HABITAT CHARACTERISTICS OF REFUGE WETLANDS AND TARO LO’I USED BY ENDANGERED WATERBIRDS AT HANALEI NATIONAL WILDLIFE REFUGE, HAWAI’I BY HUGO K. W. GEE A thesis submitted in partial fulfillment of the requirements for the Master of Science Major in Wildlife Sciences South Dakota State University 2007 ii HABITAT CHARACTERISTICS OF REFUGE WETLANDS AND TARO LO’I USED BY ENDANGERED WATERBIRDS AT HANALEI NATIONAL WILDLIFE REFUGE, HAWAI’I This thesis is approved as a creditable and independent investigation by a candidate for the Master of Science degree and is acceptable for meeting the thesis requirements for this degree. Acceptance of this thesis does not imply that the conclusions reached by the candidate are necessarily the conclusions of the major department. ____________________________ Leigh H. Fredrickson Thesis Advisor Date ____________________________ Charles Scalet Head, Wildlife and Fisheries Date iii ACKNOWLEDGMENTS I would like to thank Dr. Leigh H. Fredrickson for providing me with guidance and opportunities during my education. I especially enjoyed talking with him about conservation issues and sampling the local cuisine on our travels together. I want to thank Dr. W. Carter Johnson, Dr. Daniel E. Hubbard, and Dr. Dennis Todey for being on my committee. I also am grateful to Fred Paveglio, Kevin Kilbride, Mike Hawkes, Brenda Zaun, and Michael Mitchell for their suggestions throughout this study. Thank you to the U.S. Fish and Wildlife Service and Ducks Unlimited for providing funding to study tropical wetlands, taro lo’i, and endangered waterbirds on Kaua’i. I appreciate the administrative support provided by the Department of Wildlife and Fisheries Sciences at South Dakota State University. I also would like to thank the staff of Kilauea National Wildlife Refuge Complex for securing housing and transportation for this project. I am indebted to Ashley B. Hitt for her countless hours of field work and companionship. I acknowledge the efforts of Tandi Perkins and Scott Becker in pilot studies that were critical to this project. I am grateful to Chadd Smith for sharing his knowledge and enthusiasm of wetland management at Hanalei National Wildlife Refuge. I value Steve Wall’s help with ArcView and Dr. Gary Larson’s assistance with identifying plants. I also appreciate the cooperation and friendship of taro farmers at the refuge during this study. My time in South Dakota was made memorable by the company of friends such as Sharon N. Kahara, Bernard M. Hien, Sheila K. Thomson, Madeline R. Schickel, Nick iv L. Wirwa, Jennifer L. Gutscher, Jessica F. Lee, Kari A. Ranallo, and Karen E. Arnold. My siblings Maggie, Albert, and Trudy (you too Karen) made trips back to Montreal special by keeping the home fires burning. I am blessed to have two loving parents who tolerate my wandering nature. Special thanks to Anil K. Patel, Ming Cheung, Peter Levidis, and Ngaio Richards for their moral support throughout the years. I would especially like to thank Grandma Gee for always pushing me to new limits. Finally, this thesis is dedicated to Grandma and Grandpa Tam who will always be in my thoughts. v ABSTRACT HABITAT CHARACTERISTICS OF REFUGE WETLANDS AND TARO LO’I USED BY ENDANGERED WATERBIRDS AT HANALEI NATIONAL WILDLIFE REFUGE, HAWAI’I HUGO K. W. GEE 2007 Hanalei National Wildlife Refuge (HNWR) was established to protect habitat for the endangered Hawaiian common moorhen (Gallinula chloropus sandvicensis), Hawaiian coot (Fulica alai), Hawaiian duck or Koloa (Anas wyvilliana), and Hawaiian stilt (Himantopus mexicanus knudseni). I studied two major habitat types (refuge wetlands and taro lo’i) used by these endangered waterbirds (EWBs) as well as wetland vegetation and EWB response to moist-soil management at HNWR. Furthermore, the links are discussed between habitat conditions that occur during the taro agricultural cycle and how each stage contributed to the life-history requirements for foraging, loafing, and nesting of these four EWBs. During the study, effective management of refuge wetlands at HNWR was not possible because the water control infrastructure did not allow for the effective transfer or discharge of water needed to promote optimal habitat conditions for all EWBs. As a result, invasive plant species dominated most of the wetland management units. A few wetland subunits (3.3 ha) did have rototiller treatments combined with water level manipulations in an attempt to create conditions for the germination of native and/or naturalized vegetation. vi These manipulations produced moist-soil vegetation such as the annual sedge, fimbry (Fimbristylis littoralis), or the perennial knotgrass (Paspalum distichum). Moorhen were commonly observed foraging in F. littoralis or P. distichum. Koloa also were observed feeding and loafing in F. littoralis after water levels were raised to overtop seedheads. The majority of EWBs were observed foraging or loafing on grass-covered dikes surrounding taro because frequent mowing encouraged vigorous plant growth and increased visibility of predators. Moorhen preferred lo’i being harvested and stilt preferred unvegetated wet fallow lo’i; whereas, coots and Koloa preferred vegetated wet fallow lo’i. EWBs may prefer open water conditions in these taro habitat categories because they provide important invertebrate food resources for breeding and brood- rearing. Moorhen preferred less intensively managed taro lo’i in early growth or mature and medium to dense growth stages because these habitats provided structural support for overwater nests and the complex habitat structure for invertebrates. Furthermore, annual wetland plants also provide forage for moorhen, coot and Koloa. Call response surveys indicated that moorhen were more abundant in taro lo’i (3.6 birds/ha) compared to refuge wetlands (1.6 birds/ha); however, wetland infrastructure was not in place throughout the study to promote optimal conditions in all refuge wetlands. Taro lo’i in early growth or mature and medium to dense growth stages provided important cover and obscurity for 41% (24 of 58) of the moorhen nests found throughout the study. Nest success for moorhens was 64%; however, recruitment may have been as low as 2.5%, likely as result of predation. vii Wetland management (rototiller treatments and partial water drawdowns) for EWB nesting was implemented on 20.2 ha of refuge wetlands either by partial water drawdowns on subunits (16.9 ha) with limited vegetation cover or by removing dense cover (3.3 ha) with rototiller treatments. These manipulations were important in creating suitable nesting habitat for stilts (42 of 48 nests). Partial water drawdowns within managed wetlands with limited residual vegetation created suitable habitat for nesting stilts in 15 days; whereas, response to rototiller treatments took 42 days. Nest success for stilts was 43% and recruitment was 3%. Predator removal, flood control by refuge staff and taro farmers, and limiting human disturbance can potentially improve nest success rate and EWB recruitment at HNWR. In conclusion, the availability of taro habitat conditions suitable for EWBs is influenced by the stage of taro that varies with market demand throughout the year. In contrast, refuge staff can manipulate wetland conditions to create EWB feeding or nesting habitat as needed. However, refuge wetlands must be reconfigured into larger units based on soil texture and a water distribution system developed to allow management activities that will result in habitat conditions for all EWB life-history stages. viii TABLE OF CONTENTS Page ACKNOWLEDGMENTS ................................................................................................ III ABSTRACT....................................................................................................................... V LIST OF TABLES......................................................................................................... XIII LIST OF FIGURES .........................................................................................................XV LIST OF APPENDICES...............................................................................................XVII INTRODUCTION .............................................................................................................. 1 LITERATURE REVIEW ................................................................................................... 5 MOIST-SOIL MANAGEMENT AT HANALEI NATIONAL WILDLIFE REFUGE ......................... 5 TARO CULTIVATION AT HANALEI NATIONAL WILDLIFE REFUGE..................................... 7 HAWAIIAN COMMON MOORHEN ECOLOGY ....................................................................... 9 HAWAIIAN COOT ECOLOGY ............................................................................................ 11 HAWAIIAN DUCK (KOLOA) ECOLOGY............................................................................. 13 HAWAIIAN STILT ECOLOGY ............................................................................................ 14 RATIONALE OF STUDY................................................................................................ 17 STUDY SITE.................................................................................................................... 19 STUDY PLOTS ................................................................................................................. 20 Taro lo’i .................................................................................................................... 21