The end of a long winter
As soon as the snow had melted away in Montréal, Karoline and Manuel travelled back north for the first field trip of the season in mid-March. Together with colleagues from Wilfrid-Laurier University, they drove from Yellowknife to Checkpoint, where they had their first opportunity to enjoy the northern lights in the cold subarctic night. Checkpoint is the last stop on the highway at the junction of the Liard and the Mackenzie Highway before one has to depart from the road into the bush to reach the Scotty Creek field camp.
The snowpack at the end of this year’s winter was particularly thick and early in the spring they could still use the snowmobiles to shuttle all our gear and equipment from Checkpoint through the frozen boreal forests, bogs, and fens to our field camp at Scotty Creek. With the help of all available hands and lots of snow shovelling, the camp was set up within two days and the scientific work could begin.
The first priority was to check if the eddy covariance instrumentation had survived the long and harsh winter in the Northwest Territories. Once arrived at the tower sites, a bad surprise was awaiting: both CO2/H2O infrared gas analysers (IRGA) had stopped working during the winter. The eddy covariance system in the bog broke down in late November after the solar panel had been covered in snow. Without any power recharge, the system shut down within a few weeks after our departure from the site last fall. At the taller landscape tower, the EC150 IRGA conked out in early 2015 and all attempts to reanimate the two sensors remained unsuccessful.
Nevertheless, they continued setting up the planned intercomparison experiment between three different IRGA’s: the enclosed LI7200, the open-path IRGA’s LI7500 and the open-path IRGASON. Within a few days, the LI7200 was running smoothly alongside the LI7500. Unfortunately, the EC150 IRGA’s had to be sent back to Edmonton for repair and they could not be included in the intercomparison setup. However, Campbell Scientific reacted quickly to the emails and sent an IRGASON instrument to the field camp. In a record time of four days, the sensor arrived at the camp, delivered by snow mobile from Checkpoint. And on March 29, all three IRGA’s were running in parallel and everybody was happy about the successful damage control.
In the beginning of April, Jessica arrived at the camp while Manuel left for the second part of his field trip to Inuvik. An early warm spell in spring warmed up the snow pack quickly at Scotty Creek, but luckily Jessica was just in time to capture this year’s snow melt. The snow melt period was long and interrupted by several cold spells, which prolonged the melt process. As the snow cover was particularly thick this year, some of the flume boxes, that are used to measure channel runoff across the landscape, were close to be affected by overflow. Some adjustments had to be done on the side arms of the weirs as water levels rose quickly and the original structure was not designed for such an intense snow melt event. Thankfully, the cold spells saved the day and a complete overflow could be avoided at all sampling spots. The sampling of dissolved organic carbon (DOC) and total dissolved nitrogen (TDN) of the snow melt 2015 provides an excellent opportunity to compare the dynamics compared to the snow melt season of 2014, which was characterised by considerably smaller runoff.
Arrived at the tundra field site Trail Valley Creek, the luck was again not on their side. Manuel realised that the eddy covariance system was without power supply. A fall storm knocked over the solar panels during the previous year and they were now buried under 50 cm of dense snow and ice layers. With the help of Chuck, the helicopter pilot, and Dawn, a post-doc from the University of Saskatchewan, they managed to dig out the panel and to re-erect it. At the end of the day, they saw again green lights in the battery box, the batteries were recharging, and the freshly calibrated eddy covariance system was running again.
This field trip has shown us that an early visit to the sites is crucial to ensure high quality eddy covariance measurements during the snowmelt at such remote northern sites. Hopefully, the systems will now run smoothly until the next field trip in June.
Shortly after his return to Montréal, Manuel participated in the General Assembly of the European Geosciences Union (EGU) in Vienna, where he could already incorporate the first results of the intercomparison study at Scotty Creek in his presentation. With this intercomparison study, we could show that CO2 flux measurements by the IRGASON compare well (within ±0.5 μmol CO2 m-2 s-1) with CO2 fluxes derived from the enclosed LI7200 over a cold boreal forest in spring when high frequency corrections for the IRGASON’s temperature sensitivity are applied. These findings give us more confidence in the accuracy and the intersite comparability of our CO2 flux measurements across the Taiga Plains transect.