Contributions of the remote sensing techniques for the estimation of thermal field.
Mihailescu, Denis Iosif ; Savin, Elena L. ; Poenaru, Violeta Domnica 等
1. INTRODUCTION
Romania has a temperate continental climate with cold winters and
hot summers. During spring time occurs numerous episodes of cooling
ground frost. Average date of last spring frost calculated for the
period 1961-2000 in the plains and hills is generally in the second
decade of April.
Climate scenarios for the country projected average temperature
increases, which will cause an early development of vegetation and
therefore greater sensitivity to cold. In the context of climate change
with many uncertainties concerning the intensity and frequency of
freezing phenomenon, shaping the spatial scale of minimum temperatures
will allow accurate estimation of the consequences of climate change and
planning for adapted methods (use of plants less sensitive are more
likely to experience frost Spring).
The main objective of this study is to develop frost risk maps of
measured data from weather stations and the satellite data and
presenting a case study in Romania (March 2009)
[FIGURE 1 OMITTED]
2. METHODOLOGY
Analysis of weather data minimum soil temperature and minimum air
temperature shows a large variability and large spatial variability of
frost damage caused phenomena (Tait & Zheng, 2003). Protect plants
from the effects of low temperatures is a matter of considerable
importance in agricultural and horticultural production of fruits and
vegetables. Accurate forecasting and monitoring of the phenomenon of
frost is needed and knowing of areas most likely to be affected.
Degree and specific type of frost damage depends on several
factors: the type of crop, its variety, maturity, tissue, culture size,
phenological stage, temperature decrease rate and the time during which
the culture is exposed to lower temperature. Plant species differ
greatly in terms of susceptibility to damage caused by cold and points
of freezing are different to each other.
Remote sensing offers unique opportunities in this regard by the
channel thermal infrared satellite images that allow spatial mapping of
soil surface temperature (Stancalie, 2005).
In this context, estimating a minimum temperature scale in specific
situations, spring frost is the main objective of this study. A mapping
is presented as a result of spatial interpolation of data coming from
the network of weather stations and thermal infrared satellite imagery
of MODIS and MSG.
[FIGURE 2 OMITTED]
The MODIS instrument provides high radiometric sensitivity (12 bit)
in 36 spectral bands ranging in wavelength from 0.4 [micro]m to 14.4
[micro]m. Land surface temperature is one of the MODIS Land products.
Emitted spectral radiance L at wavelength [lambda] from a surface at
thermodynamic temperature [T.sub.s] is given by multiplying the Planck
function by spectral emissivity [epsilon]([lambda]) (Zhengming, 1999).
L([lambda], T)= [epsilon]([lambda])B([lambda], Ts) (1)
In general, azimuthally depended radiance in an absorbing,
emitting, and scattering layer is governed by the monochromatic
radiative transfer equation (Caselles et al., 1997).
Land surface temperature is one of the main quantities governing
the energy exchange between surface and atmosphere. LST based on
MSG-SEVIRI measurements is an operational product of the Land Surface
Analysis--Satellite Application Facility (LSA-SAF)
[FIGURE 3 OMITTED]
In case of satellite measurements, the surface associated with a
pixel is not in thermal equilibrium and is composed of areas with
different characteristics of reflectance and emissivity.
A surface temperature is well defined for homogeneous surfaces at
thermal equilibrium. Remote sensing sensors for determining the
temperature related parameters, such as brightness temperature, Tb,
radiation temperature, Tr, Radiometric temperature T. Brightness
temperature is the temperature of a blackbody having the same radiation
emitted from a given area. Radiometric temperature depends on surface
temperature distribution, of its emissivity and the spectral range to be
measured (Stancalie et al., 2006)
To achieve frost risk map, interpolation of LST data was performed,
data obtained from weather stations, processing of satellite images,
consisting of georeferencing (bringing images in stereo 70 projection),
redesign, mosaic and calculating pixels values in ENVI 4.5 software.
Eventually we realized the correlation between meteorological and
satellite data to observe and interpret the results.
3. RESULTS
Using weather data daily minimum soil temperature were made frost
risk maps for March with thresholds of 0, -1, -2, -3 [degrees] C.
To analyze the cooling of March 2009, minimum soil temperature maps
were made and satellite images were processed: MODIS, MSG and SPOT
VEGETATION.
Minimum soil temperature data from 26 and March 27, were measured
at 130 stations and were interpolated and compared with soil surface
temperature product from MODIS satellite images, and MSG. MODIS images
have spatial resolution of 1km and are daily. MSG images have a spatial
resolution of 4 km and 15 minutes temporary resolution that allows a
synthesis and obtaining an image without clouds. To estimate the effect
of low temperatures on vegetation development were used images SPOT
VEGETATION NDVI, FPAR and FVC. The images were processed and integrated
into a GIS environment.
[FIGURE 4 OMITTED]
4. CONCLUSION
The role of satellite data to determine the thermal field has
become particularly important given the development of remote sensing
systems and products generated by them.
Due to temporary good resolution (15 min.), land surface
temperature from MSG, allow an image synthesis without clouds.
Land surface temperature from MODIS images allows estimation of
temperature surface for each pixel of lxl km.
Data obtained from satellite imagery can be used for numerical
forecasting models.
It can be seen much better spatialization of information, making
comparison between data obtained from weather stations and data from
satellite imagery. Data from weather stations provide information in one
point and does not cover a large area while satellite imagery provides
more detailed information, a value on each pixel of lxl km, in case of
MODIS images.
The analysis of meteorological data shows a high spatial
variability for the low temperatures as well as damages due to frost
events. Indeed, important differences in the temperatures of weather
stations as close as a few kilometres apart, sometimes records coming
from the same hill side, are observed. These contrasts are accounted for
by numerous factors, the main ones being the atmospheric conditions, the
topography and soil state. In this context, an estimate of minimum
temperatures at a fine scale using remote sensing data in spring frost
situations is useful.
Because of spatial resolution of 1 km (MODIS) and 4 km (MSG), the
results of research on smaller areas may be inaccurate. The next step of
research will be used of other satellite images, for example Landsat 7
(band 6 thermal infrared) with spatial resolution of 60m.
5. REFERENCES
Caselles, V., Valor, E., Coll, C., Rubio, E., 1997: "Thermal
band selection for the PRISM instrument 1.Analysis of
emissivity-temperature separation algorithms", J. Geophs. Res.,
102, D10, 11145-11164.
Stancalie G., Simona Catana, Anis,oara Irimescu, Dr. Elena Savin,
A. Diamandi, Alina Hofnar, Dr. Simona Oancea "Contribution of Earth
Observation data supplied by the new satellite sensors to flood
management", Nato Science Series: IV. Earth and Environmental
Sciences, vol. 72, p. 287-304, ISBN-10 1-4020-4901-3
Stancalie G., "Study on the utilization of RS data to estimate
land surface temperature", COST actions 718 Meteorological
applications for agriculture, Use and availability of meteorological
information from different sources as input in agro meteorological
models, 2005, 217-226
Tait, A., Zheng, X., "Mapping Frost Occurrence Using Satellite
Data", National Institute of Water and Atmospheric Research,
Wellington, New Zealand, Journal of Applied Meteorology, volume 42, 2003
Zhengming, W. (1999). MODIS "Land-Surface Temperature
Algorithm Theoretical Basis Document", Available from:
http://modis.gsfc.nasa.gov/data/atbd/atbd_mod11.pdf
