Sensor Data

This is the first time I’ve played at sensors and this is a very quick and dirty look a the data. There are 3 axis in the accelerometer data. The data is not always clean. I need to consult with Ornitila, there are often steps in the data where sensor seem to get fixed around a new level. Apparently it can be related to battery drain.

Figure 1 shows raw sensor data from the z-axis for 2 birds. Hen 210751 (blue) has a lot of noise (steps) that is not likely due to movement. For hen 210747 (red) the intensity the sensor collected data was much lower as the frequency the sensor collected data was tied to the battery status. This bird appears to have nested in dense cover and the battery was not recharing and had drained to about 30% by 19 May.

Figure 1. Z-axis sensor data for 2 birds.

Figure 1. Z-axis sensor data for 2 birds.

In Figure 2, I’ve quickly reset the sensor data from hen 210751 to a baseline of 0. As I say, this is quick and dirty - most of the large spikes in the sensor readings are due to the steps see in Figure 1 and not due to activity. However, the main purpose of this transformation was to make make it easier to see the changes in patterns as the bird actvity changes from moving arount to sedentary.

Figure 2. Z-axis sensor data for 210751 reset to baseline of 0.

Figure 2. Z-axis sensor data for 210751 reset to baseline of 0.

Results for a Couple of Birds.

Hen 210746 was the first bird to initiate nesting and it looks like it compeleted incubation last weekend. Based on the GPS data this bird initiated nesting around 02:00 on 3 May (times in UTC), but the sensor data suggest it did not settle down until 08:00 on 4 May (black point on Fig. 3). The area defined by the light blue background outlines the period the sensor data suggest the bird was inactive or incubuating - about 24 days. This is followed by an increase in activity - I guess this corrisonds to hatch. The GPS indicates the bird did not leave the nest until ~22:00 on 29 May (red point, Fig. 3) is the first point where the GPS suggest the bird left the nest. I assume this hen is with a brood. Last we heard from her she was on an aquaculture site in side the bay (Fig. 4)

Figure 3. All 3 sensor axis for 210746.

Figure 3. All 3 sensor axis for 210746.

Figure 4 is an interactive map of hen’s 210746 track. You can zoom to a locaiton and if you hover over a point the date and time will popup.

Figure 4. Movements of hen 210746 though incubation and post-hatch.

For comparison I’ve plotted the sensor data for 2 hens. Figure 5 shows the sensor data hen 210748 that did not breed. She remained very active for the entire period making daily movements between the inner and outer part of the bay (Fig. 6).

Figure 5. All 3 sensor axis for 210748.

Figure 5. All 3 sensor axis for 210748.

Figure 6. Movements of non-breeding hen 210748.

Finally Figure 7 shows the sensor data for hen 210756. It appears the GSM fell off around 20 May. We loose the signal for most tags when they fall off but we are still getting data.

Figure 7. All 3 sensor axis for hen 210756

Figure 7. All 3 sensor axis for hen 210756