Invasive species are well known to have dramatic impacts on ecosystems, and it is hypothesized that such impacts can feed back to exacerbate invader abundance. However, long-term data on ecosystem impacts of invasive species are often unavailable, so it is not known whether hypothesized relationships persist over time. In Hawai′i Volcanoes National Park the African C4 grass Melinis minutiflora became invasive in the 1970’s and 80’s after wildfires. It was found to have positive feedbacks with soil nitrogen (N): Melinis increased net N mineralization rates in post-fire invaded sites as compared to unburned native woodland, and Melinis growth was N-limited. Invaded sites however, had high potential for N loss via leaching due to low NPP and the lack of active plant N uptake during the winter when rainfall is plentiful and mineralization rates are high. If high N loss has occurred over the course of Melinis domination then feedbacks between Melinis and soil N may have broken down over time. To investigate this, we resampled invaded and native sites to re-quantify ANPP, plant N-uptake, and soil N cycling. If N has been lost from Melinis invaded sites, then we would expect that net N mineralization in invaded sites would have decreased since 1995, lessening differences between native intact woodland and invaded sites.
Results/Conclusions
Indeed, 17 years later, net N mineralization quantified with the same intact core method shows that net N mineralization rates in native woodlands have remained constant, while rates have halved in invaded grasslands. In addition, Melinis abundance and ANPP have decreased since 1995. These results suggest that positive effects of Melinis on net N mineralization have disappeared. We also evaluated N dynamics in native woodland sites that have been newly invaded by Melinis to ask whether N mineralization was increasing (as was found 17 years ago after Melinis invaded burned sites). Interestingly, more recent Melinis invasion in unburned woodland sites show dramatic net N immobilization with the onset of the rainy season, even though available N (NH4+ + NO3-) pools were much higher compared to intact woodlands before the rains. One potential explanation for this is high C availability due to substantial root inputs from young Melinis plants with high growth rates early in invasion. While the mechanism is yet unclear, these results point to important interactions between species invasions, fire, and time in predicting the long-term impacts of non-native species.