Figure 1 – Map of Pallas catchment and location in
Finland. The location of the DTS cable, stream, and ditch systems are
shown. Classification of vegetation in the catchment is derived from
Räsänen et al. (2021).
The video (Video S1) shows daily mean water temperature at a 10m
resolution for a 2 km reach in a headwater subarctic stream in Pallas,
Northern Finland (68°02′N, 24°16′W; Figure 1). For context, moving
graphs of precipitation, air temperature, and snow depth are also shown.
A Halo DTS system (Sensornet, UK) was installed into the stream and made
double-ended measurements. The DTS measured water temperature at 2 m
spatial resolution and at a 30-minute temporal resolution from 01 May
2021 to 16 Sep 2021. The data collection captured the onset of snowmelt
period and continued to early autumn which comprises the period where
the stream shows the strongest connectivity with the wider catchment in
Pallas (Marttila et al., 2021). DTS temperature records were calibrated
against a reference HOBO Pendant water temperature logger (Onset, MA,
USA) located in an upstream spring. The DTS data were then cleaned, with
areas the cable that were exposed to air temperature due to dewatering
filtered out using cross-correlation analysis, and 10 m averages
calculated (n = 5). To facilitate assessment of spatial and seasonal
dynamics daily mean water temperature records are presented. All data
processing and visualization was completed using R version 4.04 (R Core
Team, 2022). Graphs were created using the “ggplot ” package and
“gganimate ” was used to generate moving images.
Our visualization shows snowmelt to be a dominant control on the thermal
regime of the whole Lompolonjängänoja stream system. Throughout the
snowmelt period, water temperature remains relatively low and constant
at all locations along the cable, likely because of cold meltwater
runoff from the catchment dominates temperature dynamics at all points
during this period. Thus, length of snowmelt season is highlighted as a
crucial control on stream temperature dynamics (Slemmons et al., 2013).
After snowmelt, localized fluctuations in water temperature regime
become apparent. The 500-650 m segment of the stream had consistently
above average stream temperature during the summer period. Thus, it may
be inferred that localized inputs of warm surface water occur here,
creating a “hotspot” of potential increased biogeochemical activity
(Marruedo Arricibita et al., 2018). GIS investigation revealed drainage
ditches may be a possible source of these warm water inputs, a prominent
feature in Nordic peatlands due to historical drainage practices
(Hasselquist et al., 2018; Nieminen et al., 2018).
The cable terminates in a large natural spring (located at the 1900-2000
m portion of the cable), which is seen as a very constant temperature in
the spring throughout the study period. Interestingly, while DTS has
previously been used to identify groundwater inputs into streams
(Marruedo Arricibita et al., 2018; Matheswaran et al., 2014), and
despite numerous other springs visible in the channel, the impact of
springs on stream temperature appears minimal during the summer period.
Only one area of the cable (~800m) appeared consistently
cooler than the rest of the stream during the warmest period of early
July. While cold water refugia have previously been found to be
important refugia from high temperatures in other systems (Fullerton et
al., 2018), the low impact visible for this subarctic stream suggests
that the downstream influence of springs provides minimal mitigation to
high temperature events. By autumn (September), water temperature across
the cable once again became relatively spatially homogenous despite air
temperatures remaining elevated. This indicates the hydrological
connectivity providing localized ‘hotspots’ in water temperature become
disconnected from the main channel by early Autumn. Thus, it is notable
the most dynamic period of water temperature response to catchment
behavior appears constrained to the post-melt summer months.
Our dataset has potential to investigate dominant controlling processes
for water temperature at high spatial resolution in logistically
challenging high latitude environments. Identification of patterns in
the resolved spatial and temporal data become apparent through animation
and the great potential of DTS to assess seasonal drivers of water
temperature can be observed.