Three Parks Savanna Fire-effects Plot Network (Fauna): Terrestrial Vertebrate Monitoring, Kakadu National Park, Northern Territory, Australia, 2012–2014

Abstract: This Terrestrial Vertebrate Monitoring data package comprises data for a sub-sample of the 133 plots situated in Kakadu National Park. Terrestrial mammals, birds, reptiles and amphibians are sampled in a 50 x 50 m plot adjacent to each fire plot monitoring site (below). Each vertebrate plot is sampled every five years. Within each five yearly-interval plots are sampled in different months over a three-year period. Each individual plot is always sampled in the same month.

The plots are assessed for fire occurrence at least annually by means of on-ground visits by ranger staff. This record is used, in conjunction with aerial surveys, to assess the reliability of the fire history which has been developed from satellite imagery. A photo record assembled for field visits to Kakadu National Park (and Nitmiluk National Park) visits spanning 1995-2004 has also been used to develop a practical field-based index of fire severity. The Three Parks Savanna Plot Network’s full program uses a total of 220 permanent monitoring plots established in the Kakadu, Litchfield and Nitmiluk National Parks which commenced in 1994-1995. The research is aimed at improving our understanding of impacts associated with imposed fire on vegetation structure, plant species and vertebrate fauna, in order to improve our knowledge and management of fire in tropical savanna landscapes.

A synopsis of related data packages which have been collected as part of the Three Park Savanna's full program is provided at https://doi.org/10.25911/5c35959d4530d.

Sampling method: The network program at Kakadu National Park uses a core of 133 sites to sample a range of landform and vegetation type / habitat conditions. The plots are assessed for fire occurrence at least annually by means of on-ground visits by ranger staff. This record is used, in conjunction with aerial surveys, to assess the reliability of the fire history which has been developed from satellite imagery. A photo record assembled for field visits to Kakadu National Park (and Nitmiluk National Park) visits spanning 1995-2004 has also been used to develop a practical field-based index of fire severity.

Where possible fauna survey sites are set up adjacent to the left-hand edge of the fire plots. In the case of creek line site, the fauna quadrat often needs to be upstream or downstream of the site.

Study extent: The monitoring program was designed originally to sample vegetation. Its applicability for the monitoring of fauna was recognised subsequently. As a consequence, the monitoring of fauna generally has lagged behind the vegetation monitoring program. Whereas all 133 monitoring plots in Kakadu have been sampled for vegetation four times over the period 1995-2009 (and the fifth
resampling commenced April 2013), sampling for fauna has been far less substantial and regular: 15 of the Kakadu plots have been monitored for fauna four times, 121 plots three times, 39 plots only twice, and 6 plots have not been sampled at all.

Project funding: Between 2012 and 2018 this project was part of, and funded through the Long Term Ecological Research Network (LTERN) a facility within the Terrestrial Ecosystem Research Network (TERN) and supported by the Australian Government through the National Collaborative Research Infrastructure Strategy.

This project has also received funding through Parks Australia, NT Parks & Wildlife Service, Tropical Savannas CRC, and Darwin Centre for Bushfires Research (Charles Darwin University).