(pictures coming!)
Lake Rotoiti is located in the north west of the South Island, New Zealand. Much of the land surrounding the lake is covered by the Department of Conservation’s (DoC) restoration efforts. In NZ, restoration primarily means pest control because the country has been overrun by four legged, furry, mammalian pests.
The invasion began with the first humans who came to NZ: Polynesians. They brought with them kiore, the Polynesian rat, and kuri, the Polynesian dog. Later, when the Europeans colonized the land they brought the usual agricultural resources: cows, sheep, deer, rabbits, and our very best friends dogs and cats.
The only native mammals in NZ are two bat species so the flora and fauna were not evolved to cope with the grazing and predation pressures caused by the invasive mammals. Unfortunately, when humans began to realize what was happening they introduced more mammals, (mostly mustelids) as a biological control, and these creatures added to the destruction of the natives.
So, the biggest threat to native and endemic species of NZ (which, by the way, are unusually numerous and wonderful, but that’s a whole other story) are invasive mammals. Appropriately, DoC spends the majority of its resources on controlling these animals. They use two methods: trapping and poison. Trapping is effective, but takes an enormous amount of human-power to clear and reset traps, so DoC has been steadily shifting to using a poison called 1080. At Lake Rotoiti, DoC uses both poison and trapping to control the invasive mammals. Their goal is to keep the pests below a 5% threshold, which is determined by a proportion built by pest census data.
While at the lake, our goal was to design and conduct a study that quantitatively measured the effects of DoC’s pest control efforts. We had a lot of freedom, really only limited by time. I worked with two other students, Cate Turner and Bobby Brandt, to measure the differences in stream quality inside and outside the DoC pest control area. We used benthic stream invertebrates to indicate the quality of a stream using a technique developed in NZ called Macro-invertebrate Community Index, or MCI for short. The method is similar to the one used in the LC BIO 141 stream ecology project at Tryon Creek. The theory rests in the fact that certain families of invertebrates are more sensitive to water pollution than others, so if a stream contains critters from these orders, we can infer that the stream is clean, or of high quality. In NZ, the sensitive orders are ephemeroptera (mayflies), plecoptera (stoneflies), and trichoptera (caddisflies).
Cate, Bobby, and I visited a total of six different sites and sampled the invertebrates. We roughly categorized and described the invertebrates in the field and took pictures to identify them with back at the lodge with photo ID guides. We collected data for the first two days of the three day challenge. We hiked over 15 miles in total and carefully, but not without difficulty, navigated the thick bush surrounding the lake as we made our way up streams.
From our on-site identifications it seemed like we were seeing the same little critters everywhere we went, with a few interesting ones here and there, but it wasn’t until we spent all of the third day pouring over our data and working through analysis that we realized all the sites we visited were pretty much the same. Each one contained representatives from at least one of the three sensitive families I mentioned earlier, indicating that they were all clean. Only one out of the six sites was in direct human contact before emptying into the lake, but even so, it was clean. Our data indicated that stream quality is not different between DoC pest control and non-control areas.
On the third evening we presented our preliminary findings to the rest of the group, who had been working through their own projects and were also presenting that night. The atmosphere was buzzing with the delight of a room full of scientists all curious and excited about the biological systems around us. The projects were a bit ragtag, given that we had limited time to plan and conduct the studies, and the results were generally inconclusive based on the little analysis that we had time to do.
Overall the time we spent a Rotoiti was exciting and informative. Not only did we learn quite a bit about the biology of the area, we also learned about the difficulties and shortcomings of the scientific method, particularly in terms of replication and comparison between sites. Cate, Bobby, and I realized that each of our streams had different rates of flow and different volumes, but by the time we had collected all of our data, we had no way to normalize or control for these differences. The problem is not dire, but it does weaken our data and we are limited in the conclusions we can make from the invertebrate populations we observed.
When we get back to Wellington the three of us will finish analysis, completing the MCI assessment for each stream, and write up a formal report. I’m looking forward to running statistical analysis on our data to confirm our preliminary conclusions, that the quality of each stream is not significantly different from the others. This would indicate that pest control probably does not have an effect on the aquatic ecosystems of DoC lands. The implications of this are that any changes caused by the altered terrestrial system (loss of pests) do not trickle down to the aquatic systems and that any 1080 that enters the streams is, apparently harmless, given the results of our investigation.
Our project was designed to be broad and plastic, such that it could fit the many possibilities that we might have faced upon arriving at Lake Rotoiti. This means that our results are also widely applicable to other streams in NZ, but it is of the utmost importance to remember the limitations of our data. It was collected over the course of only two days and is a miniscule sample of the available data (i.e. all the streams and their inverts in NZ). There is a distinct possibility that the same study conducted at a different site, or even at a different time of year, might yield completely different or even opposite results. There are innumerable factors that might be causing the invertebrate populations that we observed. Recall that five out of six of our streams were pristine, untouched by humans from their mountain spring beginnings to their lake endings, this alone means that there will be far less pollutants present in them. In the future it would be important to have an equal number of pristine and human-contact streams so that we could measure any difference and control for it. This shortcoming, and others, means that our study is just a brief glimpse into the complex interactions that may or may not be happening between DoC pest control and streams.
The most important thing I gained from this project was an appreciation for the time and effort that goes into designing and conducting the most basic of ecological studies, and how those basic studies can open the door to a wonderfully intricate web of new problems and questions. And leaping into that web is daunting and exhilarating. I am scientist, after all.