Remote Sensing Study
Detecting Change in Ice Presence with Correlations to Climatic Variations in Svalbard, Norway
Satellite Observations
Having assigned the 2013/2014 delta NDSI image to red, 2015/2016 to green, and 2017/2018 to blue, we can see the progression of the glacier overtime. The moraines of the glacier can be visibly seen (in its most noticeable location just north of E in Fig. 3.1) as red, then it moved eastward and shows as green, and eastward again and shows as blue. We can see the glacier heads that appear to be red, then a green prograde of the glacier, and then a blue prograde of the glacier where it is currently at today.
Analysis of displacement in the RGB composite images revealed glacial activity over the time interval studied. We observed that the right side group of glaciers primarily experienced moraine migration, while the left side group of glaciers experienced advancement of the glaciers themselves
Fig 3.1: Master composite of ΔNDSI between individual years
On the right group of glaciers (Fig 3.1) we have A, B, C, D, and E. The right side medial moraine of B, C, D and E show eastward displacement throughout our studied years. The medial moraine between D and E show the biggest amount of displacement totaling ~600m between 2013 and 2018. The medial moraine between A and B shifted at the tip 275m eastward.
The left side group (Fig. 3.1) contains glaciers 1, 2, 3 and 4. 3 and 4 showed advancement along the right flank of the valley between 2014 and 2016. Glacier 1 and 2 surged 2,027m and 1678m respectively from 2013 to 2018. Glacier 1 surged 1,250m from 2014 to 2016 and 778m from 2016 to 2018. Glacier 2 surged 752m from 2014 to 2016 and 465m from 2016 to 2018
Fig 3.2: ΔNDSI for July from 2013 to 2018.
Satellite Displacement Measurements
Meteorological Data Observations
The data plotted on the graphs (Fig. 3.3) reveals trends for each meteorological property over time. There is an overall trend of rising temperatures seen in the air temperature record and both soil depth records. One interesting result to note is that throughout the air temperature graph, the summer temperatures remain more or less consistent or show a very steady rise but the winter air temperatures show more and more erratic variation over time. This feature can be seen more clearly in the smoothed air temperature graph. Another visible trend is the very slight decrease in wavelength which suggests that the winter freeze is coming later and the summer melt is coming earlier as time goes on, decreasing the total ‘winter’ time.
The 50 cm soil depth graph shows relatively consistent winter temperatures from 2006 to 2008 but the amplitude of the graph seems to be increasing with time since then. The upward trend in the temperatures appears to be pushing the average closer to the 0-degree mark and, if this trend continues, the top of the permafrost active layer will be pushed deeper underground. Due to missing data, no safe interpretations can be made about the possible changes to the 1 m soil depth temperatures in recent years.
Perhaps the most alarming of these is the rainfall depth graph. Up until about 2013, there was steady precipitation throughout the year. However, even if we disregard the corrupted data from recent years that have been filtered out, there definitely appears to be a general trend of rising rainfall amounts. Unfortunately, due to the quality and reliability of the data, it would be disingenuous to draw definitive conclusions. With more reliable data, we would be able to address the concern of rainfall coming earlier in the spring and staying later in the winter as we move toward a more rain-based arctic.
Fig 3.3: Graphs depicting weather data plotted against time (2004-2018)
NOTE: Due to sensor malfunctions, some data from 2013 and 2015 is missing.
Previous Research
Fig. 3.4 shows an overlay of average glacial extents over time. The purpose of acquiring this image was to compare our results against the historical record. While it does show a striking trend of decline, this particular region experienced glacial surges up to 2km over the next few years after this dataset ends. However, without a multi-decade average as presented in this dataset, the surge of 3-5 years cannot be used to definitively suggest that the trend has reversed.
