Dryland agricultural expansion and intensification in Kohala, Hawai'i island.

Dryland agricultural expansion and intensification in Kohala, Hawai'i island. Throughout the Hawaiian archipelago, dryland agricultural fieldsystems constituted a significant component of the late prehistoricsubsistence economy A subsistence economy is an economy in which a group generally obtains the necessities of life, but do not attempt to accumulate wealth. In such a system, a concept of wealth does not exist, and only minimal surpluses generally are created, therefore there is a reliance on renewal . Of the various dryland systems, the leeward or westside of the island of Hawai'i is notable for the three largesystems of Kohala, Kona, and Waimea-Lalamilo [ILLUSTRATION FOR FIGURE 1OMITTED]. The Kohala and Kona areas were archaeologically documentedduring the late 1960s and early 1970s by the University of Hawai'iand the Bishop Museum of Honolulu (Barrera 1971; Bevacqua 1972; Ching1971; Crozier croziersee crosier. 1971; Newman 1970; 1972; Pearson 1968; Rosendahl 1972a;1972b; 1994; Soehren & Newman 1968; Tuggle & Griffin 1973; seealso Kelly 1983; Schilt 1984). The Waimea-Lalamilo field system wassurveyed and tested in the 1970s and 1980s during a contract archaeologyproject undertaken by the Bishop Museum of Honolulu (Clark 1981; 1986;1987; Clark & Kirch 1983). These studies have suggested that thefield systems produced large quantities of food to support local farmersand residents as well as local and district-level chiefly elites. It isgenerally thought that the dryland agricultural systems had spread totheir maximum extent, nearly reaching the edge of productive lands. Andit is presumed that these systems had approached the limits ofintensification, with smaller and smaller gains in agriculturalproduction relative to new labour inputs. The focus of this paper is toevaluate these hypotheses in relation to the Kohala field system using ageographic information system geographic information system (GIS)Computerized system that relates and displays data collected from a geographic entity in the form of a map. The ability of GIS to overlay existing data with new information and display it in colour on a computer screen is used primarily to (GIS) to analyse the spatial role ofenvironmental variables on agricultural development. We also considerwhether social processes are implicated in the extent and intensity towhich dryland agricultural production was developed in Hawai'i.The Kohala field systemThe 19x4-km Kohala field system is situated on the leeward-facing anddrier slopes of the Kohala mountains on the northwestern portion ofHawai'i Island [ILLUSTRATION FOR FIGURE 2 OMITTED]. Thearchaeological features associated with the system include many rock andearthen walls and alignments, terraced garden areas, planting andclearing mounds, enclosures, religious features, trails, as well astemporary and permanent residential features (Kirch 1985; 1994; Newman1970; Rosendahl 1972; 1994). Sweet potato sweet potato,trailing perennial plant (Ipomoea batatas) of the family Convolvulaceae (morning glory family), native to the New World tropics. Cultivated from ancient times by the Aztecs for its edible tubers, it was introduced into Europe in the 16th cent. (Ipomoea Ipomoeawidespread genus of poisonous vines of the family Convolvulaceae; may contain various toxins including the indole alkaloid lysergic acid, furanoterpenes, indolizidine alkaloids (swainsonine). Includes I. asarifolia (salsa), I. batatas Ba`ta´tasn. 1. An aboriginal American name for the sweet potato (Ipomæa batatas). ) wasapparently the main cultigen cul��ti��gen?n.An organism, especially a cultivated plant, such as a banana, not known to have a wild or uncultivated counterpart.[culti(vated) + -gen. of the field system, with other plants -dryland taro taro:see arum. taroHerbaceous plant (Colocasia esculenta) of the arum family, probably native to Southeast Asia and taken to the Pacific islands. (Colocasia esculenta), yams (Dioscorea spp.), bananas (Musahybrids), sugarcane (Saccharum officinarum), breadfruit breadfruit:see mulberry. breadfruitFruit of either of two closely related trees belonging to the mulberry family. Artocarpus communis (also called A. incisa or A. altilis) provides a staple food of the South Pacific. (Artocarpusaltilis), paper mulberry (Broussonetia papyrifera - on the field edgesor at upper elevations (Newman 1970: 119; Rosendahl 1994: 63-4). Thelong axes of fields throughout the Kohala system were oriented parallelto the elevational contours, and the walls would have functioned aswindbreaks from the trade winds which sweep down the slopes of theKohala mountains (Newman 1970: 28, 143). Configured in this way, thewalls would have reduced evapotranspiration evapotranspirationLoss of water from the soil both by evaporation from the soil surface and by transpiration from the leaves of the plants growing on it. Factors that affect the rate of evapotranspiration include the amount of solar radiation, atmospheric vapor pressure, and - with heavy mulching -retained essential moisture for the crops (Smith & Schilt 1973:314). This alignment of fields also conserved water by retaining anddispersing surface run-off, and inhibited wind erosion wind erosionn → erosi��n f del vientoand soil creep(Rosendahl 1994: 35).Models for intensification at KohalaField systems such as those at Kohala are the remnants of formeragricultural practices and processes. Throughout Polynesia, butespecially in Hawai'i, anthropologists have attempted to identifyand reconstruct the development of traditional and prehistoricagriculture (Barrau 1961; Earle 1978; 1980; Handy 1940; Handy &Handy 1972; Sahlins 1958; Yen 1971; 1973; Yen et al. 1972). Agriculturalchange in prehistoric Hawai'i is generally viewed as the outcome ofadaptation, expansion and intensification (Kirch 1985: 216; Tuggle &Tomonari-Tuggle 1980): adaptation is the process whereby cultigens andcultivation practices are introduced and adjusted to local conditions;expansion involves increasing the area covered by an establishedagricultural complex; intensification refers to increased inputs inorder to achieve increased agricultural output (Kirch 1994: 19).Morrison (1994: 115) notes that intensification usually involves anincreased input of capital, labour or skill, in relation to a constantplot of land. Labour investment can take a number of different forms;one is to increase mulching and shorten the fallow fallowa pale cream, light fawn, or pale yellow coat color in dogs. period in order toincrease the cropping cycle. Alternatively, labour investment caninvolve capital investment on the land (Kirch 1994: 19), where permanentagricultural infrastructure such as terraces, irrigation irrigation,in agriculture, artificial watering of the land. Although used chiefly in regions with annual rainfall of less than 20 in. (51 cm), it is also used in wetter areas to grow certain crops, e.g., rice. canals oralignments are constructed. These modifications of the landscape - theresult of labour investment - in turn create capital and demarcate de��mar��cate?tr.v. de��mar��cat��ed, de��mar��cat��ing, de��mar��cates1. To set the boundaries of; delimit.2. To separate clearly as if by boundaries; distinguish: demarcate categories. userights by successive generations (Brookfield 1972; Kirch 1994; Lansing1991; Morrison 1994).We distinguish here the interpretive investigation of intensificationas a global process from analytical studies. Interpretations ofintensification regard increases in population size and density, effortsto alleviate dietary risk and the tributary support of elite socialgroups, as the mechanisms spurring agricultural change (Morrison 1994:118-30). Population increase within a group's territory and chieflydemands for surplus production are reasons commonly suggested forintensification at Kohala (Kirch 1984: 164; 1985: 235; 1994: 261-8;Tuggle & Griffin 1973: 63). Kirch (1984: 164; 191-2) and Hommon(1986: 65-7) treat sociopolitical so��ci��o��po��li��ti��cal?adj.Involving both social and political factors.sociopoliticalAdjectiveof or involving political and social factors factors, especially in the lateprehistoric period, as the proximate proximate/prox��i��mate/ (prok��si-mit) immediate or nearest. prox��i��mateadj.Closely related in space, time, or order; very near; proximal.proximateimmediate; nearest. variable in Hawaiian agriculturalintensification. Still, the Boserup (1965) model as modified in thePacific by the geographer Brookfield (1972; 1984; 1986), with itsemphasis on population pressure as the ultimate mechanism, is thought tospur the development of intensified agriculture in the archipelago(Kirch 1984: 164; Hommon 1986: 65). Analytical studies, such as thisone, focus on variation in intensification as the result of differentpatterns of interacting variables.Chronology and hypotheses of agricultural development at KohalaTwo chronologies have been identified for the Kohala field system;both are based for the most part on the archaeological research from asingle ahupua'a (the traditional term for a Hawaiian territoryassociated with a community) called Lapakahi. From radiocarbon andvolcanic glass volcanic glassAny glassy rock formed from lava or magma that has a chemical composition close to that of granite. Such molten material may reach very low temperatures without crystallizing, but its viscosity may become very high. dates, Rosendahl originally (1972; subsequently Kirch1984: 181-92; 1985; 1990a; 1994: 258-9) suggested that the initialexpansion of agriculture into the uplands surrounding the Kohala coasttook place AD 1300-1500 through a strategy of shifting cultivation,likely involving regular burning of primary and secondary vegetation.The main development of the Kohala field system took place AD 1450-1800.By either AD 1600 (Kirch 1990a: 333) or the late 1600s (Kirch 1984: 189)the lateral expansion of the field system had been reached, and by AD1800 the system was highly intensified. Kirch proposes that the processof intensification involved shortened fallow periods, and agriculturalplots divided into successively smaller units. He (1994: 258) goes on tosuggest the system was abandoned shortly after European contact in thelate 18th or early 19th century.Recently, Rosendahl (1994:20-22) has modified his earlier views onthe chronology, perhaps because of unreliability of volcanic glass datesin Hawai'i (Graves & Ladefoged 1991a; 1991b). Still emphasizinga late expansion, Rosendahl now places the development of bounded fieldsno earlier than the 16th century and has intensified dryland cultivationin Kohala continuing well into the contact period, perhaps as late as AD1850. Archaeological research at the Waimea-Lalamilo system southeast ofKohala documents its use well into the 19th century (Clark 1988; Graves& Ladefoged 1991b).Previous research concluded that within this period of 300-500 years,the Kohala dryland field system had reached its extensive and intensivelimits. Rainfall is thought to have been the primary variable limitingthe extensive limits of the system (Kirch 1984; 1985; 1994; Murabayashi1970; Newman 1970: 143; Rosendahl 1994: 20; Smith & Schilt 1973),with agriculture being restricted to areas receiving more than c. 500 mm(20 ins.) of rain annually (Newman 1970: 117). Most of the rain in theKohala district falls between November and April, and in the lowerelevations there would have been a 5-7-month period of difficult farmingdue to lack of water, higher temperatures and greater evapotranspiration(Murabayashi 1970: 261-3). Kirch (1984: 188) cites several attempts toexpand agricultural production below the 20-inch (500 mm) rainfallisohyetal on Hawai'i Island, none appearing successful. Soildistribution is also thought to have curbed the expansion of the fieldsystem (Kirch 1985: 233-4; Murabayashi 1970), although few details areprovided. Finally, elevation has been suggested as limiting the upperextent (Newman 1970), as soil temperatures and the amount of sunlightdecrease at higher elevations due to orographic o��rog��ra��phy?n.The study of the physical geography of mountains and mountain ranges.oro��graph clouds; both wouldaffect the sweet potato.Similar evidence is used to conclude the field system was intensifiedduring approximately the same period of time. Newman (1970: 152)apparently originated this idea when he suggested 'It would appearthat nearly all usable land (on Hawai'i Island) was under the mostintense cultivation possible, given the limitations of environment andculture.' Both Tuggle & Tomonari-Tuggle (1980: 311) and Kirch(1984: 191; 1994: 265) have reiterated the position that the limits ofintensification had been approached or reached. Kirch (1994: 265)recently proposed that the level of intensification within the Kohalafield system 'had approached the limits of increased productivityeven with significant labor inputs'.A GIS approach to variables and limits at KohalaNeither hypothesis concerning limits to the expansion orintensification of the Kohala field system has been adequately tested,nor have the environmental factors been systematically evaluated. Inthis paper we use a GIS to model environmental contexts of the northwestportion of Hawai'i Island in relation to the system of earthen andstone walls. Lacking temporal information on the geographic spread andinfilling of the system we do not model the process of expansion orintensification (although see Ladefoged & Graves 1996 for an attemptat modelling a relative chronology of the field system). Rather weconsider its endpoint to determine whether the geographic limits toexpansion and intensification had in fact been reached. We also ask ifvariability in the density of walls and alignments - our measure ofintensification - was environmentally conditioned. Additionally, weexamine one expression of social factors in agriculturalintensification: the distribution of the field system across the variousahupua'a territories in Kohala.If the margins of the system - measured by the occurrence of rockalignments and walls - correspond with particular parameters (generally,quantitative limits) of the environmental variables, then it is likelythat agricultural expansion had been reached in terms of that variable.If it does not, then it is unlikely that variable had constrainedfurther expansion.To assess the level of intensification, we follow Kirch's (1994)lead and use differences in the scale of infrastructural improvements.We take the total length of rock and earthen walls in each hectare ofthe field system as a measure of labour input. Areas of highintensification will have a greater length of walls per hectare thenareas of low intensification. The environmental characteristicsassociated with different levels of intensification are compared. Wewill conclude that the limits to intensification had been reached if theenvironmental characteristics of higher intensification areas are bothdifferent and better suited to the primary subsistence crop, sweetpotatoes, than lower intensification areas. If areas associated withdifferent levels of intensity have similar environments, it is likelythat further intensification would have been possible. To measureintensification potential we calculate the relative proportions of areasassigned to different categories (or intervals) of environmentalvariables for each intensity level.The data for our analysis consist of an archaeological base-map ofthe field system, historic territorial boundaries and five environmentalvariables: rainfall, elevation, slope, aspect, and soil. Rainfall datawere collected by the National Weather Service at three climatological cli��ma��tol��o��gy?n.The meteorological study of climates and their phenomena.clima��to��log stations (Armstrong 1983), and soil classifications were made by theUnited States Department of Agriculture United States Department of Agriculture (USDA),n.pr established in 1862, USDA is responsible for the safety of meat, poultry, and egg products. It conducts ongoing research in areas from human nutrition to new crop technologies and also helps ensure open (1973). These two data-sets weredigitized and coded within a vector GIS (pc Arc/Info and ArcView 2).Elevation data came from the United States Geological Survey The United States Geological Survey (USGS) is a scientific agency of the United States government. The scientists of the USGS study the landscape of the United States, its natural resources, and the natural hazards that threaten it. as adigital elevation model A digital map of the elevation of an area on the earth. The data are either collected by a private party or purchased from an organization such as the U.S. Geological Survey (USGS) that has already undertaken the exploration of the area. (DEM See digital elevation model. ) (down-loaded via the Internet) with ahorizontal resolution The number of elements, dots or columns from left to right on a printed page, display screen or fixed area such as one inch. Contrast with "vertical resolution," which is the number of rows, dots or lines from top to bottom. of 30 m and a vertical resolution of I m. Weanalysed the DEM with a raster GIS (IDRISI 4.1) to produce slope andaspect data. These data-sets have a spatial resolution (Data West Research Agency definition: see GIS glossary.) A measure of the accuracy or detail of a graphic display, expressed as dots per inch, pixels per line, lines per millimeter, etc. It is a measure of how fine an image is, usually expressed in dots per inch (dpi). of 30 m and wereclassified by 1 [degrees] increments. For analytical purposes theelevation, slope and aspect data were converted to vector format In CAD and computer graphics, a vector format is used to represent the vector graphics on the screen.This approach was used earlier in the 1970s when cathode ray tubes would draw line segments representing edges of a BREP. with aspatial resolution of 90 m.During the late prehistory prehistory,period of human evolution before writing was invented and records kept. The term was coined by Daniel Wilson in 1851. It is followed by protohistory, the period for which we have some records but must still rely largely on archaeological evidence to and early history of Hawai'i, aseries of localized communities held territories called ahupua'a(Handy & Handy 1972: 48; Hommon 1986; Kirch 1985). Ideally theseunits formed wedge-shaped territories that extended from the coast tothe interior uplands. A number of contiguous ahupua'a wereintegrated into larger districts (moku) ruled by a political elite. Atdifferent times, districts were integrated into island-wide, or evenmultiple-island, polities. The boundaries of ahupua'a wereoccasionally marked by cairns and stone walls. In Kohala, somearchaeological walls mark the boundaries of ahupua'a, but these arenot found throughout the whole area. To approximate the location of theahupua'a boundaries, we incorporated into the GIS the ahupua'aboundaries indicated on contemporary USGS USGS United States Geological Survey (US Department of the Interior)7.5 minute topographic mapsfor Hawai'i which derive from early historical accounts[ILLUSTRATION FOR FIGURE 2 OMITTED].The archaeological base-map of the Kohala field system, originallyproduced by Myra Tomonari-Tuggle (n.d.) from aerial photographs, wasdigitized and incorporated into the GIS. Tomonari-Tuggle's map, notproduced using photogrammetric techniques, contains a degree of spatialdistortion and a number of smaller alignments are not shown on this map.When we compared the GIS base map with two detailed maps (Newman 1970;Rosendahl 1994) of archaeological features (including walls within thefield system) from a portion of Lapakahi ahupua'a, we found someshorter alignments are not depicted. Newman (1970: 138) found, whenusing aerial photographs for his map, 'Archaeological featuresmeasuring less than about 10 feet (3 metres) across or less than 3 feet(1 metre) high cannot always be seen'. On the Tomonari-Tuggle map,alignments less than c. 20-5 m in length were often omitted.TABLE 1. Intensification level categories.level of total length percentageintensification of rock align- of field ments per system hectare (m)high (H) 400-614 6medium-high (M-H) 300-400 1 9medium (M) 200-300 2 4medium-low (M-L) 100-200 2 6low (L) 2-100 2 5The archaeological map of the Kohala field system depicts over 5400segments of rock alignments and walls with a total length of just over800 km [ILLUSTRATION FOR FIGURE 2 OMITTED]. The fields begin near thenorth tip of Hawai'i very close to the coast. The western marginextends southward south��ward?adv. & adj.Toward, to, or in the south.n.A southward direction, point, or region.south at an increasing distance from the coast, with theeastern margin at a higher elevation and also an increasing distancefrom the coast. From north to south the field system is more than 19 kmin length. At its maximum it is more than 4 km in width. At an estimated40 sq. km in total area, it is one of the largest archaeological sitesin Polynesia.The level of intensification was calculated by the total length ofwalls within each 1-ha grid. Nearly 4000 (3969) ha in Kohala containrock or earthen walls, with the total length of the walls per hectareranging from 2 to as much as 614 m. The grid squares were classifiedinto five categories: high (H), medium-high (M-H), medium (M),medium-low (M-L), and low (L) (TABLE 1, [ILLUSTRATION FOR FIGURE 3OMITTED]). Areas of low intensity are not emphasized in this analysisbecause in some instances they result from an edge effect of the grid.Between the zones of intensified agriculture are areas that lackalignments. These areas may never have been used, or more likely in thenorthern sections of the study area, the walls once occurring there havebeen destroyed by historic activities.We define a final region in Kohala that we refer to as the'study area', that takes in a 2-km wide buffer around the edgeof the field system, the area within the field system proper, and theareas within the field system with no alignments. We use theenvironmental characteristics of the study area as a control.The field-system and the environmental variablesAn indication of the extensive and intensive limits of the fieldsystem is developed from the five environmental variables.(1)RainfallPrecipitation in Hawai'i is influenced by the dominant northeasttrade winds and the topography of the area. Annual rainfall[ILLUSTRATION FOR FIGURE 4 OMITTED] in the study area ranges from lessthan 250 mm (10 ins.) per year to 3175 mm (125 ins.). When the fieldsystem is plotted against the rainfall map it falls within the460-1400-mm (18-55-in,) rainfall band. Its southwestern boundary isdefined by the 460-mm (18-in.) rainfall isohyetal, a parameter valuesimilar to previous estimates (Kirch 1985; 1994; Murabayashi 1970: 257;Newman 1970; Rosendahl 1972; 1994; Smith & Schilt 1973: 312).The proportion of areas failing within each rainfall band was graphedfor each level of intensification as well as the study area[ILLUSTRATION FOR FIGURE 5 OMITTED]. An indication that the field systemhad reached its extensive limits is suggested by the marked differencebetween the study area proportions and the intensity level proportions.Much of the highly intensified zones are in the 510-760-mm (20-30-in.),760-1020-mm (30-40-in.) and 1020-1270-mm (40-50-in.) rainfall bands, andthe M-H to M-L intensity areas show a similar pattern of rainfalldistribution. This similarity between intensity levels suggests that thelimits to intensification had not been reached in terms of rainfall.ElevationElevation in the larger study area ranges from sea level to 1410 mabove sea level [ILLUSTRATION FOR FIGURE 6 OMITTED]. The field system ison the leeward slope of the Kohala mountains, the dominant topographicfeature of north Hawai'i, separated by the ridge from the windwardside. The upper extent of the field system, especially along thesoutheastern boundary, commonly falls at an elevation of 800 m (with afew walls higher up), a relationship shown graphically in FIGURE 7.There is a lower proportion of H, M-H and M intensity areas above the700-800-m zone, and virtually no features above the 900-1000-m zone.Below the 200-300-m zone, there are fewer areas of H and M-H intensitythan would be expected based on the distribution of such lands in thestudy area. These differences suggest that the elevation limits toexpansion occur at about 800 m and just below 200 m.The hypothesis that elevation was implicated in establishing limitsto intensification is tested by comparing the distributions of H and M-Hintensity areas with those of M and M-L areas. These distributionsshould be different, corresponding to agricultural productivity Agricultural productivity is measured as the ratio of agricultural inputs to agricultural outputs. While individual products are usually measured by weight, their varying densities make measuring overall agricultural output difficult. . Thedistribution of H and M-H intensity areas are roughly bimodal bi��mod��al?adj.1. Having or exhibiting two contrasting modes or forms: "American supermarket shopping shows bimodal behavior forelevation: about 40% in lower, 20% in middle and about 40% in upperelevations. The M and M-L intensity areas are also bimodal indistribution, but less strongly. This pattern only weakly fulfils thefirst criterion for demonstrating that intensification was approachingan environmental limit. It does not fulfil the second at all, as thereis no reason to suppose that the middle elevations are less productivefor growing sweet potato. The bimodality Bimodality is the simultaneous use of two distinct pitch collections. It is more general than bitonality since the "scales" involved need not be traditional scales; if diatonic collections are involved, their pitch centers need not be the familiar major and minor-scale tonics. more likely results fromdifferential intensification within ahupua'a (see below), and weconclude that the elevational data do not suggest limits ofintensification had been reached.AspectA ridge of the Kohala mountains oriented northwest-southeast dictatesaspect (or direction of orientation) in the study area [ILLUSTRATION FORFIGURE 8 OMITTED]. The field system, located on the southwest leewardside Noun 1. leeward side - the side sheltered from the windto leewardleeward - the direction in which the wind is blowing with its northeastern margin largely on the ridge-line, ispredominately located in areas with aspects of 190-300 [degrees][ILLUSTRATION FOR FIGURE 9 OMITTED]. The study area has higherpercentages in the northeasterly north��east��er��ly?adj.1. Situated toward the northeast.2. Coming or being from the northeast.north��east facing aspects, reflecting the effectof the ridge. As there is little difference in aspect between H and M-Hintensity areas and M and M-L intensity areas, we conclude thatintensification was not limited by some parameter of this environmentalvariable.SlopeSmall gulches and volcanic cinder cones create local variation inslope throughout the study area. Most of the area, whether underagriculture or not, has a slope of 3-9 [degrees]. In FIGURE 10, steepareas are shown in darker shades, more gentle slopes in lighter. Otherthan the southeast section of the field system where volcanic conesrestrict the construction of walls, slope does not appear to direct theexpansion of the Kohala field system. In FIGURE 11, displaying theproportion of area represented by each slope interval, areas withdifferent intensity levels have distributions similar to one another andto the entire study.SoilNine different soils are found in the study area [ILLUSTRATION FORFIGURE 12 OMITTED]. The most common soils, Pu'u Pa and Kahana, area well-drained stony sandy loam and a stony silty silt?n.A sedimentary material consisting of very fine particles intermediate in size between sand and clay.v. silt��ed, silt��ing, siltsv.intr. clay (USDA USDA,n.pr See United States Department of Agriculture. 1973).Kawaihae, the next most frequent soil, is an excessively porous stonysilty loam. The margin of the field system is associated with thesedifferent soils. Relatively few walls were found on the Kawaihae andKahana soils, which form the western and eastern boundariesrespectively. The combination of porous soils and low rainfall producesthe sharp western edge of the field system. The failure to extendagriculture onto the Kahana soils may reflect its association with highrainfall and relatively dense vegetation. Highly intensified portions ofthe field system are often on Pu'u Pa and Moloka'i soils[ILLUSTRATION FOR FIGURE 13 OMITTED], and so are the M-H and M intensitysectors (but with less of each on Moloka'i soils). Overallpercentages for the areas with agricultural intensification differ fromthose of the entire study area, suggesting that the type of soilaffected choices about agricultural expansion and to some extent thedegree of agricultural intensification.Ahupua'aThe level of intensification within each of the 32 ahupua'aalong the Kohala coast differs ([ILLUSTRATION FOR FIGURE 14 OMITTED],TABLE 2, and [ILLUSTRATION FOR FIGURE 2 OMITTED]). In FIGURE 14,ahupua'a are arranged from north to south along the X axis, thegraph showing the percentage of H and M-H intensity areas versus M, M-Land L intensity areas within each ahupua'a. In general, thenorthern ahupua'a from Pu'uepa 2 to Kapa'a 1-2 have lessland at higher levels of agricultural intensification, and the middleahupua'a more intensified land. For five ahupua'a in thecentral portion (Kapa'anui, Lapakahi, Kaupalaoa, Kehena 1 andPuaili) more than 40% of land is at H and M-H levels. Intensification inthe southern ahupua'a is less than the centre communities buthigher than the north. This pattern of variation does not directlycorrespond to any simple environmental pattern. We conclude that thelimits to intensification throughout the entire field system had notbeen reached with respect to the traditional Hawaiian land divisions.Discussion and conclusionsOur analysis suggests that further expansion of the Kohala fieldsystem was limited by several environmental parameters. The southwestedge was defined by the minimum amount of rainfall necessary to reliablygrow sweet potatoes. This boundary is also linked to the distribution ofrocky and porous Kawaihae soils, which were largely avoided. Thesoutheastern upper reaches of the field system were confined byincreasing elevations - and, to the north, by the edge of the Kohalaridge - to a limit of around 800 m, a zone of colder temperatures,higher rainfall, wet soils and longer maturation periods. Expansion tothe north was limited by the coastline edge. This area has also beendisturbed by recent development and a number of blank areas in the fieldsystem here are undoubtedly due to post-depositional processes. Othergaps in the field system appear to correspond to localized areas of highrelief.While the extensive limits of the Kohala field system were reached inthese environmental terms, the intensive limits had not yet beenapproached. There is little difference between H and L intensity areaswith respect to rainfall, slope and aspect. There are minor differenceswith soils and with elevation - but the elevational differences do notcorrespond to the expected pattern. The variation in the level ofintensification between ahupua'a indicates that furtherintensification was possible.Our results have implications for understanding the late prehistoricand protohistoric period in Hawai'i, where it has been suggested(most often by Kirch 1984; 1985; 1994: 261-8, but also by Hommon 1986:66-7; Tuggle & Griffin 1973: 63; Tuggle & Tomonari-Tuggle 1980)that agricultural intensification was spurred by the pressure applied bydistrict level chiefs to local farmers within their affiliatedcommunities to produce an agricultural surplus. If this were the case,one would expect a more uniform degree of intensification inahupua'a with similar environmental characteristics under thejurisdiction of the Kohala district chiefs. Our data suggest otherwise,indicating the level of intensification might have varied acrossahupua'a as the result of local production needs or strategies, inaddition to the demands of district level elite.Two hypotheses suggest themselves. First, that population size drovethe level of dryland agricultural intensification. A positivecorrelation Noun 1. positive correlation - a correlation in which large values of one variable are associated with large values of the other and small with small; the correlation coefficient is between 0 and +1direct correlation between the size of coastal villages (or number ofresidential features), the amount of associated marine resources and thelevel of agricultural intensification would support this hypothesis.Alternatively, some more complex combination of environmental andgeographic variables may account for differences in ahupua'aagricultural intensification. The geographic relationship between theminimum rainfall parameter and distance from the coast may be linked tosubsistence - traditionally in Hawai'i marine resources (fish) werecombined with terrestrial cultigens to maintain an adequate diet. To thenorth, sufficient rainfall occurs along the coast to enable drylandagriculture and fishing to be undertaken in close proximity. As onemoves south, the distance increases from the coast to the point whererainfall becomes adequate for dryland agriculture, until the lower-mostfields are several kilometres inland from the coast. In a cost-effectiveadaptive strategy, the central ahupua'a would be intensified morethan either the north or south. Where sufficient rainfall fell by thecoast, no substantial intensification may have been necessary. To thefar south, the distance separating farming from fishing was aconstraint. Only in the central portion of the district were conditionsoptimal for intensification - far enough from the coast to be necessary,close enough to the coast to be cost-effective (see Allen & McAnany1994 for a discussion of environmental variability and Hawaiian landuse).TABLE 2. The percentage of each intensification level within eachof the 32 ahupua'a along the Kohala coast (ordered from north tosouth).ahupua'a low medium-low medium medium-high highPu'uepa 2 39 53 8 0 0Kokoili 61 21 18 0 0'Upolu 42 22 31 3 2Honoipu 27 25 29 14 5Puakea 62 21 16 1 0Kukuipahu 54 33 10 3 0Awalua 57 19 2 4 0Ha'ena 50 27 17 7 0Kapunapuna 33 31 24 9 4Kapa'a 1-2 37 26 25 8 4Kapa'anui 20 24 13 27 17Kou 20 25 24 21 10Kamano 31 21 23 18 7Mahukona 20 27 24 21 8Lapakahi 11 21 24 31 13Lamaloloa 23 29 27 17 4Kaiholena 11 28 35 23 3Makeanehu 20 22 24 19 16Kaupalaoa 10 11 23 39 17Kehena 1 11 23 23 30 14Kehena 2 15 24 45 16 0Puanui 9 21 37 22 12Puaili 6 20 20 33 21Ki'iokalani 6 23 37 25 9Kaiho'oa 14 17 29 33 7Pohakulua Ahula 19 32 29 13 6Pohakulua 52 18 21 0 9Kalala 23 29 22 21 6Makiloa 18 29 33 17 3Pahinahina 15 28 20 28 9Kahua 1 23 25 28 20 4Kahua 2 41 33 20 5 1It has also been suggested that intensification of drylandagriculture in west Hawai'i Island was approaching the point whereno further significant investments were possible and consequentlyalternative strategies, including warfare, were employed to obtain thefood resources necessary to feed the population and the requirements ofthe chiefly elites (Kirch 1994: 261-8). While there was likely amaterial basis for inter-group competition in Hawai'i, our worksuggests that further intensification of the Kohala field system waspossible. Recent evidence from the windward location of Anahulu on theisland of O'ahu shows intensification of taro production during thehistoric era (Kirch 1990b; Spriggs & Kirch 1992; Kirch & Sahlins1992). Other instances of historic agricultural intensifications are theWaimea-Lalamilo field system on Hawai'i Island (Clark 1986), and onKalaupapa, Moloka'i Island (Ladefoged 1993). In Kohala, the limitsto expansion (as opposed to intensification) had been reached andperhaps when this threshold was approached, other alternatives,including cycles of inter-group aggression, began to appear less costlyand more reliable as a means of resource acquisition.Our analysis extends the pioneering work of several archaeologists(including Paul Rosendahl and Stell Newman) on the development ofdryland agricultural systems in Hawai'i. These remarkable, largeand relatively well-preserved complexes are well-suited to GISapproaches, even in the absence of fine-grained chronologicalinformation. With GIS, we can examine the effects of several variableson the entire system which may only be partially represented withinsmaller sections (i.e. the Lapakahi ahupua'a). With a better recordof the timing of construction and use of different portions of the fieldsystem, we could model changes in the process of agriculturalintensification during the late prehistory of the Hawaiian Islands. Atthat point we expect more completely to resolve the environmental andsocial factors identified in this analysis which may have stimulated theexpansion and intensification of dryland agriculture.Acknowledgements. This paper would not have been possible without thearchaeological base map drawn by Myra Tomonari-Tuggle and discovered byMWG MWG Men with Guts (sports apparel company)MWG Match-Winning Goal (soccer)mWG Microworld of Gems (e-commerce business)MWG Measurements Working GroupMWG Model Working Group in the Department of Anthropology at the University of Hawai'i.Myra graciously allowed us to use her map of the Kohala field system forthis analysis. Additionally, this paper builds on the previous work ofRichard Pearson (University of British Columbia LocationsVancouverThe Vancouver campus is located at Point Grey, a twenty-minute drive from downtown Vancouver. It is near several beaches and has views of the North Shore mountains. The 7. ), Roger Green(University of Auckland Not to be confused with Auckland University of Technology.The University of Auckland (Māori: Te Whare Wānanga o Tāmaki Makaurau) is New Zealand's largest university. ), P. Bion Griffin (University of Hawai'i),H. David Tuggle (International Archaeological Research Institute), PaulH. Rosendahl (P.H. Rosendahl, Incorporated) and Patrick V. Kirch(University of California, Berkeley The University of California, Berkeley is a public research university located in Berkeley, California, United States. Commonly referred to as UC Berkeley, Berkeley and Cal ) and their students and colleagues.Discussions with Sarina Pearson were particularly helpful in preparingthe manuscript, as were the comments and suggestions of Roger Green. PipForer, Graeme Glen, Terry Hunt, Geoff Irwin, Patrick Kirch, RichardPearson, Anna Romaniuk, Peter Sheppard and Matthew Spriggs also providedhelpful comments and suggestions. The Auckland University ResearchCommittee awarded TNL TNL The Next Level (church; gaming clan)TnL Transform and Lighting (video games)TNL Technical NewsletterTNL target nomination list (US DoD)a grant which facilitated this research. Finally,the late T. Stell Newman pioneered archaeological research on thedryland agricultural systems of Kona and Kohala on the Island ofHawai'i. We would like to dedicate this paper to him and the visionhe established for the analysis of large-scale archaeological patternsin Hawai'i.1 In the following maps the full extent of the field system is shown,but the full extent of the 2-km buffer study area is not. This was doneto enhance the visual display of the field system data. The maximumvalues of the environmental parameters in the graphs are therefore notalways shown on the corresponding maps.ReferencesALLEN, M.S. & P.A. MCANANY. 1994. Environmental variability andtraditional Hawaiian land use patterns: Manuka's cultural islandsin seas of lava, Asian Perspectives 33(1): 19-55.ARMSTRONG, R.W (ed.). 1983. Atlas of Hawaii. 2nd edition. Honolulu(HI): University of Hawaii Press The University of Hawaiʻi Press is a university press that is part of the University of Hawaiʻi. .BARRAU, J. 1961. Subsistence agriculture Subsistence agriculture (also known as self sufficiency in terms of agriculture) is a method of farming in which farmers plan to grow only enough food to feed the family farming, pay taxes or feudal dues, and perhaps provide a small marketable surplus. in Polynesia and Micronesia.Honolulu (HI): B.P. Bishop Museum. Bulletin 223.BARRERA, W.M. 1972. Archaeological excavations and survey at Keauhou,North Kona, Hawaii. Honolulu (HI): B.P. Bishop Museum, Department ofAnthropology. Report 71-10.BEVACQUA, R.F. (ed.). 1972. Archaeological survey of portions ofWaikoloa, South Kohala district, Island of Hawaii. Honolulu (HI): B.P.Bishop Museum, Department of Anthropology Report 72-4.BOSERUP, E. 1965. The conditions of agricultural growth: theeconomics of agrarian change under population pressure. Chicago (IL):Aldine.BROOKFIELD, H.C. 1972. Intensification and disintensification inPacific agriculture: a theoretical approach, Pacific Viewpoint 13:309-48.1984. Intensification revisited, Pacific Viewpoint 25: 15-44.1986. Intensification intensified, Archaeology in Oceania 21: 177-80.CHING, F. 1971. The archaeology of south Kohala & North Kona:from the Ahupua'a of Lalamilo to the ahupua'a of Hamanamana.Honolulu (HI): Division of State Parks, Department of Land and NaturalResources.CLARK, J.T. 1981. Archaeological survey of the proposed Lalamiloagricultural park, South Kohala, Island of Hawai'i. Honolulu (HI):DOWALD, Department of Land and Natural Resources, State of Hawai'i.1986. Continuity and change in Hawaiian agriculture, AgriculturalHistory 60(3): 1-21.1987. Waimea-Kawaihae: a leeward Hawaii settlement system. Ph.Ddissertation. Urbana (IL): University of Illinois University of Illinois may refer to: University of Illinois at Urbana-Champaign (flagship campus) University of Illinois at Chicago University of Illinois at Springfield University of Illinois system It can also refer to: .1988. Paleodemography in leeward Hawaii, Archaeology in Oceania 23:22-30.CLARK, J.T. & P.V. KIRCH (ed.). 1983. Archaeologicalinvestigations of the Mudlane-Waimea-Kawaihae road corridor, Island ofHawaii: an interdisciplinary study of an environmental transect.Honolulu (HI): B.P. Bishop Museum. Departmental Report 83-1.CROZIER, S.N. 1971. Archaeological excavations at Kamehameha IIIroad, north Kona, island of Hawaii - Phase II. Honolulu (HI): B.P.Bishop Museum, Department of Anthropology. Report 71-11.EARLE, T.K. 1978. Economic and social organization of a complexchiefdom: the Halelea District, Kaua'i, Hawaii. Ann Arbor Ann Arbor,city (1990 pop. 109,592), seat of Washtenaw co., S Mich., on the Huron River; inc. 1851. It is a research and educational center, with a large number of government and industrial research and development firms, many in high-technology fields such as (MI):Museum of Anthropology, University of Michigan (body, education) University of Michigan - A large cosmopolitan university in the Midwest USA. Over 50000 students are enrolled at the University of Michigan's three campuses. The students come from 50 states and over 100 foreign countries. . Anthropological Papers63.1980. Prehistoric irrigation in the Hawaiian Islands: an evaluationof evolutionary significance, Archaeology and Physical Anthropology inOceania, 15(1): 1-28.GRAVES, M.W. & T.N. LADEFOGED. 1991a. The disparity betweenradiocarbon and volcanic glass dates: new evidence from the island ofLanai Lanai(lənī`), island, 141 sq mi (365 sq km), central Hawaii, W of Maui island across the Auau Channel; Mt. Lanaihale (3,370 ft/1,027 m) is the island's highest point. For many years the island was used for sugarcane raising and cattle grazing. , Hawaii, Archaeology in Oceania 26: 70-77.1991b. Agricultural expansion in prehistoric Hawai'i: its timingand impact. Paper presented at the XVII Pacific Science Congress,Honolulu (HI).HANDY, E.S.C. 1940. The Hawaiian planter planter,farm or garden implement that places propagating material such as seeds or seedlings into the ground, usually in rows. Broadcasting, i.e., scattering seed in all directions, by hand followed by harrowing (see harrow) to cover the seed with soil was an early 1. Honolulu (HI): B.P.Bishop Museum. Bulletin 161.HANDY, E.S.C. & E.G. HANDY. 1972. Native planters of old Hawaii:their life, lore, and environment. Honolulu (HI): B.P. Bishop Museum.Bulletin 233.HOMMON, R.J. 1986. Social evolution in ancient Hawaii, in P.V. Kirch(ed.), Island societies: archaeological approaches to evolution andtransformation in Oceania: 55-68. Cambridge: Cambridge University Press Cambridge University Press (known colloquially as CUP) is a publisher given a Royal Charter by Henry VIII in 1534, and one of the two privileged presses (the other being Oxford University Press). .KELLY, M. 1983. Na mala o Kona: gardens of Kona. Honolulu (HI): B.P.Bishop Museum, Department of Anthropology. Report 83-2.KIRCH, P.V. 1984. The evolution of Polynesian chiefdoms. Cambridge:Cambridge University Press.1985. Feathered feath��ered?adj.1. Covered, provided, or adorned with feathers.2. Having feathering, as an animal's coat.3. Moving swiftly: feathered feet.4. gods and fishhooks: an introduction to Hawaiianarchaeology and prehistory. Honolulu (HI): University of Hawaii Press.1990a. The evolution of socio-political complexity in prehistoricHawaii: An assessment of the archaeological evidence, Journal of WorldPrehistory 4: 311-45.1990b. Production, intensification, and the early Hawaiian Kingdom,in D.E. Yen & J.M.J. Mummery (ed.), Pacific production systems:190-210. Canberra: Department of Prehistory, RSPAS RSPAS Research School of Pacific and Asian Studies (Australia), Australian NationalUniversity.1994. The wet and the dry: irrigation and agriculturalintensification in Polynesia. Chicago (IL): University of Chicago Press The University of Chicago Press is the largest university press in the United States. It is operated by the University of Chicago and publishes a wide variety of academic titles, including The Chicago Manual of Style, dozens of academic journals, including .KIRCH, P.V. & M.D. SAHLINS. 1992. Anahulu: the anthropology ofhistory in the kingdom of Hawaii 2: The archaeology of history, Chicago(IL): University of Chicago Press.LADEFOGED, T.N. 1993. Hawaiian dryland agricultural intensificationand the Pacific economy, Pacific Studies 16(2): 119-31.LADEFOGED, T.N. & M.W. GRAVES. 1996. The process of agriculturaldevelopment in Kohala, Hawai'i. Paper presented at the 1996 Societyfor American Archaeology The Society for American Archaeology (SAA) is the largest organization of professional archaeologists of the Americas in the world. The Society was founded in 1934 and today has over 7000 members. Meeting, New Orleans New Orleans(ôr`lēənz –lənz, ôrlēnz`), city (2006 pop. 187,525), coextensive with Orleans parish, SE La., between the Mississippi River and Lake Pontchartrain, 107 mi (172 km) by water from the river mouth; founded .LANSING, J.S. 1991. Priests and programmers: technologies of power inthe engineered landscape of Bali. Princeton (NJ): Princeton University Princeton University,at Princeton, N.J.; coeducational; chartered 1746, opened 1747, rechartered 1748, called the College of New Jersey until 1896.Schools and Research Facilities Press.MORRISON, K.D. 1994. The intensification of production:archaeological approaches, Journal of Archaeological Method and Theory1(2): 111-60.MURABAYASHI, E.T. 1970 An analysis of soils and their earlyagricultural implications in the Lapakahi area of Kohala, Hawaii Kohala is the name of two districts — North Kohala and South Kohala — on the northwest portion of the island of Hawai‘i in the Hawaiian Archipelago. , inNewman (1970): 253-79.NEWMAN, T.S. 1970 Hawaiian fishing and farming on the island ofHawaii in AD 1778. Honolulu (HI): Division of State Parks, Department ofLand and Natural Resources.1972. Two early Hawaiian field systems on the island of Hawai'i,Journal of the Polynesian Society The Polynesian Society is a non-profit organization based at the University of Auckland, New Zealand, dedicated to the scholarly study of the history, ethnography, and mythology of Oceania. 118: 87-9.PEARSON, R.J. (ed.). 1969. Archaeology on the island of Hawaii.Honolulu (HI): Social Science Research Institute, University of Hawaii (body, education) University of Hawaii - A University spread over 10 campuses on 4 islands throughout the state.http://hawaii.edu/uhinfo.html.See also Aloha, Aloha Net. .Asian & Pacific Archaeology Series 3.ROSENDAHL, P.H. 1972a. Aboriginal agriculture and residence patternsin upland Lapakahi, island of Hawaii. Ph.D dissertation, University ofHawaii, Honolulu (HI).1972b. Archaeological salvage of the Hapuna-Anaehoomalu section ofthe Kailua- Kawaihae road (Queen Kaahumanu Highway), island of Hawaii.Honolulu (HI): B.P. Bishop Museum, Department of Anthropology. Report72-5.1973. Archaeological salvage of the Ke-ahole to Anaehoomalu sectionof the Kailua-Kawaihae road (Queen Kaahumanu Highway), island of Hawaii.Honolulu (HI): B.P. Bishop Museum, Department of Anthropology. Report73-3.1994. Aboriginal Hawaiian structural remains and settlement patternsin the upland agricultural zone at Lapakahi, island of Hawaii, HawaiianArchaeology 3: 14-70.SAHLINS, M.D. 1958. Social stratification Noun 1. social stratification - the condition of being arranged in social strata or classes within a groupstratificationcondition - a mode of being or form of existence of a person or thing; "the human condition" in Polynesia. Seattle (WA):University of Washington Press.SCHILT, R. 1984, Subsistence and conflict in Kona, Hawaii. Honolulu(HI): Bernice P. Bishop Museum The Bernice Pauahi Bishop Museum, designated the Hawai'i State Museum of Natural and Cultural History, is a museum of history and science in the historic Kalihi district of Honolulu on the Hawaiian island of O'ahu. , Department of Anthropology. Report 84-1.SMITH, K. & A. SCHILT. 1973. North Kohala: agricultural fieldsystems and geographic variables, in Tuggle & Griffin (ed.): 309-20.SOEHREN, L., & T. S. NEWMAN. 1968. The archaeology of Kealakekua.Honolulu (HI): B.P. Bishop Museum and the University of Hawai'i,Departments of Anthropology. Special Publication.SPRIGGS, M.T. & P.V. KIRCH. 1992, 'Auwai, Kanawai, andWaiwai: Irrigation in Kawailoa-uka, in Kirch & Sahlins: 118-64.TOMONARI-TUGGLE, M.J. N.d. Archaeological base map of field systemwalls from the Kohala district, Hawaii Island, Map on file, Departmentof Anthropology, University of Hawaii, Honolulu.TUGGLE, H.D. & P.B. GRIFFIN (ed.). 1973. Lapakahi, Hawaii:archaeological studies. Honolulu (HI): Social Science ResearchInstitute, University of Hawaii. Asian and Pacific Archaeological Series5.TUGGLE, H.D. & M.J. TOMONARI-TUGGLE. 1980, Prehistoricagriculture in Kohala, Hawaii, Journal of Field Archaeology 7: 297-312.USDA (United States Department of Agriculture, Soil ConservationService), 1973. Soil survey of island of Hawaii, State of Hawaii.Washington (DC): US Government Printing Office.YEN, D.E. 1971. The development of agriculture in Oceania, in R.C.Green & M. Kelly (ed.), Studies in oceanic culture history 2: 1-12.Honolulu (HI): B.P. Bishop Museum. Pacific Anthropological Records 12.1973. The origins of Oceanic agriculture, Archaeology and PhysicalAnthropology in Oceania 8: 68-85.YEN, D.E., P.V. KIRCH, P. ROSENDAHL & T. RILEY. 1972. Prehistoricagriculture in the upper Makaha Valley, Oahu, in D.E. Yen (ed.), MakahaValley Historical Project: Interim Report 3: 59-94. Honolulu (HI): B.P.Bishop Museum. Pacific Anthropological Records 18.