Characterizing weed management activities for archeological site preservation and grass-fire mitigation at Kaloko-Honokōhau National Historical Park

Leary, James; Gross, Jacob

This report summarizes recent studies conducted by the University of Hawaii, College of Tropical Agriculture and Human Resources to improve non‐native vegetation management techniques within Kaloko‐Honokohau National Historical Park. The first section examines the timing of mechanical and herbicide control methods for suppressing the invasive C4 tussock, fountain grass (Pennisetum setaceum). All treatment combinations initially suppressed live fountain grass cover, but the timing of treatments significantly influenced suppression longevity. We associated this timing effect with precipitation events. Mechanical treatments showed extended suppression when applied during a dry period (7.4 mm for the month), while herbicide treatments were most effective when applied during a much wetter period (74.9 mm for the month). This result encourages managers to closely monitor and exploit weather patterns when scheduling management practices to suppress fountain grass. The second section focused on the development of a Dissected Archeological Site Clearing (DASC) protocol to measure all aspects of a woody vegetation removal operation (i.e. cutting, loading and hauling) within an archeological site as a way to accurately estimate relative effort (i.e. labor resources) to complete the task. Results from six assigned sites determined that hauling the cut vegetation accounted for a considerable portion of the entire effort (46‐81%). Ease of site access, adjacency to disposal site and the number of cut vegetation loads were all major determinants on the labor resources required to accomplish a task. The third section of this report utilized this DASC field data to develop a GIS modeling tool to estimate labor resources for future site‐clearing efforts. Major determinants for calculating labor resources included: (i) vegetation type, area and density for estimating cutting time; (ii) the shortest distance from the arch site to load biomass on to the utility vehicle (UTV); (iii) shortest hauling distance to the disposal site. By utilizing GIS data layers from a 2010 vegetation survey and the existing UTV trail network, outputs can be estimated remotely with simple designations of the site clearing area and disposal site. The output includes an estimate of total labor resources (i.e. person‐hrs) and an overall timeline (based on the number of personnel) to accomplish an assigned task. This model, in its current form, only utilizes a small data set derived from our DASC field calculations, but is designed to accept new data sets provided by field technicians in future tasks. The purpose of this tool is to facilitate management decisions by prioritizing limited resources to a wide range of site clearing activities and projecting future budget proposals more accurately.