Penelope Godwin ^1,2* Siyuan Tian ¹ Clément Duvert ² Penny Wurm ² Norman Riwu Kaho ³ Andrew Edwards ²

• ¹ Australian National University, Canberra, Australia
• ² Charles Darwin University, Darwin, Northern Territory, Australia
• ³ University of Nusa Cendana, Kupang, East Nusa Tenggara, Indonesia

Revegetation projects are an important feature of landscape function in Indonesian karst savannas. An understanding of the relationship between available moisture and vegetation condition can assist with the planning and implementation of revegetation efforts. Observing interannual trends in normalised difference vegetation index (NDVI) may be an effective approach to remotely monitor the relative success of small-scale revegetation efforts. Working at vegetation restoration sites in the Haharu district of East Sumba, East Nusa Tenggara Indonesia, we identified increasing trends in Landsat 7 NDVI at two of four restoration sites using Mann-Kendall and Theil-Sen trend analyses. A key determinant of vegetation response to changing available moisture (as rainfall or soil moisture), is the degree of access to groundwater. We found that rainfall dependent sites had significant Pearson's correlations with NDVI ranging from 0.52 and 0.71, while NDVI was not correlated with rainfall at shallow groundwater sites. There was a clear negative effect of the very dry El Niño period in the year 2005 on NDVI across all sites, but it was less pronounced at the three shallow groundwater sites. Wet years, associated with La Niña events, including the year 2018 had a positive response to NDVI across all sites, while the response was lower in very wet years with annual rainfall above 1200 mm (years 2000 and 2011). We applied a windowed cross correlation method to mean values of district NDVI to examine the lag between moisture input and NDVI response for both rainfall and soil moisture between 1999 and 2018. We found that the strongest correlation between rainfall and NDVI was with four months of antecedent rainfall (r = 0.71, p <0.001), and between 2015 to 2018 two months of antecedent rainfall gave the highest correlation (r = 0.75, p <0.001). We found the highest correlation between soil moisture and NDVI was with a one-month lag (r = 0.40, p <0.001) followed by no lag (r = 0.35, p <0.001). Through this study, we demonstrated the applicability of using NDVI, rainfall and soil moisture trend analyses to identify groundwater dependent vegetation patches and to monitor the effectiveness of vegetation restoration.