I am seeing magnetometer readings change by ~0.5 uT/degC (in each axis) when reading earth fields of about 30 uT. This seems much larger than the specified sensitivity change due to temperature (0.1%) for this device, since 0.1% of 30 uT is 0.03 uT. This is the case for multiple units. Is this expected performance, and if not is there anything else I should be checking?