Measurement of emissivity in Namibia, GOBABEB site
Posté le 21 November 2018 dans News
Measuring device:
The device used is the emissivity measuring device EM3 from THEMACS Engineering. This device is the result of a patent deposited in 2012 by the CERTES (University Paris Est Créteil). The patent [1] describes the apparatus, the operation thereof in [2]. Figure 1 shows the apparatus.
Broadband Emissivity Measurements:
The measurements were made around the mast managed by the KIT. The numbers of the measurement points correspond to the points marked around the fence surrounding the mast. These points were taken up by Laurent Poutier and Stéphane Langlois from ONERA for spectral measurements. The standard deviation on the emissivity is between 0.005 and 0.009. This brings the expanded uncertainty to a level of 0.015 to 0.027.
In view of the measurements made on this type of soil, an average emissivity of 0.915 may be considered as the reference value. The raw measurements are given in Table 1. The averages made on the different measurements made on the points at different times are given in Table 2.
Table 1
N° measurement | Point | Emissivity |
1 | Point1 | 0.92327535 |
2 | Point2 | 0.90424594 |
3 | Point2 | 0.92339245 |
4 | Point2 | 0.91850336 |
5 | Point3 | 0.901494 |
6 | Point3 | 0.91510734 |
7 | Point4 | 0.92652498 |
8 | Point4 | 0.91449254 |
9 | Point5 | 0.91273598 |
10 | Point5 | 0.92465132 |
11 | Point6 | 0.90190386 |
12 | Point6 | 0.91785928 |
13 | Point6 | 0.92599801 |
14 | Point7 | 0.91800566 |
15 | Point7 | 0.90679296 |
moyenne | 0.91566554 | |
standard dev. |
Table 2
Point | Emissivity |
Point1 | 0.92327535 |
Point2 | 0.91538058 |
Point3 | 0.90830067 |
Point4 | 0.92050876 |
Point5 | 0.91869365 |
Point6 | 0.91525372 |
Point7 | 0.91239931 |
moyenne | 0.91625886 |
standard dev. | 0.00505357 |
Measurements of emissivity in band III:
The following measures were carried out in Band III. The filter supplied by Dexter-Research is shown in figure 2. The thermopile was changed to replace it in a band closer to that of the radiometers KT15 and KT19. However, as shown in figure 2, the strip is much wider than that of these radiometers. The emissivity measurements cannot be completely compared. The results are presented in Tables 3 and 4.
Table 3
Point | Emissivity |
point1 | 0.92783419 |
point1 | 0.90714245 |
point1 | 0.91019695 |
point1 | 0.91059108 |
point2 | 0.89965401 |
point2 | 0.89935842 |
point3 | 0.89029346 |
point3 | 0.89216557 |
point3 | 0.90152612 |
point3 | 0.88694337 |
point4 | 0.91689713 |
point5 | 0.87728723 |
point6 | 0.90182172 |
point6 | 0.90211732 |
point6 | 0.90172319 |
moyenne | 0.90170348 |
standard dev. | 0.01236199 |
Table 4
Point | Emissivity |
point1 | 0.91394117 |
point2 | 0.89950621 |
point3 | 0.89273213 |
point4 | 0.91689713 |
point5 | 0.87728723 |
point6 | 0.90188741 |
moyenne | 0.90037521 |
standard dev. | 0.01450188 |
Laboratory Spectral Measurements:
These measurements were done in the laboratory and not in the field. The spectrometer is a Frontier® Perkin-Elmer®, with an integrating sphere Pike®. Measurements were made separately on sand and gravel (supposed quartz) pieces. The samples are shown in figure 3.
The results are shown in figure 4 and 5. It can be seen that the gravel has a high reflection factor (around 0.5) and very variable in the 8-14 μm band, whereas the sand has a very low reflection factor (Between 0.04 and 0.15). The ratio between quartz gravel and sand will strongly influence the reflection factor and therefore the emissivity. The emissivity is a mean of emissivity of sand and gravel weighted by the surface of each components.
Calculation of emissivity is presented in table 5.
Calculation is done for a temperature of 303K.
Calculation in band 2-17µm is similar at the broad band measurement with EM3. For the sand the value of reflectance greater than 0.4 for wavelength under 2µm explain the difference of these values (EM3 εground=0.915 , spectral measurement εsand=0.937 εgravel=0.873).
Calculations are proceed with differents band :
- 2-17µm (the maximum with an integrating sphere
- 8-14µm The window used in EM3 in band III
- 8-12µm The sensitivity of the A35 thermal camera
- 6-11.5µm The sensitivity of the KT-19.85
For the last one we took the sensitivity of the KT-19.85.II showed in figure 6 for the calculation of an apparent emissivity.
Table 5
Emissivity for different spectral band | ||||
Material | 2-17µm | 8-14µm | 8-12µm | 9.6-11.5µm |
Sand | 0.937 | 0.925 | 0.909 | 0.9365 |
Gravel | 0.873 | 0.806 | 0.758 | 0.894 |
A) Frontier® spectrometer; B) Gobabeb quartz gravel; C and D) Gobabeb sand
Conclusion:
The difference of substrate between sand and gravel, and the great difference of spectral behaviour show that the emissivity depends on the ratio of these two components. The only comparison possible is the 8-14µm bandwidth. The measurement with EM3 shows an emissivity of 0.901 and the spectral measurement shows a value of 0.925 if we don’t take in account the value of the gravel (0.806). We have supposed a flat response of the detector in EM3.
The difference between the different measurements comes from the spectral bandwidth of the detectors. We need to build an emissivity measurement device fitted for the radiometer used for thermal measurement. Emissivity with spectral measurement of sand shows a value of 0.9365. The right value could be under if there is more gravels, upper with the roughness of the soil. This value is similar to the precedents measurements.
New investigation is necessary to explain some differences in measurement values.
References:
[1] Brevet : « dispositif de mesure de l’émissivité ou de la réflectivité d’une surface », N° 1251476
- CANDAU, L. IBOS, J.P.MONCHAU
[2] Infrared Emissivity Measurements of Building and Civil Engineering Materials A new device for measuring emissivity J.P.MONCHAU, L. IBOS, Y. CANDAU , V. FEUILLET (CERTES, Université Paris Est, Créteil), M. MARCHETTI (CETE de l’Est-LRPC Nancy-ERA Tomblaine ), J. DUMOULIN (IFSTTAR, MACS, Bouguenais ) (International Journal Of Thermophysics), Vol. 34, N°4, Avril 2013