Soil Phosphorus and Agricultural Development in the Leeward Kohala Field System, Island of Hawai‘i

The leeward Kohala Field System on the island of Hawai‘i was one of the most intensive pre–European contact dryland agricultural systems. Archaeological and soil analysis has documented changes in soil nutrients over time. Soils were collected under agricultural field walls of different relative ages within the Kohala Field System. These field walls preserved soil from the time of their construction ( between ca. A.D. 1400 and 1800), so soil samples from underneath older field walls have been exposed to a shorter period of cultivation than the soils under more recent field walls. Total P and P : Nb ratios of these buried soils were greater under walls than in once-cultivated surface soils, and greater under older walls than under younger walls. These results suggest that precontact cultivation decreased soil P reserves in this intensive agricultural landscape. Before European contact, Hawaiians created and maintained a number of rain-fed dryland agricultural systems, mostly on the geologically younger islands of Hawai‘i and Maui (Kirch 1985, 1994). These systems covered many kilometers with a fine-scale matrix of earth or rock field walls and paved trails (Clark 1987, Ladefoged 1993, Rosendahl 1994, Ladefoged and Graves 2000, 2005, Allen 2001, 2004, Ladefoged et al. 2003, Burtchard and Tomonari-Tuggle 2004, McCoy 2005). Both contemporary accounts at the time of European contact and the scope of the agricultural infrastructure make clear that these represented intensive agricultural systems by any measure. At least some of these systems were maintained for 200–500 yr before European contact (Rosendahl 1994, Coil 2004, Ladefoged et al. 2005; T.N.L. and M. W. Graves, unpubl. data), though most were abandoned shortly after contact as Hawaiian populations were devastated by introduced diseases. To be sustained, intensive agriculture must replenish the nutrients that are removed in agricultural products or otherwise lost from perennially disturbed sites—and the more intensive the agriculture, the greater the challenge (Sanchez 1976). Modern agriculture replenishes nutrients through fertilization, whether by industrial or organic amendments. Hawaiians had no access to industrial fertilizer and no access to forms of energy (fossil fuels or even draft animals) that could help to distribute organic material. Although Hawaiian cultivators mulched extensively (Handy and Handy 1972), replacing the nutrients removed from soils by their crops cannot have been easy, and on the scale of the entire field systems it may not have been possible. One way that Hawaiians did address the nutrient requirements of intensive agriculture was through their choice of areas to intensify. All of the large rain-fed systems occur in areas that represent ‘‘sweet spots’’ of rainfall and soil fertility. They receive enough rainfall to support crop growth in most years (Kirch Pacific Science (2007), vol. 61, no. 3:347–353 : 2007 by University of Hawai‘i Press All rights reserved 1 Research was funded by NSF biocomplexity grant BCS-0119819 and grants from the University of Auckland. Manuscript accepted 29 August 2006. 2 Earth Systems Program, Stanford University, Stanford, California 94305. 3 Current address: 1616 East Howell Street, Apartment 206, Seattle, Washington 98122. 4 Department of Anthropology, University of Auckland, Auckland, New Zealand. 5 Department of Biological Sciences, Stanford University, Stanford, California 94305. 6 Corresponding author: [email protected]. 1985) but not so much rain that nutrients have been removed from soils by tens to hundreds of thousands of years of leaching (Chadwick et al. 2003, Kirch et al. 2004, Vitousek et al. 2004). As soils grow progressively older, the cumulative effects of leaching cause these sweet spots to narrow and eventually to disappear, effectively confining most of the large rain-fed intensive agricultural systems to portions of the island of Hawai‘i and to Haleakalā on Maui. Could long-sustained intensive agriculture have caused the depletion of essential nutrients, even from the relatively nutrient-rich soils that were cultivated? If so, decreases in soil fertility could have decreased crop yield, stressing the productive system in the late precontact period, when Hawaiian populations were at or near their peak and demands for social production were high. Because of its mobility, nitrogen (N) is the nutrient most likely to be depleted by intensive agriculture, but that mobility makes it difficult to infer past N dynamics from modern N pools. However, phosphorus (P) also can limit crop production, especially in tropical systems, and its very low mobility makes its past dynamics more amenable to analysis. Recent research shows that soil P is enriched within dryland agricultural systems, relative to adjacent drier and especially wetter sites (Kirch et al. 2004, Vitousek et al. 2004). The enhancement of resin-extractable P (more or less biologically available) is particularly striking on older Pololū substrates in leeward Kohala, Hawai‘i: concentrations within the abandoned field system average >125 mg/g, versus <20 mg/g for adjacent sites outside the system (Vitousek et al. 2004). Total P also is enriched within the field system, by twoto threefold relative to adjacent areas outside it and (in the younger Hāwı̄ substrates) relative to basalt parent material as well. Vitousek et al. (2004) evaluated possible reasons for this enrichment in total P by analyzing soils that had been buried under field walls when the dryland system was established. They reasoned that if mulching by Hawaiian cultivators were responsible, then any enrichment should be less under walls; if a natural process (such as ‘‘mining’’ of subsoils by native forest vegetation, over millennia [ Jobaggy and Jackson 2001]) were responsible, then underwall soils should be at least as enriched as surface soils within the field system. They found buried soils to be substantially more enriched in P than modern soils and concluded that a natural process was responsible for the elevated P within the field system. They further speculated that the difference in total P between the buried soils and surface soils within the field system could reflect nutrient depletion under centuries of intensive cultivation by Hawaiian farmers. In this paper, we pursue that speculation by making use of additional samples of soils from under field walls in leeward Kohala. These additional collections were based upon relative chronologies of wall construction in three areas of the system. We tested whether the degree of P enrichment in buried soils correlates with the age of the walls that cover them, as would be expected if sustained cultivation causes P depletion.