The ecology of testate amoebae and Cladocera in Hawaiian montane peatlands and development of a hydrological transfer function

Peatland complexes in the humid highlands of Hawai'i are vital refuges of biodiversity and freshwater resources. Hawaiian peat deposits are also rare repositories of terrestrial ecosystem archives located in an otherwise vast expanse of ocean. We investigated the potential for researching the paleohydrological history of Hawaiian montane peatlands on Kohala, Hawai'i Island through analyses of testate amoebae and Cladocera. Surface peat was collected from a variety of ecohydrological habitats (from water pools to hummocks) and analyzed for modern testate amoeba and cladoceran species relative abundance. We identified 54 morphotype taxa from 21 genera of testate amoebae, 4 taxa and genera of littoral Cladocera, and the common peat rotifer Habrotrocha angusticollis. Testate amoeba diversity and morphotype occurrence mirrored observations from many high-latitude peatland studies. Constrained and unconstrained ordinations support the hypothesis that surface moisture, measured as water-table depth, is an important control on the distribution of testate amoebae and Cladocera in Hawaiian peatlands. Transfer functions relying on weighted-averaging and modern analogs were developed to predict water-table depths from species relative abundance data, and these perform well under leave-one-site-out cross-validation: RMSEP = 9.75-10.3 cm, R-2 = 0.56-0.62. Including cladoceran abundance data in the calibration dataset produced modest model improvement: RMSEP = 1-8% and R-2 = 2-13%. A weighted-average partial-least-squares transfer function was applied to microfossil assemblages from a 0.5 m-long peat core with a Pb-210 decay chronology anchored by ten existing measurements of Pb-210 activity and a Bayesian statistical framework. Microfossils were well-preserved in the peat core. The water-table depth optima of an abundant down-core taxa, Hyalosphenia subflava, is not precisely constrained in the calibration data set, but estimates match those of other tropical studies. A reconstruction of water-table depth indicates dry early nineteenth-century conditions, wet conditions in the late 19th to early twentieth centuries, followed by progressive drying for much of the twentieth-century. Testate amoeba composition appears to have been sensitive to severe drought in recent decades. The results signal that assemblages of testate amoebae and Cladocera are useful proxies of Hawaiian peatland paleohydrology and should be considered alongside other archives of Hawaiian environmental history.