Peng Guangxiong
College of Resources Science and Technology (IRES)
Beijing Normal University (BNU), Beijing, 100875 CHINA
Tel: 86(10) 5880 3675
[email protected]
Li Jing
College of Resources Science and Technology (IRES)
Beijing Normal University (BNU), Beijing, 100875 CHINA
Tel: 86(10) 5880 3675
Su Wei
College of Resources Science and Technology (IRES)
Beijing Normal University (BNU), Beijing, 100875 CHINA
Tel: 86(10) 5880 3675
Tay Lip Hong
Cilix Corp. Sdn. Bhd.,
Lot L4-I-6, Enterprise 4, Technology Park Malaysia, Bukit Jalil,
57000 Kuala Lumpur, MALASIA.
Tel: (603)8996 9430 Fax: (603)8996 9431
Lim Boon Sheng
Cilix Corp. Sdn. Bhd.,
Lot L4-I-6, Enterprise 4, Technology Park Malaysia, Bukit Jalil,
57000 Kuala Lumpur, MALASIA.
Tel: (603)8996 9430 Fax: (603)8996 9431
Norizan Abdul Patah
Malaysian Centre for Remote Sensing (MACRES)
13, Jalan Tun Dr. Ismail, 50480 Kuala Lumpur, MALAYSIA.
Tel: (603)2697 3400 Fax: (603) 2697 3360
Siti Atikah Mohammed Hashim
Malaysian Centre for Remote Sensing (MACRES)
13, Jalan Tun Dr. Ismail, 50480 Kuala Lumpur, MALAYSIA.
Tel: (603)2697 3400 Fax: (603) 2697 3360
ABSTRACT
At present, the column water vapor can be retrieved from Moderate Resolution Imaging Spectroradiometer (MODIS) instruments. In this paper, we used several MODIS near-IR channels to retrieve column water vapour of Malaysia. Many parameters were adjusted to fit with the climate and environment condition of this region. However instead of column water vapor, relative humidity (RH) is often used to describe the moisture content of air in various modeling and application. Therefore, it is very necessary to convert the column water vapor to relative humidity. Based on the relation between surface level specific humidity and column water vapor in tropical region, an empirical expression was developed to derive specific humidity. The surface level relative humidity can be calculated using specific humidity and air temperature, which is very useful to estimate and forecast the forest fire risk.
1. INTRODUCTION
At present, two Moderate Resolution Imaging Spectroradiometer (MODIS) instruments on board the NASA Terra and Aqua Spacecraft platforms are operational for global remote sensing of the land, ocean, and atmosphere. In recent years, a number of papers have reported water vapor retrieval algorithms, which use backscattered solar radiation near 1 μm measured with aircraft and satellite instruments[1-7]. In this paper, we used several MODIS near-IR channels to retrieval column water vapour of Malaysia. Techniques employing ratios of water vapor absorbing channels centered near 0.905, 0.936, and 0.940μm with atmospheric window channels at 0.865μm are used. Due to the special climate and environment condition of this region, some parameters need to be adjusted.
The column water vapor is a useful parameter for meteorological study which indicates total atmospheric water vapor. However instead of column water vapor, relative humidity (RH) is often used to describe the moisture content of air in various modeling and application, for example, the relative humidity is often used to estimate and forecast the risk of forest fire. Therefore, it is very necessary to convert the column water vapor to relative humidity. Based on the relation between surface level specific humidity and column water vapor in tropical region[6], an empirical expression was developed to derive specific humidity. The surface level relative humidity can be calculated using specific humidity and air temperature, which is very useful to estimate and forecast the forest fire risk.
2. COLUMN WATER VAPOUR RETRIEVING
Descriptions of techniques for water vapor remote sensing using near-IR channels were previously reported[1,5]. The remote sensing method is based on detecting the absorption by water vapor of the reflected solar radiation after it has been transmitted down to the surface, reflected at the surface, and transmitted up through the atmosphere to the sensor. The equivalent total vertical amount of water vapor can be derived from a comparison between the reflected solar radiation in the absorption channel, and the reflected solar radiation in nearby non-absorption channels (see Fig. 1). Fig. 1 shows the positions and widths of five MODIS near-IR channels marked in thick horizontal bars, and two-way atmospheric water vapor transmittance spectra for the tropical and sub-arctic winter models in LOWTRAN-7[7] with a solar zenith angle of 45 degrees and a nadir-looking geometry.
2. 1 The Relationship Between Tobs and Water Vapor
Remote sensing of water vapor can be based on a ratio of absorbing to non-absorbing channels (e.g., a ratio of the measured radiation at 0.94μm to that at 0.865μm). Kaufman and Gao retrieved column water vapor using channel ratio techniques[1,5]. A two-channel ratio of an absorption channel with a window channel gives the water vapor transmittance of the absorption channel. For example, the transmittance of the channel at 0.940, 0.905 and 0.036μm can be expressed as:
Tobs (0.940μm) = ρ*(0.940μm) / ρ*(0.865μm) (1)
Tobs (0.905μm) = ρ*(0. 905μm) / ρ*(0.865μm) (2)
Tobs (0.936μm) = ρ*(0. 936μm) / ρ*(0.865μm) (3)
Where, Tobs is transmittance, ρ* is the computed apparent reflectance at the top of the atmosphere for the specified channel. Fig. 2 shows the relationship between Tobs and total precipitable water vapor. It can be expressed by an exponential formula. Moreover, due to the saturation of the water vapor absorption[5,7], a square root of W is used as the independent variable:
Where, Tobs is transmittance, W is total precipitable water vapor. We do many experiments and get the coefficient of a and β for Malaysia. For Tobs (0.940μm), a=0.12 and β=0.651, for Tobs (0.905μm), a=0.025 and β=0.3, for Tobs (0.936μm), a=0.056 and β=0.60.
Fig. 1: Water Vapor Transmittance Spectra Curve[5]. Fig. 2: Transmittance Ratio – Total Water Vapor Curve[5].
2. 2 The Weighting Functions
Atmospheric water vapor has very different absorption coefficients over the band passes of MODIS channels centered near 0.936, 0.940, and 0.905μm. As a result, the three channels have different water vapor sensitivities under the same atmospheric condition. The strong absorption channel at 0.936μm is most sensitive under dry conditions, while the weak absorption channel at 0.905μm is most sensitive under humid conditions. Under a given atmospheric condition, the derived water vapor values from the three channels can be different. A mean water vapor value (W) is obtained according to the following equation:
W = f1w 1+ f2w2 + f3w3 (5)
Where, w 1, w 2 and w 3 are water vapor values derived from the 0.936, 0.940, and 0.905μm channels, respectively, and f1, f2, and f3 are the corresponding weighting functions. In Malaysia, f0.936=0.24, f0.940=0.40 and f0.905=0.36.
Fig 4a shows the column water vapor map of Peninsular Malaysia on 12 August, 2004. Kaufman and Gao (1992) have performed a lot of experiments and analysis on the retrieval of column water vapor from MODIS for different conditions and conclude that the accuracy of the retrieved column water vapor is about 87% in the cloud-free conditions[1,5].
3. SPECIFIC HUMIDITY
There is no direct method of estimating relative humidity and specific humidity accurately using remote sensing. It has been demonstrated that the water vapor in the atmospheric column can be determined very accurately by remote sensing method. However, a simple statistical regression between specific humidity and column water vapor were found by W. Timothy (1984)[8]. He had analyzed 9 years time series of monthly mean column water vapor (W) and specific humidity (Q) of temperate, subtropical and tropical zone and produced the scatter plots of Q versus W for three zones as shown in Fig. 3 a, b, c. Each curve is a quadratic regression and the bar represents 4×10-4 in Q. The scatters (r.m.s difference between the measured values and those predicted by the regression) are 3.3×10-4, 6.6×10-4 and 2.8×10-4 and the correlation coefficients are 0.97, 0.97 and 0.90 respectively.
Fig 3: Scatter plot of Q versus W[8].
Fig. 3 shows that subtropical stations have larger range of Q and the correlation coefficients however are very good (Fig. 3b). Tropical station show slightly lower correlation but the scatters are generally less than 4×10-4. Therefore the specific humidity (Q) in tropical zone can be calculated from column water vapor (W) according to the following experiential formula:
Q = -0.0252w2 + 1.2622w + 13.574 (6)
4. RELATIVE HUMIDITY
Relative humidity is the ratio of vapor pressure (e) and saturation vapor pressure (es):
RH = e / es (7)
vapor pressure (e) can be calculated from specific humidity(Q) and air pressure (P):
e = 0.622Q × P (8)
saturation vapor pressure (es) can be calculated from temperature(T):
Air pressure will decrease with the increment of elevation. Tab. 1 shows the corresponding relationship of Air pressure (P) with elevation (H). Air pressure (P) can also be calculated from elevation (H) according to the following experiential formula:
P = 1013.3 – 0.1038H (10)
Tab.1: The Look Up Table Of Elevation and Pressure.
Column water vapor can be retrieved from MODIS data while elevation can be obtained from DEM data and air temperature (T) can be obtained from weather data or retrieving from MODIS data. MOD07 of MODIS level-2 product provides atmospheric profiles. Since the accuracy of surface level air temperature from MODIS level-2 product is not validated in Peninsular Malaysia, hence the air temperature of weather data were used to calculate relative humidity of Peninsular Malaysia in this study. Once the column water vapor (W), air temperature (T) and elevation (H) were obtained, we can calculate the relative humidity using formula (6)-(10).
Fig. 4: Column Water Vapor, Air Temperature and Relative Humidity Map of Peninsular Malaysia.
We have calculated the relative humidity of Peninsular Malaysia at 11 a.m. on 12 August, 2004 (see Fig 4c) by using column water vapor, air temperature and DEM data. Fig. 4c shows that the relative humidity of most regions is about 85%. The relative humidity observed in mountain area is relatively higher due to the dense forest and lower air temperature. However, the relative humidity of some area is less than 73% due to the cloud effect.
5. CONCLUSION AND DISCUSSION
The column water vapor retrieved from MODIS with the accuracy of 87% has important applications in meteorology and hydrology. Combination of column water vapor with a resolution of 1 km and other earth and atmospheric parameters by MODIS, such as temperature, vegetation and fire mask will enable an intensive study of the environment and the effect of human activity on climate change
The air temperature interpolated from weather data is not logic in mountain area, for air temperature is much lower than that of plain areas due to the higher elevation. Therefore, in the area which the elevation is higher than 400m, the elevation must be used to correct the air temperature. In future study, air temperature from MOD07 atmospheric profiles of MODIS level-2 product will be validated and subsequently used to retrieve relative humidity of Peninsular Malaysia.
Because there was no relative humidity record from weather data at 11 a.m., the relative humidity retrieved from MODIS in Peninsular Malaysia in this study can not be validated. The validation will be performed in future.
ACKNOWLEDGMENT
The authors would like to thank Malaysian Centre for Remote Sensing for their help and support in this research. This data are provided by MACRES under MACRES Airborne Remote Sensing (MARS) programme.
REFERENCES
- B.-C. Gao and Kaufman, Y. J. Water Vapor Retrievals Using Moderate Resolution Imaging Spectroradiometer (MODIS) Near-Infrared Channels. Journal of Geophysical Research, 108, 13, 4389-23-43.2003.
- Gao, B.-C., and A. F. H. Goetz, Column Atmospheric Water Vapor and Vegetation Liquid Water Retrievals from Airborne Imaging Spectrometer Data, J. Geophys. Res., 95, 3549- 3564, 1990.
- Frouin, R., P. Y. Deschamps, and P. Lecomte, Determination from Space of Atmospheric Total Water Vapor Amounts by Differential Absorption Near 940 nm: Theory and Airborne Verification, J. Appl. Meteorol., 29, 448- 460, 1990.
- J.A.Reagan, K.Thome, B. Herman. And R.Gall, Water Vapor Measurements In the 0.94μm Absorption Band: Calibration, Measurements and Data Application. Proc. IGARSS, pp. 63-67, 1987.
- Kaufman, Y. J., and B.-C. Gao, Remote Sensing of Water Vapor in the Near IR from EOS/MODIS, IEEE Trans. Geosci. Remote Sens., 30, 871-884, 1992.
- Kaufman, Y. J., D. Tanre´, L. Remer, E. F. Vermote, D. A. Chu, and B. N.Holben, Operational Remote Sensing of Tropospheric Aerosol Over the Landfrom EOS-MODIS, J. Geophys. Res., 102, 17, 051-17,068, 1997.
- Kneizys, F. X., E. P. Shettle, L. W. Abreu, J. H. Chetwynd, G. P. Anderson, W. O. Gallery, J. E. A. Selby, and S. A. Clough, Users Guide to LOWTRAN 7, AFGL-TR-8-0177, Air Force Geophys. Lab., Bedford, Mass., 1988.
- W. Timothy Liu, Remote Sensing of Near Surface Humidity over North Pacific. IEEE Trans. Geosci. Remote Sens., 115-118, 1984.
