I ran 4 HMP sensors in the lab for a day and then at Oxford Tract for a week: a new HMP155 as our reference sensor (sn R0360206), a HMP45 (sn T2130005) in the lab, and two HMP60s (sn L1740398, sn G4540023).

The two HMP60s were most recently at Bouldin Alfalfa for Camilo's temperature profile, and sn G4540023 was dropping out and spiking a lot. In the lab, I removed the teflon shielding from this sensor, which seemed to remove AirT/RH spikes. In the field, however, there were still spikes with and without the teflon shielding, so this sensor is probably bad.

The sensors were aspirated. The sensors were connected to a CR6, and I recorded data every minute.

I recommend using the field calibration equation if you would like to correct AirT or RH because the field calibration ran for more days, and had higher R2 values for the regressions.

 

 

Figure 1. Field comparison - Air temperature time series. The HMP60 sn G4540023 drops out frequently and should not be used.

Figure 2. Field comparison - Air temperature time series with bad sensor (HMP60 sn G4540023) removed. The reference sensor and the HMP45C (sn T2130005) match well, and the HMP60 (sn L1740398) is consistently about 0.5 degC cooler than the reference sensor.

Figure 3. Field comparison - Regression between air temperature data streams.

 

 

Figure 4. Field comparison - Relative humidity time series.

Figure 5. Field comparison - Relative humidity time series with bad sensor (HMP60 sn G4540023) removed. The reference sensor and the good HMP60 (sn L1740398) match well, and the HMP45 (sn T2130005) is consistently ~5% lower than the reference sensor.

Figure 6. Field comparison - Regression between relative humidity data streams.

 

Figure 7. Lab comparison - Air temperature time series.

Figure 8. Lab comparison - Relative humidity time series.

Figure 9. Lab comparison - Regression between air temperature data streams.

Figure 10. Lab comparison - Regression between relative humidity data streams.