MEASUREMENTS OF SOIL HEAT FLUX AND SOIL TEMPERATURE USING A LOCALLY DEVELOPED SENSOR SYSTEM

Abstract

The aim of this study was to design, fabricate and evaluate the field performance of a soil heat flux and soil temperature sensor system from locally sourced materials.The specific objectives were to: measurethe soil temperature and soil heat flux with a locally developed sensor system and compare the results of its measurements to that of the standard sensors.To construct the sensorsa temperature transducer(LM35IC) and a heat flux transducer (Thermocouple) were selected and used. The sensorswere then calibrated against standard soil heat flux and temperature sensors. The sensors’ signals were digitized and programmed using theArdiuno Uno Board. Testing was carried out to certify the functioning of the system. The sensor system was deployed in the field for measurements at WASCAL Observatory in the Federal University of Technology, Akure. This observatory has a standard soil heat flux sensor (HFP01SC Soil Heat Flux Plate) and soil temperature sensor (10TCTR-107 Soil Temperature Probe)both made by Campbell Scientific Ltd.The accuracy of the standard soil temperature sensor is ±0.3oC over the range -25oC to +50oC while that of the standard soil heat flux sensor is ±3% of reading.Thedeveloped soil heat flux sensor was embedded at 10cm soil depth while thedeveloped soil temperature sensor wasembedded at 5cm soil depth supposed to be the depths of installation of the standard Campbell soil heat flux and temperature sensors respectively. Afterinstallation the developed sensor system was set to the same time interval with the standard.The data collected during the field work were subjected totime series, statistical analysis and mathematical estimations.The calibration yielded the following calibration constants,5.253W/m2/mV, 4.327W/m2/mV,6.445W/m2/mV, 0.543oC/mV, 0.4490C/mV and 0.1170C/mV for the heat flux sensor 1 (hfs1), heat flux(hfs2), heat flux sensor (hfs3), temperature sensor1(ts1), temperature sensor (ts2) and temperature sensor (ts3) respectively. The offsets were respectively,12.08W/m2, 11.22W/m2, 11.22W/m2, 14.74oC, 15.64oC, 27.57oC for the hfs1, hfs2, hfs3, ts1, ts2 and ts3. The performance evaluation of the sensors against standard sensors was also carried out using some statistical tests of performancenamely,correlation coefficient(R), means bias error (MBE), root mean square error (RMSE) and standard deviation (STDEV). The R values for the heat flux sensorshfs1, hfs2 and hfs3were respectively,0.837, 0.852 and 0.848.The MBE valueswere 0.00044, 0.00018 and 0.00021, for the hfs1, hfs2 and hfs3 sensors while the RMSE values were 7.324, 6.0422, and 7.026 respectively. For the temperature sensors, ts1, ts2 and ts3, the R valueswere 0.889, 0.638 and 0.653, the MBE values 0.0000257, 0.0000179 and 0.0001528while theRMSE values were, 0.00572, 0.00461and 0.00883 respectively.The mean soil temperature during the given period ranges from 29.2 to 32.90C. Also the minimum soil temperature ranges from 25.1 to 29.10C while the maximum soil temperature ranges from 33.2 to 37.40C. The mean heat flux that entered the soil during the period ranges from 32.13 to 75.53 W/m2.The system’s field observations agree with model predictions for soil heat flux and soil temperature.The study therefore concluded that the soil heat flux and soil temperature sensors developed from locally sourced materials compared favourably with imported standard sensors and could be deployed for field observations.