The Hawaiian Islands as a Model System for Ecosystem Studies

The Hawaiian Islands encompass an extraordinary range of variation in climate and soil age in a small area; the younger volcanoes are also extraordinary for their lack of variation in relief or topography, parent material, and biota (before widespread invasions by alien species). Consequently, in Hawai'i the independent and interactive effects of temperature, precipitation, and soil age on ecosystem structure and function can be evaluated with a power that is beyond the reach of studies elsewhere. Not only are extreme conditions well represented in Hawai'i, but there are also complete gradients between the extremes, allowing the determination of the relationships as well as the differences among sites. My colleagues and I have established two sets of sites that make use of these gradients: the Mauna Loa Environmental Matrix, a set of lava flows ('a'a versus pahoehoe, old versus young) that cover a broad elevational range on the wet east versus dry northwest flank of Mauna Loa; and a chronosequence of sites that reaches from KIlauea (~300 yr old) to Kaua'i (~4,100,000yr old) at 1200 m elevation, 2500 mm annual precipitation. These sites are being used to determine climatic and developmental controls of ecosystem function. I report some of the early results here. A NUMBER OF BIOLOGICAL, geological, and climatic features combine to make the Hawaiian Islands a unique system for ecosystem studies. In fact, I suggest that Hawai'i can be to ecosystem studies what Escherichia coli has been to molecular genetics or the Hawaiian Drosophila to population genetics: a relatively well-defined model system in which fundamental mechanisms can be identified, understood, and tested. Hawai'i is useful for ecosystem studies in part because environmental factors vary across an extraordinarily broad but extraordinarily well-defined range of conditions. Jenny (1941, 1980) proposed that the characteristics of soils and ecosystems are determined by a small number of variables that he termed "state factors" (climate, relief, parent material, organisms, time, etc.). This apI The research described here was supported by NSF grants BSR-8414821, BSR-8718003, and BSR-8918~82 to Stanford University. Manuscript accepted 27 April 1994. 2 Department of Biological Sciences, Stanford University, Stanford, California 94305. 2 proach represents a powerful framework for determining ultimate controls on ecosystem structure and function, and Jenny and his successors (including many Hawaiian soil scientists) have applied it widely. Except under unusual circumstances, however, analyses based solely on state factors cannot determine the regulation of ecosystem dynamics. The state factors interact strongly with each other, and there are often several layers of mechanisms and feedbacks between these ultimate controls and the proximate mechanisms that regulate ecosystem structure and function. Nonetheless, an explicit combination of the context provided by state factors with the insight developed from detailed process studies (together with the conceptual and mathematical models that integrate them) remains the most productive way to analyze proximate as well as ultimate controls on ecosystems. In this paper, I review several theories that relate the time, climate, and organism state factors to the control of ecosystem properties and processes, and show how each of the major factors varies in Hawai'i. I then dis"4i Yi£Zfii4 I, 61 iZiifJtJl!Ldlil d1t&!J2&ii&llZZ&H. Z!O!ll&J2J&IilJE!£&M1,'!U M Hawaiian Islands as Model for Ecosystem Studies-VrrousEK 3 cuss how the wide, well-defined, continuous, and independent variation in these factors makes the Hawaiian Islands an extraordinary resource for the development and testing of ecosystem theory and its application.