total bone, considering it is only a single species and its bones have had much less time to accumulate than the native avian bones.
[FIGURE 10 OMITTED]
As Table 3 suggests, there is little stratigraphic integrity in the subfossil deposits despite the evident layering of the sediments. R. exulans and the lizards, which are Polynesian introductions, occur throughout the deposits even though sediments almost certainly have been accumulating in the Site 5108-F1 sink since long before the arrival of Polynesians. Bones of M. musculus, an historic introduction, are also found in all but the lowest layer. Trace amounts of charcoal, associated with human activity, are also found in all but the lowest two layers.
The lack of stratigraphic integrity in the Site 5108-F1 deposits is confirmed when radiocarbon dates on bird bone and R. exulans bone from several excavations are plotted (Fig. 11; Athens et al., 1999). As may be seen, surface bone may be quite old (all dates calibrated, 2 sigma range). The Figure 11 graph illustrates surface dates of AD 569-971 and AD 1214-1520 (both Pterodroma phaeopygia) from Sites 1752-F67 and 1753-F2. Also, four bone samples from near the base of Layer III at Site 5108-F1 display strikingly different ages: 1573-1066 BC (Oceanodroma castro), AD 942-1172 (Branta sp.), AD 1268-1401 (R. exulans), and AD 892-1160 (R. exulans--probably a different individual). Note that for Site 1752-F60 there is an inversion of ages from samples deriving from the same strata (Layer III) but slightly different levels. The disparity of ages, however, is no less striking. Here dates were obtained of AD 783-1016 from the 30-35 cm level (Branta sp.) and AD 1669-1950 from the 40-45 cm level (charcoal--primarily Scaevola cf. coriacea). Finally, from Site 5094-F1, two of the deepest sinkhole samples produced dates of AD 211-429 (Thambetochen xanion) and AD 972-1179 (R. exulans). These results are also stratigraphically inverted, with the older T. xanion date deriving from Layer II at 70-75 cmbs, while the R. exulans date derives from Layer II at 75-85 cmbs. In sum, there are obviously major problems in regard to the stratigraphic integrity of the sinkhole paleontological deposits.
[FIGURE 11 OMITTED]
There seem to be several natural disturbance processes affecting most sinkholes (Athens et al., 1999:147-148; Hammatt and Shideler, 1995; Tuggle, 1997; Franklin et al., 1995; Beardsley, 2001). Surface flooding from occasional heavy rains is one such process because the water induces percolation as it drains into the sinkholes. These rainfall events may also lead to redeposition of outside surface bird bones and other materials into the sinkholes. The flooding problem is no doubt aggravated by the relatively coarsely textured sinkhole sediments, facilitating downward migration of materials that are either within the sinkhole or have been introduced from the outside the sinkhole. Although sinkhole deposits possess sedimentary layers, it is a mistake to assume that paleontological materials associated with R. exulans or other Polynesian horizon markers in sinkhole deposits have a coeval age (see also radiocarbon dating evidence below).
Another example of the dating problem concerns the finding of bones of two types of extinct geese associated with an archaeological hearth feature discovered inside a large sinkhole (Olson and James, 1982:28,31). A charcoal sample from the hearth dated to AD 1205-1299 (2 sig. cal: range). Bones of P. phaeopygia (a petrel that no longer occurs on O`ahu), Branta sp. (an extinct goose), another type of unnamed extinct goose, and R. exulans were found "in and around [the] hearth" (Olson and James, 1982:31). The difficulty is that paleontological remains extend to the surface of many sinkhole deposits (e.g., Franklin et al., 1995; Wickler and Tuggle, 1997). This suggests that mixing of archaeological and paleontological deposits of very different ages inside sinkholes is virtually a foregone conclusion. Without radiocarbon determinations performed directly on the bird bone samples to demonstrate contemporaneity, or other compelling evidence to confirm a human association (e.g., butcher marks on the bone or evidence of burning), it is unwarranted to assume that the land bird remains are anything other than a part of the natural pre-Polynesian paleontological assemblage. In this case the excavator of the sinkhole hearth, P. McCoy, has stated that the cultural context of the bird bone is uncertain (pers. comm. cited by Tuggle and Tomonari-Tuggle, 1997a:92).
In terms of the overall chronology of avian subfossil remains in the `Ewa Plain sinkholes, there are a total of 46 usable radiocarbon determinations (Table 5). Although dating of such remains in the absence of meaningful stratigraphic context has obvious sampling problems, there are a number of important points that can be considered.
One is that a significant change occurs about AD 1000. Prior to this time a range of taxa is found, indicating a general accumulation of avian remains from the early mid-Holocene. However, after AD 1000 only Pterodroma phaeopygia is present in the sinkholes. This is completely consistent with the archaeological record. Lowland archaeological deposits, if they contain any avian remains, have almost exclusively seabirds.
Further, the dating sequence for the `Ewa Plain avian remains indicates the temporal overlap with the early period of Hawaiian settlement for some extinct taxa. Four individual Large Anatid (including three identified as Branta sp.), one flightless rail (Porzana ziegleri) and one passeriform (Chaetopila angustipluma) have radiocarbon dates that overlap the AD 800-1000 range that defines the Hawaiian colonization period, indicating that these taxa did not become extinct in the pre-Polynesian period. The problems of sampling and the statistical nature of radiocarbon dating ranges do not allow the conclusion that these birds necessarily became extinct during the early Polynesian period, but the evidence is consistent with this interpretation given the absence of such remains in later archaeological deposits.
`Ewa Plain archaeological sites and avian remains
Excavation of non-sinkhole habitation sites on the `Ewa Plain has produced very little bird bone, despite good bone preservation. The few bird bones present are almost exclusively sea birds (procellariids; see summary in Tuggle and Tomonari-Tuggle, 1997a:91-93). There are virtually no land birds represented, such as the presumably highly desirable anatids (ducks, geese). (7) This is, in fact, a common pattern of lowland sites throughout Hawai`i; indisputable archaeological remains indicating human use of land birds are extremely rare (see also Olson and James, 1982:29; Collins, 1995; Moniz, 1997). Upper elevation archaeological deposits in lava tube caves and shelters, however, do contain avian land bird remains (e.g., Athens et al., 1991 for discussion of sites in the saddle area of Hawai`i Island at elevations between 1,500 and 2,000 m), indicating that their scarcity in lowland sites is not a question of archaeological recognition.
The avian extinction problem
Research on the `Ewa Plain indicates that subfossil avian remains have been accumulating in the sinkholes since the early Holocene and that most of the extinctions occurred around the time of human colonization of Hawai`i or shortly thereafter. The four major possible explanations for avian extinctions are predation by humans, predation by rats, loss of habitat, and disease. There is no evidence to address the question of disease, and predation by rats (R. exulans) is unlikely given their food preferences, as discussed below.
Concerning human predation, unlike many other places in the Pacific, early Hawaiian lowland archaeological deposits show no evidence for accumulation of remains of now-extinct or extirpated birds, and avian bones from non-archaeological deposits have no indication of human modification. They also show that marine birds continued to exist and to be utilized throughout the period of prehistoric Hawaiian occupation. Thus, the suggestion of predation as a cause of extinction/extirpation is not supported by the terrestrial avifauna remains of the `Ewa Plain or anywhere else in the Hawaiian lowlands.
Paleoenvironmental coring demonstrates there was a rapid loss of the lowland forest roughly concurrent with the time of Polynesian colonization of Hawai`i and largely coincident with the avian extinctions. This timing suggests a highly probable causal model of habitat loss leading to extinction. However, the paleontological and archaeological data indicate that this habitat destruction occurred prior to human occupation of the `Ewa Plain, and certainly prior to any Hawaiian agricultural expansion or forest burning. The absence of evidence for land bird predation in archaeological sites on the `Ewa Plain and elsewhere in the Hawaiian lowlands thus has an explanation. Most of the land birds were gone before the Hawaiians even had a chance to make use of them for subsistence.
Vegetation and rats
If forest loss is the answer to avian extinction, the problem refocuses on the cause of the forest decline, and we suggest that a good case can be made that the Polynesian rat, R. exulans, was a significant destructive agent. The Polynesian rat has been considered as a factor in avian extinction through direct predation (e.g. Kirch, 1985:291), and has been discussed in regard to possible effects on vegetation (e.g. Cuddihy and Stone 1990), but the `Ewa Plain research provides a set of data that suggests it played a dramatic role in environmental change.
This rat, which was the first non-human land mammal to reach the shores of Hawai`i, was introduced either by the initial Polynesian discoverers/settlers, or soon thereafter by other Polynesian voyagers. There are several points to make regarding the suitability of R. exulans as a prime suspect in the demise of the forest. The first concerns the dating evidence.
Following the lead of James et al. (1987), we undertook the radiocarbon dating of three rat bones in our recent project (Athens et al., 1999:347) to use as a proxy indicator for the arrival of humans in the islands, as well as for their history on the `Ewa Plain. Another series of five bone dates for R. exulans also has been recently reported by McDermott et al. (2000). The results of these dates are listed in Table 6 and also shown in Figure 12.
[FIGURE 12 OMITTED]
Although there is no stratigraphic basis for insuring that one is dating the earliest occurrences of R. exulans in the sinkhole deposits, the results indicate that R. exulans was probably present on the `Ewa Plan at the time the native forest disappeared and well before human settlement on the Plain. Because of the statistical nature of radiocarbon dating, it may be impossible to narrow the ranges much further. Dye's (2000b) Bayesian calibration of bird bone and rat bone dates obtained by McDermott et al. (2000) from sinkholes (see Table 6), shows that there was some likelihood for the overlap of now-extinct anatids with R. exulans on the `Ewa Plain.
The second point is that rat bones are relatively common in the sinkhole deposits (see Fig. 10). This may indicate that R. exulans attained relatively high population densities in the past in comparison to the native fauna. It would be valuable to test this model by the radiocarbon dating of a large sample of rat bones randomly selected from sinkhole deposits to develop a population curve for their relative abundance through time.
Third, rats arrived in Hawai`i with no predators except possibly the Hawaiian hawk or the now extinct eagle, which may have restricted their diets to familiar birds in any case (see Olson and James, 1991:64, 67). Competition for food from other animals (such as folivorous anatids--e.g., James and Burney, 1997) may also have been of little consequence. It is also hard to imagine seed eating birds providing very effective competition for the rats since the latter are excellent climbers and can reach almost anywhere the birds can for obtaining seeds. Further, with their teeth, the rats can open even the hardest and thickest seed cases, possibly enabling them to consume some seeds before they might be available to birds when they break open naturally. Although not a competitive factor, it is also interesting to note that seeds, after consumption by rats, presumably would have no chance of viability unlike some seeds that pass through the digestive tract of birds. Thus, given the fact that R. exulans is a very fecund mammal, capable of having four to six litters of just over 4 young each per year on average (Kramer, 1971), it seems likely that their population could have expanded very quickly at an exponential rate. In just a decade or two, rats could have densely covered an entire island like O`ahu up to their elevation maximum.
Fourth, numerous studies show that the primary food preference of R. exulans is plant matter (Mosby et al., 1973; Norman, 1975: Strecker and Jackson, 1982; Temme, 1979; and Wodzicki, 1978/79:442). The argument has been made that rats may have preyed on ground-dwelling birds (Cuddihy and Stone, 1990:34), but a review of relevant case studies indicates that this is rare and occurred only when R. exulans was under severe survival stress in the absence of plant material (Wirtz, 1972; Tomich, 1986), or is occasional and minor (cf. Mosby et al., 1973:808). These rats have been found to rely on plant material as diverse as coconut, sugarcane, pandanus, various fruits, and grasses (summarized in Atkinson and Moiler, 1990). In short term studies, Polynesian rats have been shown to affect coastal forest composition during regeneration (Campbell, 1978). That the Hawaiian native flora was particularly attractive to R. exulans is suggested by a statement made by a horticulturalist who specializes in the propagation of endangered native Hawaiian plants: "Let rats near them and they'll be eaten.... They're like candy to rats" (TenBruggencate, 1997). The rats could have produced their damage through seed and fruit predation, consumption of seedlings and new leaf production, and girding of soft-barked trees, as well as consumption of invertebrates critical to plant pollination and the nutrient cycle (see discussion in Cuddihy and Stone, 1990:34,68-70).
While the above only serves to demonstrate the potential for rats to have been responsible for the demise of the native lowland forest and offers no proof, there is further interesting evidence in Hawai`i suggesting that rats may have played a significant role in this regard. This involves the contrast between two small islands off the north coast of Moloka`i. Huelo Island is a vertically-sided pinnacle rising 60 meters above the sea. The 30 x 60 meter sloping surface of this island is densely covered with both mature and young Pritchardia hillebrandii palms (demonstrating the palms are reproducing) and other native plants (e.g., Diospyros sandwicensis), besides being home to many Wedge-tailed Shearwater chicks in the ground litter. Wildlife surveys have not documented rats on this island. In contrast, nearby Mokapu Island (unoccupied and consisting of steeply sloping--but not vertical--surfaces) has very few Pritchardia palms, including only a single immature plant (suggesting that it is having a difficult time reproducing), but rats have been documented (K.R. Wood, pers. comm., 2000).
The discovery of two Kanaloa kahoolawensis shrubs, a previously unknown plant, on another nearly inaccessible rock spire off the south coast of Kaho`olawe Island in 1992 (Lorence and Wood, 1994) is also relevant to the possible role of rats in the destruction of native vegetation. K. kahoolawensis, representing a genus and species new to Hawaiian botany, is common in the prehuman pollen intervals of lowland cores, but as demonstrated in the Ordy Pond and other cores, it became all but extinct with the arrival of humans and associated rats. With respect to the Kaho`olawe discovery, presumably rats cannot access this extremely difficult location, possibly accounting for the survival of K. kahoolawensis.
Finally, it is of interest that native vegetation in Hawai`i is relatively common above about 1,500 m (although Pritchardia palms do not extend to this elevation with the exception of one species on Hawai'i Island, and many other lowland taxa are also absent in the higher elevations--see Gagne and Cuddihy 1990; Wagner et al. 1990). In spite of rare documented instances of its occurrence at higher elevations (Tomich, 1986:42; Cuddihy and Stone, 1990:68), the maximum elevation range for R. exulans in Hawai`i is also about 1,500 m. Thus, the persistence of native vegetation at higher elevations (and consequently, the native avifauna), may owe much to the primarily lowland natural habitat range of rats.
Given the cumulative set of data, we thus propose that R. exulans, quickly radiating throughout the islands ahead of the human settlers who had brought them, destroyed much of the native Hawaiian lowland forests by consumption of the leafy and reproductive portions of the plants. This may be at least part of the explanation why certain formerly common native woods (e.g., K. kahoolawensis) are never found in archaeological contexts despite the examination of over 600 archaeological charcoal samples (containing multiple charcoal fragments) and perhaps 1,000 single specimens of archaeological charcoal (G. Murakami, pers. comm., 2001).
The rapid forest decline in Hawai`i is not matched on other islands where R. exulans seems to have been introduced prehistorically (e.g., Palau and Guam--Athens and Ward, 1999a, 1999b). There is no immediate explanation for this, but one possible factor could relate to the high degree of endemism that characterizes Hawai`i's native vegetation--between 91 and 96+% of the angiosperms (Loope and Mueller-Dombois, 1989; see also Carlquist, 1980). The endemism may also have fostered the development of seed production characteristics that made native plants susceptible to the eating habits of the rats (see Campbell, 1978). If endemic plants of the lowland forest were an ideal food for the rats, the conditions were right for a population explosion of this new species, with the resulting vegetation collapse.
The paleoenvironmental coring data document major and very rapid vegetation change (i.e., disappearance of the dryland forest) prior to the significant presence of humans in the `Ewa Plain region, and almost certainly before any vegetation clearance by burning. The model proposed here suggests that drastic change of the native plant community led to the relatively rapid extinction or extirpation of some land bird species, particularly passerines and flightless taxa, prior to the advent of humans on the `Ewa Plain. The model thus explains the absence of early sites of bird hunters, the general absence of bird remains (except for some seabird bones) in most Hawaiian occupational sites of the region, and the general pattern of bird bones underlying cultural deposits in sinkholes (with acknowledgment of the common problem of mixing in the interface zone).
Forest decline as a cause of bird extinction also seems compatible with the limited evidence concerning the pattern of extinction; that is, the survival of procellariids well into the period of human settlement. Presumably the procellariids would have been as susceptible as forest birds (or more so) to population stress from rat or human predation, but less susceptible to stress from forest decline.
To evaluate our model we would like to find and analyze another high resolution laminated wetland deposit like Ordy Pond on perhaps another Hawaiian island. Secondly, we would like to radiocarbon date (using the XAD resin protocol--Stafford et al., 1991) more R. exulans bones not only to determine the date of their earliest occurrence on the `Ewa Plain with greater precision and assurance than is possible at present, but also to provide a statistically sound means for estimating relative population numbers through time. Finally, we believe that many more subfossil avian bones of the different extinct/extirpated taxa need to be dated using the XAD resin protocol to more firmly establish the date of their disappearance.
Table 1. Bird Bone (NISP, lowest taxon) from `Ewa Plain Sinkholes; Distribution by Site and Feature. Key: epp = extinct prehuman/prehistorically; eh = extinct historically; xpp = extirpated prehuman/prehistorically; xh = extirpated historically; pi = Polynesian introduction; mi = historic/modern introduction; p = presently surviving native; mig = migratory. TAXON Survival Status Indeterminate Bird, large -- Procellariiformes Puffinus newelli xpp/xh(?) Puffinus sp. -- Pterodroma phaeopygia xpp Indet. Procellariid, small -- Indet. Procellariid, medium -- Oceanodroma castro xpp/xh(?) Anseriformes Branta sp., cf. and/or Thambe. x. epp Branta sp. epp Thambetochen xanion epp Indet. Anatid, small epp(?) Indet. Anatid epp(?) Galliformes Gallus gallus pi Gruiformes Porzana sp. epp/eh Porzana ziegleri epp Rallid, medium epp&eh Columbiformes Streptopelia chinensis mi Geopelia striata mi Strigiformes Asio flammeus pi(?)/p Strigid, medium -- Passeriformes Corvus (large sp.) epp Chaetoptila sp. epp Passeriform, small -- Other Pluvialis fulva mig Numenius tahitiensis mig Total TAXON 5108-1 Indeterminate Bird, large 4661 Procellariiformes Puffinus newelli 1 Puffinus sp. 4 Pterodroma phaeopygia 51 Indet. Procellariid, small 58 Indet. Procellariid, medium 296 Oceanodroma castro 22 Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. 1 Thambetochen xanion -- Indet. Anatid, small -- Indet. Anatid 1 Galliformes Gallus gallus -- Gruiformes Porzana sp. -- Porzana ziegleri 3 Rallid, medium -- Columbiformes Streptopelia chinensis 2 Geopelia striata 2 Strigiformes Asio flammeus -- Strigid, medium -- Passeriformes Corvus (large sp.) 1 Chaetoptila sp. 34 Passeriform, small 269 Other Pluvialis fulva 2 Numenius tahitiensis -- Total 5408 TAXON 5094-1 Indeterminate Bird, large 296 Procellariiformes Puffinus newelli -- Puffinus sp. -- Pterodroma phaeopygia 2 Indet. Procellariid, small 26 Indet. Procellariid, medium 10 Oceanodroma castro 2 Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. -- Thambetochen xanion 1 Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus -- Gruiformes Porzana sp. 1 Porzana ziegleri 2 Rallid, medium 2 Columbiformes Streptopelia chinensis -- Geopelia striata -- Strigiformes Asio flammeus -- Strigid, medium 1 Passeriformes Corvus (large sp.) -- Chaetoptila sp. 3 Passeriform, small 70 Other Pluvialis fulva -- Numenius tahitiensis -- Total 416 TAXON 5108-4 Indeterminate Bird, large 55 Procellariiformes Puffinus newelli -- Puffinus sp. 1 Pterodroma phaeopygia 6 Indet. Procellariid, small -- Indet. Procellariid, medium 20 Oceanodroma castro -- Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. 2 Thambetochen xanion 8 Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus -- Gruiformes Porzana sp. -- Porzana ziegleri -- Rallid, medium -- Columbiformes Streptopelia chinensis -- Geopelia striata -- Strigiformes Asio flammeus -- Strigid, medium -- Passeriformes Corvus (large sp.) -- Chaetoptila sp. -- Passeriform, small 2 Other Pluvialis fulva -- Numenius tahitiensis 1 Total 95 TAXON 1752 Indeterminate Bird, large 1 Procellariiformes Puffinus newelli -- Puffinus sp. -- Pterodroma phaeopygia -- Indet. Procellariid, small -- Indet. Procellariid, medium -- Oceanodroma castro -- Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. -- Thambetochen xanion -- Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus -- Gruiformes Porzana sp. -- Porzana ziegleri -- Rallid, medium -- Columbiformes Streptopelia chinensis -- Geopelia striata -- Strigiformes Asio flammeus -- Strigid, medium -- Passeriformes Corvus (large sp.) -- Chaetoptila sp. -- Passeriform, small -- Other Pluvialis fulva -- Numenius tahitiensis -- Total 1 TAXON 1753 Indeterminate Bird, large 2986 Procellariiformes Puffinus newelli -- Puffinus sp. -- Pterodroma phaeopygia 76 Indet. Procellariid, small 1 Indet. Procellariid, medium 521 Oceanodroma castro 0 Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. 1 Thambetochen xanion -- Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus -- Gruiformes Porzana sp. -- Porzana ziegleri -- Rallid, medium -- Columbiformes Streptopelia chinensis -- Geopelia striata 1 Strigiformes Asio flammeus -- Strigid, medium -- Passeriformes Corvus (large sp.) -- Chaetoptila sp. -- Passeriform, small 2 Other Pluvialis fulva -- Numenius tahitiensis -- Total 3588 TAXON 1754 Indeterminate Bird, large 34 Procellariiformes Puffinus newelli -- Puffinus sp. -- Pterodroma phaeopygia 2 Indet. Procellariid, small 3 Indet. Procellariid, medium 3 Oceanodroma castro -- Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. -- Thambetochen xanion -- Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus -- Gruiformes Porzana sp. -- Porzana ziegleri -- Rallid, medium -- Columbiformes Streptopelia chinensis -- Geopelia striata 1 Strigiformes Asio flammeus 2 Strigid, medium -- Passeriformes Corvus (large sp.) -- Chaetoptila sp. -- Passeriform, small -- Other Pluvialis fulva -- Numenius tahitiensis -- Total 45 TAXON 1755 Indeterminate Bird, large 1977 Procellariiformes Puffinus newelli -- Puffinus sp. -- Pterodroma phaeopygia 31 Indet. Procellariid, small 5 Indet. Procellariid, medium 222 Oceanodroma castro -- Anseriformes Branta sp., cf. and/or Thambe. x. 2 Branta sp. 8 Thambetochen xanion -- Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus 10 Gruiformes Porzana sp. -- Porzana ziegleri 2 Rallid, medium -- Columbiformes Streptopelia chinensis -- Geopelia striata -- Strigiformes Asio flammeus 11 Strigid, medium -- Passeriformes Corvus (large sp.) -- Chaetoptila sp. -- Passeriform, small 9 Other Pluvialis fulva 1 Numenius tahitiensis -- Total 2278 TAXON 1724 Indeterminate Bird, large 99 Procellariiformes Puffinus newelli -- Puffinus sp. -- Pterodroma phaeopygia -- Indet. Procellariid, small 4 Indet. Procellariid, medium 12 Oceanodroma castro -- Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. -- Thambetochen xanion 1 Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus -- Gruiformes Porzana sp. 1 Porzana ziegleri -- Rallid, medium 1 Columbiformes Streptopelia chinensis -- Geopelia striata -- Strigiformes Asio flammeus -- Strigid, medium 1 Passeriformes Corvus (large sp.) -- Chaetoptila sp. -- Passeriform, small -- Other Pluvialis fulva -- Numenius tahitiensis -- Total 119 TAXON 5129-1 Indeterminate Bird, large 84 Procellariiformes Puffinus newelli -- Puffinus sp. -- Pterodroma phaeopygia -- Indet. Procellariid, small -- Indet. Procellariid, medium 3 Oceanodroma castro -- Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. -- Thambetochen xanion -- Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus -- Gruiformes Porzana sp. -- Porzana ziegleri -- Rallid, medium -- Columbiformes Streptopelia chinensis -- Geopelia striata -- Strigiformes Asio flammeus -- Strigid, medium -- Passeriformes Corvus (large sp.) -- Chaetoptila sp. -- Passeriform, small 1 Other Pluvialis fulva -- Numenius tahitiensis -- Total 88 TAXON 1757-2 Indeterminate Bird, large 11 Procellariiformes Puffinus newelli -- Puffinus sp. -- Pterodroma phaeopygia -- Indet. Procellariid, small -- Indet. Procellariid, medium -- Oceanodroma castro -- Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. -- Thambetochen xanion -- Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus -- Gruiformes Porzana sp. -- Porzana ziegleri -- Rallid, medium -- Columbiformes Streptopelia chinensis 8 Geopelia striata 1 Strigiformes Asio flammeus -- Strigid, medium -- Passeriformes Corvus (large sp.) -- Chaetoptila sp. -- Passeriform, small 8 Other Pluvialis fulva -- Numenius tahitiensis -- Total 28 TAXON 1756 Indeterminate Bird, large 1 Procellariiformes Puffinus newelli -- Puffinus sp. -- Pterodroma phaeopygia -- Indet. Procellariid, small -- Indet. Procellariid, medium -- Oceanodroma castro -- Anseriformes Branta sp., cf. and/or Thambe. x. -- Branta sp. -- Thambetochen xanion -- Indet. Anatid, small -- Indet. Anatid -- Galliformes Gallus gallus -- Gruiformes Porzana sp. -- Porzana ziegleri -- Rallid, medium -- Columbiformes Streptopelia chinensis -- Geopelia striata -- Strigiformes Asio flammeus -- Strigid, medium -- Passeriformes Corvus (large sp.) -- Chaetoptila sp. -- Passeriform, small 1 Other Pluvialis fulva -- Numenius tahitiensis -- Total 2 TAXON Ct Indeterminate Bird, large 10205 Procellariiformes Puffinus newelli 1 Puffinus sp. 5 Pterodroma phaeopygia 168 Indet. Procellariid, small 97 Indet. Procellariid, medium 1087 Oceanodroma castro 24 Anseriformes Branta sp., cf. and/or Thambe. x. 2 Branta sp. 12 Thambetochen xanion 10 Indet. Anatid, small 0 Indet. Anatid 1 Galliformes Gallus gallus 10 Gruiformes Porzana sp. 2 Porzana ziegleri 7 Rallid, medium 3 Columbiformes Streptopelia chinensis 10 Geopelia striata 5 Strigiformes Asio flammeus 13 Strigid, medium 2 Passeriformes Corvus (large sp.) 1 Chaetoptila sp. 37 Passeriform, small 362 Other Pluvialis fulva 3 Numenius tahitiensis 1 Total 12068 Table 2. Percentages of Identified Bird Bone in `Ewa Plain Sinkholes. NAS (1): Athens et al. (1999); NAS (2): calculated from Beardsley (2001: Table V-20); Deep Draft (3) and West Beach (3): Tuggle and Tomonari-Tuggle (1997b:Table 1). Deep West AVES NAS (1) NAS (2) Draft (3) Beach (3) % NISP % NISP % NISP % NISP Procellariiformes 74 76 91 75 Pelicaniformes 0 <1 0 Anseriformes 1 9 1 6 Falconiformes 0 0 <1 <1 Galliformes <1 <1 <1 1 Gruiformes 1 2 <1 1 Columbiformes 1 <1 0 <1 Strigiformes <1 1 1 <1 Passeriformes 21 10 5 16 Other <1 <1 <1 <1 100% 100% 100% 100% Total Identified NISP 1872 1782 5285 25373 Total Bird Bone 1872 ? 15134 72781 Table 3. Site 5108-F1: Bone, Shell, and Artifacts. Bird Rattus bone exulans Strata n g n g I 47 1.26 174 3.07 II 612 31.17 712 12.55 III-1 2483 107.04 1004 16.61 (II-III) 819 41.28 199 3.94 III-2 508 15.16 159 2.72 III-3 254 9.71 167 3.22 IV 78 4.07 22 0.56 V 86 2.79 20 0.41 Misc. 521 23.95 275 4.85 Total 5408 236.43 2732 47.93 Mus R. norvegicus musculus or R. rattus Strata n g n g I 25 0.25 2 0.06 II 63 0.55 15 0.67 III-1 49 0.33 2 0.10 (II-III) 14 0.04 III-2 11 0.08 III-3 5 0.06 IV 1 0.03 V -- -- Misc. 21 0.09 Total 189 1.43 19 0.83 Fish Lizard, small Strata n g n g I 3 0.17 95 0.38 II 30 0.72 353 0.97 III-1 47 0.35 741 1.26 (II-III) 7 0.11 105 0.21 III-2 11 0.07 58 0.16 III-3 1 0.01 41 0.11 IV -- -- 4 0.05 V -- -- 10 0.01 Misc. 13 0.80 123 0.24 Total 112 2.23 1530 3.39 Vertebrate Other unidentified Strata n g n I 41 Herpestes auropunctatus -1 II 136 0.89 III-1 973 2.97 (II-III) 305 1.47 III-2 184 0.97 III-3 133 0.61 Vespertilionid Bat -1 IV 8 0.17 Vespertilionid Bat -1 V 46 0.25 Misc. 74 0.39 Total 1900 7.72 -- Shell Charcoal Artifacts Strata n g I 6 0.14 II 14 0.29 cowrie lure III-1 3 0.25 (II-III) 6 0.04 III-2 3 0.07 III-3 2 0.02 IV V Misc. 3 0.02 Total 37 0.83 Table 4. Site 5108-F1: Bird Bone Distribution (by count). ORDER FAMILY TAXON Procellariiformes Procellariidae Puffinus newelli Puffinus sp. Pterodroma phaeopygia Indet. Procellariid, small Indet. Procellariid, medium Hydrobatidae Oceanodroma castro Anseriformes Anatidae Branta sp., cf. and/or Thambe. x. Branta sp. Thambetochen xanion Indet. Anatid, small Indet. Anatid Falconiformes Accipitridae Buteo solitarus Galliformes Phasianidae Gallus gallus Gruiformes Rallidae Porzana ziegleri Porzana sp. Indet. Rallid, medium Columbiformes Columbidae Streptopelia chinensis Geopelia striata Strigiformes Strigidae Asio flammeus Strigid, medium Passeriformes Corvidae Corvus (large sp.) Meliphagidae Chaetoptila sp. Indeterminate small, or cf. Other Pluvialis fulva, cf. Numenius tahitiensis Indeterminate Indeterminate Bird (egg shell) Total number Grant Total TAXON I II III-1 (II-III) Puffinus newelli Puffinus sp. 3 Pterodroma phaeopygia 5 21 13 Indet. Procellariid, small 5 24 17 Indet. Procellariid, medium 1 42 111 64 Oceanodroma castro 1 2 8 4 Branta sp., cf. and/or Thambe. x. Branta sp. Thambetochen xanion Indet. Anatid, small Indet. Anatid 1 Buteo solitarus Gallus gallus Porzana ziegleri 1 Porzana sp. Indet. Rallid, medium Streptopelia chinensis 1 Geopelia striata 2 Asio flammeus Strigid, medium Corvus (large sp.) 1 Chaetoptila sp. 1 1 11 10 small, or cf. 8 23 94 44 Pluvialis fulva, cf. 2 Numenius tahitiensis Bird 35 530 2210 665 (3) (4) (12) (1) Total number 47 612 2483 819 Grant Total TAXON III-2 III-3 IV V Puffinus newelli 1 Puffinus sp. Pterodroma phaeopygia 1 1 1 Indet. Procellariid, small 1 2 3 Indet. Procellariid, medium 19 11 5 2 Oceanodroma castro 4 2 1 Branta sp., cf. and/or Thambe. x. Branta sp. 1 Thambetochen xanion Indet. Anatid, small Indet. Anatid Buteo solitarus Gallus gallus Porzana ziegleri 1 Porzana sp. Indet. Rallid, medium Streptopelia chinensis Geopelia striata Asio flammeus Strigid, medium Corvus (large sp.) Chaetoptila sp. 2 1 small, or cf. 30 19 10 7 Pluvialis fulva, cf. Numenius tahitiensis Bird 453 215 57 76 (5) (2) (1) Total number 508 254 78 86 Grant Total TAXON Misc. total total no. g Puffinus newelli 1 0.3 Puffinus sp. 1 4 0.8 Pterodroma phaeopygia 9 51 9.9 Indet. Procellariid, small 6 58 8.6 Indet. Procellariid, medium 41 296 86.6 Oceanodroma castro 22 0.8 Branta sp., cf. and/or Thambe. x. Branta sp. 1 0.5 Thambetochen xanion Indet. Anatid, small Indet. Anatid 1 0.1 Buteo solitarus Gallus gallus Porzana ziegleri 1 3 0.1 Porzana sp. Indet. Rallid, medium Streptopelia chinensis 1 2 0.2 Geopelia striata 2 0.1 Asio flammeus Strigid, medium Corvus (large sp.) 1 0.2 Chaetoptila sp. 8 34 2.8 small, or cf. 34 269 4.9 Pluvialis fulva, cf. 2 0.1 Numenius tahitiensis Bird 420 4661 120.7 (2) (30) Total number 521 Grant Total 5408 236.7 Table 5. Radiocarbon Dates on Bird Bone from the `Ewa Plain. Age, calibrated, Avian Taxon Location 2 sigma * AD 1651-1950 Pterodroma phaeopygia DDH AD 1549-1950 (1) Pterodroma phaeopygia DDH AD 1420-1880 Pterodroma phaeopygia DDH AD 1508-1950 Pterodroma phaeopygia DDH AD 1390-1870 Pterodroma phaeopygia DDH AD 1360-1810 Pterodroma phaeopygia DDH AD 1320-1720 Pterodroma phaeopygia DDH AD 1300-1589 Pterodroma phaeopygia DDHE AD 1290-1720 Pterodroma phaeopygia DDH AD 1280-1740 (2) Pterodroma phaeopygia DDH AD 1280-1660 Pterodroma phaeopygia DDH AD 1214-1520 Pterodroma phaeopygia NASBP AD 1200-1540 Pterodroma phaeopygia DDH AD 1150-1510 Pterodroma phaeopygia DDH AD 1030-1339 Pterodroma phaeopygia DDHE AD 1010-1410 Pterodroma phaeopygia DDH AD 942-1172 Branta sp. NASBP AD 900-1159 Branta sp. DDHE AD 890-1340 Pterodroma phaeopygia DDH AD 890-1040 Chaetoptila angustipluma cf. EM AD 810-1049 Med. or Large Anatid DDHE AD 783-1016 Branta sp. NASBP AD 650-869 Porzana ziegleri DDHE AD 580-1060 Pterodroma phaeopygia DDH AD 569-971 Pterodroma phaeopygia NASBP AD 440-639 Thambetochen xanion DDHE AD 420-629 Thambetochen xanion DDHE AD 211-429 Thambetochen xanion NASBP AD 30-239 Brant sp. DDHE AD 1-250 Thambetochen xanion EM BC 51-AD399 Oceanodroma castro DDHE BC 81-AD 399 Pterodroma phaeopygia DDHE BC 200-AD 450 Pterodroma phaeopygia DDH BC 341-AD 9 Thambetochen xanion DDHE BC 371-82 Med. or Large Anatid DDHE BC 367-275 Thambetochen xanion EM BC 400-170 Thambetochen xanion EM BC 761-391 Porzana ziegleri DDHE BC 763-370 Thambetochen xanion EM BC 1573-1066 Oceanodroma castro NASBP BC 1680-1410 Thambetochen xanion EM BC 1871-1522 Porzana ziegleri DDHE BC 3350-2920 Thambetochen xanion EM BC 5150-4840 Thambetochen xanion EM BC 5630-5330 Thambetochen xanion EM BC 6480-6210 Thambetochen xanion EM Age, calibrated, Avian Taxon Site No. 2 sigma * AD 1651-1950 Pterodroma phaeopygia 2706-18a AD 1549-1950 (1) Pterodroma phaeopygia 2702-7 AD 1420-1880 Pterodroma phaeopygia 9659-1 AD 1508-1950 Pterodroma phaeopygia 2706-18a AD 1390-1870 Pterodroma phaeopygia 2706-6e AD 1360-1810 Pterodroma phaeopygia 2706-18a AD 1320-1720 Pterodroma phaeopygia 27063q AD 1300-1589 Pterodroma phaeopygia 4903-C AD 1290-1720 Pterodroma phaeopygia 2706-18c AD 1280-1740 (2) Pterodroma phaeopygia 2702-7 AD 1280-1660 Pterodroma phaeopygia 2706-3f AD 1214-1520 Pterodroma phaeopygia 1753-02 AD 1200-1540 Pterodroma phaeopygia 1710-1 AD 1150-1510 Pterodroma phaeopygia 9659-1 AD 1030-1339 Pterodroma phaeopygia 4917-L 6 AD 1010-1410 Pterodroma phaeopygia 1710-1 AD 942-1172 Branta sp. 5108-1 AD 900-1159 Branta sp. 4907-D 1 AD 890-1340 Pterodroma phaeopygia 9659-1 AD 890-1040 Chaetoptila angustipluma cf. CLST. B, EU-46 AD 810-1049 Med. or Large Anatid 4907-D 1 AD 783-1016 Branta sp. 1752-60 AD 650-869 Porzana ziegleri 4907-D 1 AD 580-1060 Pterodroma phaeopygia 1710-1 AD 569-971 Pterodroma phaeopygia 1752-67 AD 440-639 Thambetochen xanion 4907-D 2 AD 420-629 Thambetochen xanion 4907-D 2 AD 211-429 Thambetochen xanion 5094-1 AD 30-239 Brant sp. 4907-D 5 AD 1-250 Thambetochen xanion CLST. F, EU-24 BC 51-AD399 Oceanodroma castro 4917-L 3 BC 81-AD 399 Pterodroma phaeopygia 4917-L 3 BC 200-AD 450 Pterodroma phaeopygia 1710-1 BC 341-AD 9 Thambetochen xanion 4907-D 5 BC 371-82 Med. or Large Anatid 4907-D 1 BC 367-275 Thambetochen xanion CLST. E, EU-15 BC 400-170 Thambetochen xanion CLST. E, EU-15 BC 761-391 Porzana ziegleri 4907-D 2 BC 763-370 Thambetochen xanion CLST. A, EU-9 BC 1573-1066 Oceanodroma castro 5108-1 BC 1680-1410 Thambetochen xanion CLST. F, EU-24 BC 1871-1522 Porzana ziegleri 4917-L 2 BC 3350-2920 Thambetochen xanion CLST. G, EU-26 BC 5150-4840 Thambetochen xanion CLST. B, EU-46 BC 5630-5330 Thambetochen xanion CLST. G, EU-26 BC 6480-6210 Thambetochen xanion CLST. B, EU-48 Age, calibrated, Avian Taxon Deposit 2 sigma * AD 1651-1950 Pterodroma phaeopygia C AD 1549-1950 (1) Pterodroma phaeopygia C AD 1420-1880 Pterodroma phaeopygia N AD 1508-1950 Pterodroma phaeopygia C AD 1390-1870 Pterodroma phaeopygia C AD 1360-1810 Pterodroma phaeopygia C AD 1320-1720 Pterodroma phaeopygia C AD 1300-1589 Pterodroma phaeopygia C AD 1290-1720 Pterodroma phaeopygia C AD 1280-1740 (2) Pterodroma phaeopygia C AD 1280-1660 Pterodroma phaeopygia C AD 1214-1520 Pterodroma phaeopygia N AD 1200-1540 Pterodroma phaeopygia N AD 1150-1510 Pterodroma phaeopygia N AD 1030-1339 Pterodroma phaeopygia N AD 1010-1410 Pterodroma phaeopygia N AD 942-1172 Branta sp. N AD 900-1159 Branta sp. N AD 890-1340 Pterodroma phaeopygia N AD 890-1040 Chaetoptila angustipluma cf. N AD 810-1049 Med. or Large Anatid N AD 783-1016 Branta sp. N AD 650-869 Porzana ziegleri N AD 580-1060 Pterodroma phaeopygia N AD 569-971 Pterodroma phaeopygia N AD 440-639 Thambetochen xanion N AD 420-629 Thambetochen xanion N AD 211-429 Thambetochen xanion N AD 30-239 Brant sp. N AD 1-250 Thambetochen xanion N BC 51-AD399 Oceanodroma castro N BC 81-AD 399 Pterodroma phaeopygia N BC 200-AD 450 Pterodroma phaeopygia N BC 341-AD 9 Thambetochen xanion N BC 371-82 Med. or Large Anatid N BC 367-275 Thambetochen xanion N BC 400-170 Thambetochen xanion N BC 761-391 Porzana ziegleri N BC 763-370 Thambetochen xanion N BC 1573-1066 Oceanodroma castro N BC 1680-1410 Thambetochen xanion N BC 1871-1522 Porzana ziegleri N BC 3350-2920 Thambetochen xanion N BC 5150-4840 Thambetochen xanion N BC 5630-5330 Thambetochen xanion N BC 6480-6210 Thambetochen xanion N Age, calibrated, Avian Taxon Source 2 sigma * AD 1651-1950 Pterodroma phaeopygia 3^ AD 1549-1950 (1) Pterodroma phaeopygia 3^ AD 1420-1880 Pterodroma phaeopygia 3* AD 1508-1950 Pterodroma phaeopygia 3^ AD 1390-1870 Pterodroma phaeopygia 3^ AD 1360-1810 Pterodroma phaeopygia 3^ AD 1320-1720 Pterodroma phaeopygia 3^ AD 1300-1589 Pterodroma phaeopygia 4 AD 1290-1720 Pterodroma phaeopygia 3^ AD 1280-1740 (2) Pterodroma phaeopygia 3^ AD 1280-1660 Pterodroma phaeopygia 3^ AD 1214-1520 Pterodroma phaeopygia 1 AD 1200-1540 Pterodroma phaeopygia 3* AD 1150-1510 Pterodroma phaeopygia 3* AD 1030-1339 Pterodroma phaeopygia 4 AD 1010-1410 Pterodroma phaeopygia 3* AD 942-1172 Branta sp. 1 AD 900-1159 Branta sp. 4 AD 890-1340 Pterodroma phaeopygia 3* AD 890-1040 Chaetoptila angustipluma cf. 2 AD 810-1049 Med. or Large Anatid 4 AD 783-1016 Branta sp. 1 AD 650-869 Porzana ziegleri 4 AD 580-1060 Pterodroma phaeopygia 3* AD 569-971 Pterodroma phaeopygia 1 AD 440-639 Thambetochen xanion 4 AD 420-629 Thambetochen xanion 4 AD 211-429 Thambetochen xanion 2 AD 30-239 Brant sp. 4 AD 1-250 Thambetochen xanion 2 BC 51-AD399 Oceanodroma castro 4 BC 81-AD 399 Pterodroma phaeopygia 4 BC 200-AD 450 Pterodroma phaeopygia 3* BC 341-AD 9 Thambetochen xanion 4 BC 371-82 Med. or Large Anatid 4 BC 367-275 Thambetochen xanion 1 BC 400-170 Thambetochen xanion 2 BC 761-391 Porzana ziegleri 4 BC 763-370 Thambetochen xanion 2 BC 1573-1066 Oceanodroma castro 1 BC 1680-1410 Thambetochen xanion 2 BC 1871-1522 Porzana ziegleri 4 BC 3350-2920 Thambetochen xanion 2 BC 5150-4840 Thambetochen xanion 2 BC 5630-5330 Thambetochen xanion 2 BC 6480-6210 Thambetochen xanion 2 Age, calibrated, Avian Taxon Radiocarbon 2 sigma * Lab. No. AD 1651-1950 Pterodroma phaeopygia Beta-11710 AD 1549-1950 (1) Pterodroma phaeopygia Beta-11195 AD 1420-1880 Pterodroma phaeopygia Beta-11192 AD 1508-1950 Pterodroma phaeopygia Beta-11709 AD 1390-1870 Pterodroma phaeopygia Beta-11707 AD 1360-1810 Pterodroma phaeopygia Beta-11708 AD 1320-1720 Pterodroma phaeopygia Beta-11706 AD 1300-1589 Pterodroma phaeopygia SR-5079 AD 1290-1720 Pterodroma phaeopygia Beta-11711 AD 1280-1740 (2) Pterodroma phaeopygia Beta-11196 AD 1280-1660 Pterodroma phaeopygia Beta-11705 AD 1214-1520 Pterodroma phaeopygia NSRL-2908 AD 1200-1540 Pterodroma phaeopygia Beta-11188 AD 1150-1510 Pterodroma phaeopygia Beta-11193 AD 1030-1339 Pterodroma phaeopygia SR-5314 AD 1010-1410 Pterodroma phaeopygia Beta-11189 AD 942-1172 Branta sp. NSRL-2855 AD 900-1159 Branta sp. SR-5307 AD 890-1340 Pterodroma phaeopygia Beta-11194 AD 890-1040 Chaetoptila angustipluma cf. NSRL-2443 AD 810-1049 Med. or Large Anatid SR-5305 AD 783-1016 Branta sp. NSRL-2861 AD 650-869 Porzana ziegleri SR-5303 AD 580-1060 Pterodroma phaeopygia Beta-11190 AD 569-971 Pterodroma phaeopygia NSRL-2860 AD 440-639 Thambetochen xanion SR-5306 AD 420-629 Thambetochen xanion SR-5310 AD 211-429 Thambetochen xanion NSRL-2857 AD 30-239 Brant sp. SR-5312 AD 1-250 Thambetochen xanion NSRL-2438 BC 51-AD399 Oceanodroma castro SR-5317 BC 81-AD 399 Pterodroma phaeopygia SR-5316 BC 200-AD 450 Pterodroma phaeopygia Beta-11191 BC 341-AD 9 Thambetochen xanion SR-5311 BC 371-82 Med. or Large Anatid SR-5309 BC 367-275 Thambetochen xanion NSRL-2402 BC 400-170 Thambetochen xanion NSRL-2404 BC 761-391 Porzana ziegleri SR-5308 BC 763-370 Thambetochen xanion NSRL-2441 BC 1573-1066 Oceanodroma castro NSRL-2854 BC 1680-1410 Thambetochen xanion NSRL-2401 BC 1871-1522 Porzana ziegleri SR-5315 BC 3350-2920 Thambetochen xanion NSRL-2400 BC 5150-4840 Thambetochen xanion NSRL-2442 BC 5630-5330 Thambetochen xanion NSRL-2408 BC 6480-6210 Thambetochen xanion NSRL-2440 * Dates for Pterodroma phaeopygia and Oceanodroma castro calibrated using marine reservoir curve with Delta r = 110 [+ or -] 80 (Dye, 1994b). (1) This provenience also produced a charcoal radiocarbon date: AD 1410-1650 (Beta-9543; Davis, 1990). (2) This provenience also produced a charcoal radiocarbon date: AD 1280-1440 (Beta-9052, Davis, 1990). Location: EM = `Ewa Marina; DDH = Deep Draft Harbor; DDHE = Deep Draft Harbor Expansion; NASBP = Naval Air Station, Barbers Point. Deposit. C = cultural deposit; N = natural deposit. Sources. 1 = Athens et al. (1999:Appendix G; XAD resin protocol used--Stafford et al., 1991). 2 = Franklin et al. (1995:Table 8.23; XAD resin protocol used). 3* = Davis (1990) for original data, with marine reservoir correction in Dye and Tuggle (1998:123). 3^ =Davis (1990) for original data, with marine reservoir correction calculated for the present table. 4 = McDermott et al. (2000:Table 4.29; XAD resin protocol used). Table 6. Radiocarbon dates on bone of Rattus exulans from sinkhole deposits on the `Ewa Plain. All samples processed using the XAD resin processing protocol (Stafford et al. 1991). Location: NASBP = Naval Air Station Barbers Point; DDHE = Deep Draft Harbor Expansion. AD Age, BP Age, Location Site No. calibrated, calibrated, 2 sigma 2 sigma AD 850-1289 1100-661 DDHE 4907-D 5 AD 892-1160 1058-790 NASBP 5108-F1 AD 972-1179 978-771 NASBP 5094-F1 AD 1190-1399 760-551 DDHE 4907-D 2 AD 1268-1401 682-549 NASBP 5108-F1 AD 1280-1409 670-541 DDHE 4917-L 2 AD 1330-1629 620-321 DDHE 4907-D 2 AD 1450-1639 500-311 DDHE 4907-D 1 AD Age, BP Age, Source Radiocarbon calibrated, calibrated, 2 sigma 2 sigma Lab. No. AD 850-1289 1100-661 2 SR-5082 AD 892-1160 1058-790 1 NSRL-2920; CAMS-25560 AD 972-1179 978-771 1 NSRL-2858; CAMS-26396 AD 1190-1399 760-551 2 SR-5080 AD 1268-1401 682-549 1 NSRL-2922; CAMS-25561 AD 1280-1409 670-541 2 SR-5085 AD 1330-1629 620-321 2 SR-5081 AD 1450-1639 500-311 2 SR-5304 Source: 1 = Athens et al. (1999); 2 = McDermott et al. (2000).
Financial support for these investigations came from the Department of the Navy, Pacific Division, Naval Facilities Engineering Command, Pearl Harbor, Hawai`i, through planning and environmental contracts with Belt Collins Hawaii (BCH). International Archaeological Research Institute, Inc., served as the BCH subconsultant for archaeological investigations at Barbers Point Naval Air Station. For their efforts to facilitate the research, we are grateful to Bruce Masse, former Navy archaeologist and now with the Los Alamos National Laboratory, and to the BCH planners John Goody (now retired) and Sue Sakai. The help of many collaborators, all named in our various contract reports, is also warmly appreciated. For discussions and information concerning rat ecology we wish to thank Atholl Anderson, Australian National University, Bruce McFadgen, Conservation Sciences Centre, New Zealand, and Alan Ziegler of Kane`ohe, Hawai`i. We are grateful to Ken Wood, National Tropical Botanical Garden, Kaua`i, for sharing with us his observations and information concerning Huelo and Mokapu Islands, and also to Steve Montgomery for arranging the extraordinary helicopter trip for Athens in 1999. We thank Dean Blinn for his help with diatoms and the identification of the algal mats, Gail Murakami for her help and consultations concerning wood identifications of charcoal samples, and Jane Tribble and Clark Sherman for their help with the geochemical characterization of the core sediments. Finally, a note of appreciation to Greg Burtchard for his helpful review of the paper, including his comments on the ecological arguments.
(1.) The chronology of Polynesian settlement of Hawai`i is a hotly debated topic with some investigators suggesting settlement occurred as early as the first century AD (Hunt and Holsen, 1991). Kirch's standard textbook on Hawaiian archaeology cites a date of AD 300 (Kirch, 1985:58, 68), and he more recently (2000:291-292) suggests a range between AD 300 and 600. However, we see the chronological data supporting a settlement range of AD 700-800 (Athens, 1997; Masse and Tuggle, 1998; Tuggle and Spriggs, 2000).
(2.) The radiocarbon dates and pollen analysis (Athens et al., 1999:69-79) together suggest the possibility of a sedimentary unconformity in the evaporite pan sequence during the middle Holocene, perhaps coinciding with a high stand at this time (see Fletcher and Jones, 1996; Grossman and Fletcher, 1998). However, more detailed investigations are needed before any conclusions can be drawn regarding this important sequence.
(3.) Similar snails at the base of Ordy Pond subsequently were dated to the Late Pleistocene--see footnote 4.
(4.) A subsequent project by the senior author under the direction of Jane Tribble (Dept. of Oceanography, Univ. of Hawai`i) raised a 17.4 m sediment column from Ordy Pond (Tribble et al., 1999). Like the evaporite pan, it appears to document a full Holocene sequence besides extending into the Late Pleistocene (14C determinations to 25,750 BP).
(5.) Two dates of 4480 [+ or -] 80 and 5030 [+ or -] 50 BP were eliminated from graphical representation because they obviously pertain to a pre-human time period; the samples likely were not charcoal but if so, they represent the burning of very old wood in the environment.
(6.) The activity is most likely confined to the latest part of this range. As indicated by Figure 7, there are only seven radiocarbon dates that fall within the pre-AD 1000 range. These have been evaluated and discussed at length elsewhere (Tuggle and Tomonari-Tuggle, 1997a; Athens and Tuggle, 2001). These dates suffer from a number of problems related to questionable methods of sample excavation and data analysis and none are considered reliable measures of cultural chronology. Reviews of the "tails" of radiocarbon sequences consistently demonstrate these difficulties (see, for example Dye, 2000a; Masse and Tuggle, 1998; Tuggle, 1997). The causes may be due to old undecayed wood in the environment, old driftwood, interior wood from long-lived species, confusion of non-cultural anaerobically blackened wood with charcoal, and questionable sample selection (see Murakami, 1992; Strong and Skolmen, 1963). In general, treatment of radiocarbon samples by archaeologists in Hawai`i has a very poor history, lacking protocols and controls commonly found elsewhere in Polynesian research (e.g., Davidson, 1992).
(7.) There is one excellent archaeological case of the bones of two extinct passeriforms occurring in a small pit or depression on the limestone surface that is interpreted as a possible offering (Tuggle and Tomonari-Tuggle, 1997a:92).
Athens, J.S. 1997. Hawaiian Native Lowland Vegetation in Prehistory. In Historical Ecology in the Pacific Islands, ed. by P.V. Kirch and T.L. Hunt. Yale University Press, New Haven, Connecticut, pp. 248-270.
Athens, J.S., M.W. Kaschko, and H.F. James 1991. Prehistoric Bird Hunters: High Altitude Resource Exploitation on Hawai`i Island. Bishop Museum Occasional Papers 31:63-84.
Athens, J.S. and H.D. Tuggle 2001. A Review of "early" radiocarbon dates from the `Ewa Plain. MS. International Archaeological Research Institute, Inc., Honolulu,
Athens, J.S., J.V. Ward, and S. Wickler 1992. Late Holocene Lowland Vegetation, O`ahu, Hawai`i. New Zealand Journal of Archaeology 14:9-34.
Athens, J.S. and J.V. Ward 1993. Environmental Change and Prehistoric Polynesian Settlement in Hawai`i. Asian Perspectives 32(2):205-223.
Athens, J.S. and J.V. Ward 1999a. Paleoclimate, Vegetation, and Landscape Change on Guam: the Laguas Core. Chapter V in Archaeological Inventory Survey of the Sasa Valley and Tenjo Vista Fuel Tank Farms, Piti District, Territory of Guam, Mariana Islands, by B. Dixon, J.S Athens, J.V. Ward, T. Mangieri, and T. Rieth,. Report prepared for Dept. of the Navy, Pacific Division, Naval Facilities Engineering Command, Pearl Harbor, Hawai`i. International Archaeological Research Institute, Inc., Honolulu, pp. 121-151.
Athens, J.S. and J.V. Ward 1999b. The Holocene Paleoenvironment of Palau. Vol. IV of, Archaeological Data Recovery for the Palau Compact Road, Babeldaob Island, Palau: Historic Preservation Investigations, Phase H. Draft final report prepared for the U.S. Army Corps of Engineers, Pacific Ocean Division, Ft. Shafter, Hawai`i. International Archaeological Research Institute, Inc., Honolulu.
Athens, J.S. and J.V. Ward 2000. Paleoenvironmental Study of Fishponds at Fort Shafter, O`ahu, Hawai`i. Appendix F in Fort Shafter: Cultural Resource Investigations at Hawai`i's First U.S. Military Post, by M.J. Tomonari-Tuggle, S. Hamilton, and K. Slocumb. Report prepared for the U.S. Army Corps of Engineers, Pacific Ocean Division, Fort Shafter, Hawai`i. International Archaeological Research Institute, Inc., Honolulu, pp. F-1 ,F-58.
Athens, J.S., J.V. Ward, H.D Tuggle, and D.J. Welch 1999. Environment, Vegetation Change. and Early Human Settlement on the `Ewa Plain: A Cultural Resource Inventory of Naval Air Station, Barbers Point, O`ahu, Hawai`i; Part III: Paleoenvironmental Investigations. Report prepared for Dept. of the Navy, Pacific Division, Naval Facilities Engineering Command, Pearl Harbor, Hawai`i, under contract with Belt Collins Hawaii, Honolulu. International Archaeological Research Institute, Inc., Honolulu.
Atkinson, I.A.E. and H. Moller 1990. Kiore, Polynesian Rat. In Handbook of New Zealand Mammals, ed. by C. M. King. Oxford University Press, Oxford, pp. 175-192.
Atlas of Hawaii 1983. Atlas of Hawaii, 2nd edition. Dept. of Geography, University of Hawaii. University of Hawaii Press, Honolulu.
Beardsley, F. 2001. Phase II, Intensive Survey and Testing, Naval Air Station Barbers Point. Report 1629.1-053101 prepared for Dept. of the Navy, Pacific Division, Naval Facilities Engineering Command, Pearl Harbor, Hawai`i. Paul H. Rosendahl, Ph.D., Inc., Hilo, Hawai`i.
Blinn, D.W. 1999 Diatoms. Appendix F in Environment, Vegetation Change, and Early Human Settlement on the `Ewa Plain: A Cultural Resource Inventory of Naval Air Station, Barbers Point, O`ahu, Hawai`i; Part III: Paleoenvironmental Investigations, by J.S. Athens, J.V. Ward, H.D Tuggle, and D.J. Welch. Report prepared for Dept. of the Navy, Pacific Division, Naval Facilities Engineering Command, Pearl Harbor, Hawai`i, under contract with Belt Collins Hawaii, Honolulu. International Archaeological Research Institute, Inc., Honolulu, pp. 237-244.
Carlquist, S. 1980. Hawaii: A Natural History (2nd Ed.). Pacific Tropical Botanical Garden, Lawa`i, Kaua`i, Hawai`i.
Campbell, D.J. 1978. The Effects of Rats on Vegetation. In The Ecology and Control of Rodents in New Zealand Native Reserves, ed. by P.R. Dingwall, I.A.E. Atkinson, and C. Hay, pp. 99-120. New Zealand Department of Lands, Survey Information Service 4.
Christensen, C.C. and P.V. Kirch 1986. Nonmarine Mollusks and Ecological Change at Barber's Point, O`ahu. Bishop Museum Occcasional Papers 26:52-80. Bishop Museum, Honolulu.
Clark, J.S. 1988. Particle Motion and the Theory of Charcoal Analysis: Source Area, Transport, Deposition, and Sampling. Quaternary Research 30:67-80.
Collins, S.L. 1995. Avifaunal Remains from the Kawailoa Site, O`ahu Island (BPBM Site 50-Oa-D6-62). Hawaiian Archaeology 4:4-16.
Cuddihy, L.W., and C.P. Stone 1990. Alteration of Native Hawaiian Vegetation: Effects of Humans, Their Activities and Introductions. University of Hawaii Cooperative National Park Resources Studies Unit. Distributed by University of Hawaii Press, Honolulu.
Davidson, J. 1992 Editorial. New Zealand Journal of Archaeology 14:5-7.
Davis, B.D. 1990. Human Settlement in Pristine Insular Environments: A Hawaiian Case Study from Barbers Point, Southwestern O`ahu. Ph.D. dissertation, Dept. of Anthropology, Univ. of Hawaii, Honolulu. University Microfilms International, Ann Arbor, Michigan.
Dye, T. 1994a. Population Treends in Hawai`i Before 1778. The Hawaiian Journal of History 28:1-20.
Dye, T. 1994b. Apparent Ages of Marine Shells: Implications for Archaeological Dating in Hawai`i. Radiocarbon 36(1):51-57.
Dye, T. 2000a. Effects of [sup.14]C Sample Selection in Archaeology: An Example from Hawai`i. Radiocarbon 42(2):203-217.
Dye, T. 2000b. A Bayesian Calibration of [sup.14]C Determinations from Two Sinkholes and a Surface Habitation Feature at Barbers Point, O`ahu. In Archaeological Data Recovery Report for the Archaeological Sites in the Proposed Barber's Point Harbor Expansion Area, Kalaeloa, Ahupua`a of Honouliuli, District of `Ewa, Island of O`ahu (TMK 9-1-14:2), by M. McDermott, D. Shideler, J. Winieski, and H.H. Hammatt. Report prepared for the Hawaii State Department of Transportation, Harbors Division, Honolulu. Cultural Surveys Hawaii, Inc., Kailua, O`ahu, Hawai`i.
Dye, T. and E. Komori 1992a A Pre-Censal Population History of Hawai`i. New Zealand Journal of Archaeology 14:113-128.
Dye, T. and E. Komori 1992b. Computer Programs for Creating Cumulative Probability Curves and Annual Frequency Distribution Diagrams with Radiocarbon Dates. New Zealand Journal of Archaeology 14:35-43.
Dye, T. and D.W. Steadman 1990. Polynesian Ancestors and Their Animal World. American Scientist 78:207-215.
Dye, T. and H.D. Tuggle 1998. Land Snail Extinctions at Kalaeloa, O`ahu. Pacific Science 52(2):111:140.
Dye, T. and H.D. Tuggle 2001. Land Snail Extinctions at Kalaeloa, O`ahu. Internet Archaeology, Vol. 10. http://intarch.ac.uk/journal/issue10/dye.
Fletcher, C.H., III, and A.T. Jones 1996. Sea-Level Highstand Recorded in Holocene Shoreline Deposits on Oahu, Hawaii. Journal of Sedimentary Research 66(3):632-641.
Franklin, L.J., S.T. Goodfellow, and W. Wulzen 1995. Final report, Phase II--Data Recovery Archaeological Mitigation Program, Ewa Marina Community Project, Volume I. Prepared for Haseko (Ewa), Inc., c/o Belt Collins Hawaii. Paul H. Rosendahl, Ph.D., Inc., Hilo, Hawai`i.
Gagne, W.C. and L.W. Cuddihy 1990. Vegetation. In Manual of the Flowering Plants of Hawai`i, Vol. 1, by W.L. Wagner, D.R. Herbst, and S.H. Sohmer. University of Hawaii Press and Bishop Museum Press, Honolulu, pp. 45-114.
Grossman, E.E. and C.H. Fletcher, III. 1998. Sea Level Higher Than Present 3500 Years Ago on the Northern Main Hawaiian Islands. Geology 26(4):363-366.
Hammatt, H.H. and D.W. Shideler 1995. Data Recovery Plan for Archaeological Sites in the Proposed Barber's Point Expansion Area. Prepared for the Estate of James Campbell. Cultural Surveys Hawai`i, Inc., Kailua.
Hunt, T.L. and R.M. Holsen 1991. An Early Radiocarbon Chronology for the Hawaiian Islands: A Preliminary Analysis. Asian Perspectives 30(1): 147-161.
James, H.F. 1995. Prehistoric Extinctions and Ecological Changes on Oceanic Islands. Ecological Studies 115:87-102.
James, H.F. and D.A. Burney 1997. The Diet and Ecology of Hawaii's Extinct Flightless Waterfowl: Evidence from Coprolites. Biological Journal of the Linnean Society 62:279-297.
James, H.F., T.W. Stafford, D.W. Steadman, S.L. Olson, P.S. Martin, A.T. Jull, and P.C. McCoy 1987. Radiocarbon Dates on Bones of Extinct Birds from Hawai`i. Proceedings of the National Academy of Science 84:2350-2354.
James, H.F. and S.L. Olson 1991. Descriptions of Thirty-two New Species of Birds from the Hawaiian Islands: Part II. Passeriformes. Ornithological Monographs 46:1-88.
Kirch, P.V. 1982a. The Impact of the Prehistoric Polynesians on the Hawaiian Ecosystem. Pacific Science 36(1):1-14.
Kirch, P.V. 1982b. Transported Landscapes. Natural History 91(12):32-35.
Kirch, P.V. 1985. Feathered Gods and Fishhooks: an Introduction to Hawaiian Archaeology and Prehistory. University of Hawai`i Press, Honolulu.
Kirch, P.V. 2000. On the Road of the Winds: An Archaeological History of the Pacific Islands Before European Contact. University of California Press, Berkeley.
Kramer, Raymond J. 1971. Hawaiian Land Mammals. Charles E. Tuttle Co., Tokyo, Japan.
Loope, L.L. and D. Mueller-Dombois 1989. Characteristics of Invaded Islands, with Special Reference to Hawaii. In Biological Invasions: A Global Perspective, ed. by J.A. Drake, F. DiCastri, R.H. Groves, F.M.J. Kruger, M. Rejmanek, and M. Williamson. John Wiley and Sons, Ltd., Chichester, United Kingdom, pp. 257-280.
Lorence, D.H. and K.R. Wood 1994. Kanaloa, a New Genus of Fabaceae (Mimosoideae) from Hawaii. Novon 4(2):137-145.
Masse, W.B. and H.D. Tuggle 1998. The Date of Hawaiian Colonization. In Easter Island in Pacific Context South Seas Symposium: Proceedings of the Fourth International Conference on Easter Island and East Polynesia, ed. by C.M. Stevenson. G. Lee, and F.J. Morin. Easter Island Foundation Occasional Paper 4, Easter Island Foundation, Bearsville and Cloud Mountain Presses, Los Osos, California, pp. 229-235.
McDermott, M., D. Shideler, J. Winieski, and H.H. Hammatt 2000. Archaeological Data Recovery Report for the Archaeological Sites in the Proposed Barber's Point Harbor Expansion Area. Kalaeloa, Ahupua`a of Honouliuli, District of `Ewa, Island of O`ahu (TMK 9-1-14:2). Report prepared for the Hawaii State Department of Transportation, Harbors Division, Honolulu. Cultural Surveys Hawaii, Inc., Kailua, O`ahu, Hawai`i.
Moniz, J.J. 1997. The Role of Seabirds in Hawaiian Subsistence: Implications for Interpreting Avian Extinction and Extirpation in Polynesia. Asian Perspectives 36(1):27-50.
Moore, P.D., J.A. Webb, and M.E. Collinson 1991. Pollen Analysis. Blackwell, London.
Morrison, K.D. 1994. Monitoring Regional Fire History Through Size-Specific Analysis of Microscopic Charcoal: The Last 600 Years in South India. Journal of Archaeological Science 21:675-685.
Mosby, J.M., K. Wodzicki, and H.R. Thompson 1973. Food of the Kimoa (Rattus exulans) in the Tokelau Islands and Other Habitats of the Pacific. New Zealand Journal of Science 16:799-810.
Murakami, G.M. 1992. Identification of Charcoal from Kahoolawe. Appendix H in Kahoolawe Excavations, 1982-83 Data Recovery Project, Island of Kahoolawe, Hawaii, by P.H. Rosendahl, A.E. Haun, J.B. Halbig, M. Kaschko, and M.S. Allen. Report 48-011592 prepared for Dept. of the Navy, Pacific Division, Naval Facilities Engineering Command, Pearl Harbor, Hawai`i. Paul H. Rosendahl, Ph.D., Inc., Hilo, Hawai`i.
Norman, F.I. 1975. The Murine Rodents Rattus rattus, exulans, and norvegicus as Avian Predators. Atoll Research Bulletin 182:1-13.
Oison, S.L. and H.F. James 1982. Prodromus of the Fossil Avifauna of the Hawaiian Islands. Smithsonian Contributions to Zoology No. 365. Smithsonian Institution Press, Washington, D.C.
Olson, S.L. and H.F. James 1991. Descriptions of Thirty-two New Species of Birds from the Hawaiian Islands: Part I. Non-passeriformes. Ornithological Monographs 45:1-88.
Patterson, W.A., J.J. Edwards, and D.J. Maguire 1987. Microscopic Charcoal as a Fossil Indicator of Fire. Quaternary Science Review 6:3-23.
Sinoto, A. 1976. A Report on Cultural Resources Survey at Barbers Point, Island of Oahu. Prepared for the Department of the Army, Corps of Engineers, Pacific Ocean Division. Department of Anthropology, B.P. Bishop Museum, Honolulu.
Sinoto, A. 1978. Archaeological and Paleontological Salvage at Barbers Point, Oahu. Department of Anthropology, Technical Report (ms. 030178). B.P. Bishop Museum, Honolulu.
Spriggs, M. and A. Anderson 1993. Late Colonization of East Polynesia. Antiquity 67:200-217.
Stafford, T.W., Jr., P.E. Hare, L. Currie, A.J.T. Jull, and D.J. Donahue 1991. Accelerator Radiocarbon Dating at the Molecular Level. Journal of Archaeological Science 18:35-72.
Steadman, D.W. 1995. Prehistoric Extinctions of Pacific Island Birds: Biodiversity Meets Zooarchaeology. Science 267:1123-1130.
Stolk, A., T.E. Tornqvist, K.P.V. Hekhuis, H.J. Berendsen, and J. Van der Plicht 1994. Calibration of [sup.14]C Histograms: A Comparison of Methods. Radiocarbon 36(1):1-10.
Strecker, R.L. and W.B. Jackson 1982. Habitats and Habits. In Pacific Island Rat Ecology, ed. by T.I. Storer. Bishop Museum Bulletin 225. Bernice P. Bishop Museum Press, Honolulu, pp. 64-73.
Strong, C.C. and R. G. Skolmen 1963. Origin of Drift-logs on the Beaches of Hawaii. Nature 197:890.
Temme, M. 1979. Polynesian Rat (Rattus exulans) in the Northern Marshall Islands. Ph.D. dissertation, Bowling Green State University, Bowling Green, Ohio.
TenBruggencate, J. 1997. Team Nurses Last of Dying Species. The Honolulu Advertiser, Aug. 16, 1997.
Tomich, Q. 1986. Mammals in Hawai`i. A Synopsis and Notational Bibliography. Bishop Museum Special Publication 57. B.P. Bishop Museum, Honolulu.
Tribble, J.S., G.H. Garrison, J.S. Athens, C.B. Allen, C.G Skilbeck, and E. Frankel 1999. Evidence for Early-Mid Holocene Sea-level on O`ahu, Hawai`i from Coastal Pond Sediments. In The Non-steady State of the Inner Shelf and Shoreline: Coastal Change on the Time Scale of Decades to Millennia in the late Quaternary, ed. by C.H. Fletcher and J.V. Matthews,. PDF version of Abstracts with Programs, Inaugural Meeting of IGCP Project #437, "Coastal Environmental Change During Sea Level Highstands," University of Hawaii, Honolulu, Nov. 9-12, 1999, pp. 184-189.
Tuggle, H.D. 1997. The `Ewa Plain. Hawaiian Archaeology 6:8-36.
Tuggle, H.D. and M. Spriggs 2000. The Age of the Bellows Dune Site 018, O`ahu, Hawai`i, and the Antiquity of Hawaiian Colonization. Asian Perspectives 39(1-2): 165-188.
Tuggle, H.D. and M.J. Tomonari-Tuggle 1997a. Synthesis of Cultural Resource Studies of the `Ewa Plain. Report prepared for Dept. of the Navy, Pacific Division, Naval Facilities Engineering Command, Pearl Harbor, Hawai`i, under contract with Belt Collins Hawaii, Honolulu. International Archaeological Research Institute, Inc., Honolulu.
Tuggle, H.D. and M.J. Tomonari-Tuggle 1997b. A Cultural Resource Inventory of Naval Air Station, Barbers Point, O`ahu, Hawai`i: Part I: Phase I Survey and Inventory Summary. Report prepared for Dept. of the Navy, Pacific Division, Naval Facilities Engineering Command, Pearl Harbor, Hawai`i, under contract with Belt Collins Hawaii, Honolulu. International Archaeological Research Institute, Inc., Honolulu.
Vancouver, G. 1798. A Voyage of Discovery to the North Pacific Ocean, and Round the World ... Performed in the Years 1790-1795 (3 vols.). Robinsons and Edwards, London.
Wagner, W.L., D.R. Herbst, and S.H. Sohmer1990. Manual of the Flowering Plants of Hawai`i (2 vols.). University of Hawaii Press and Bishop Museum Press, Honolulu.
Wickler, S.K. and H.D. Tuggle 1997. A Cultural Resource Inventory of Naval Air Station, Barbers Point, O`ahu, Hawai`i; Part II: Phase II Inventory Survey of Selected Sites. Report prepared for Dept. of the Navy, Pacific Division, Naval Facilities Engineering Command, Pearl Harbor, Hawai`i, under contract with Belt Collins Hawaii, Honolulu. International Archaeological Research Institute, Inc., Honolulu.
Wirtz, W.O. 1972. Population Ecology of the Polynesian Rat, Rattus exulans, Kure Atoll, Hawai`i. Pacific Science 26:433-461.
Wodzicki, K. 1978/79. Relationships Between Rats and Man in the Central Pacific. Ethnomedicine V:433-446.
JSA, HDT, DJW: International Archaeological Research Institute, Inc., 2081 Young St., Honolulu, HI 96826-2231, USA. Email: firstname.lastname@example.org
JVW: Pacific Palynology, PO Box 14, Sweet Briar, VA 24595, USA. Email: email@example.com
|Printer friendly Cite/link Email Feedback|
|Author:||Athens, J. Stephen; Tuggle, H. David; Ward, Jerome V.; Welch, David J.|
|Publication:||Archaeology in Oceania|
|Article Type:||Statistical Data Included|
|Date:||Jul 1, 2002|
|Previous Article:||Pacific Archaeology: assessments and prospects.|
|Next Article:||Holocene scraper reduction, technological organization and landuse at Ingaladdi rockshelter, Northern Australia.|