NOCTURNAL MONITORING OF THE LAND SURFACE TEMPERATURE VARIATION
IN THE RIO GRANDE DO SUL STATE USING AVHRR/NOAA SATELLITES
A. Gusso* *, D. C. Fontana?
? CEPSRM, Centro Estadual de Pesquisas em Sensoriamento Remoto e Meteorologia
Universidade Federal do Rio Grande do Sul, P.O. Box 15044, Porto Alegre, Brasil - anibal.gusso@ufrgs.br
? Faculdade de Agronomia
Universidade Federal do Rio Grande do Sul, Porto Alegre, Brasil - dfontana@vortex.ufrgs.br
PS WGVII/2
KEY WORDS: Climate, Retrieval, Surface, Temperature, Temporal, Observation, AVHRR/NOAA
ABSTRACT:
The surface temperature is the net result of the climatic energy balance to the environmental process at the Earth. Therefore, accurate
estimation of land surface temperature (LST) can be regarded as an important information to global and regional monitoring as much
as the understanding of the eletromagnetic radiation and matter interactions over the surface of the Earth. A monitoring study of the
nocturnal decreasing temperature of the land surface is developed. In the present study, the main problem is to verify the relationship
of decreasing temperature due to the rate of loss of land surface energy that can occur in the region by means of the temporal satellite
survey, compared with ground based stations. In order to quantify it, a linear regression to makes possible an operational
understanding is developed. The thermal decreasing trends with the spectral reply of the soil coverage and the land surface
emissivity has been analyzed. For the LST estimate it is used a published split-window algorithm, with strongly dependence from the
surface emissivity. The algorithm has been evaluated over the Rio Grande do Sul State in the south of Brazil, with good accuracy
and it takes data from two spectral bands in the thermal infrared radiation. The method to estimate LST measurements and emissivity
parameter use both data collect from Advanced Very High Resolution Radiometer (AVHRR) instruments from National Oceanic
and Atmospheric Administration (NOAA) satellites. Although the pioneering character in the estimate of the LST through the
satellites of NOAA series in the Rio Grande do Sul State, this study intends also, to contribute for the improvement of the
technologies of analysis in the orbital tracking for numerical forecasting and to prevent and esteem the prejudicies in crop fields in
Brazil. This paper also presents a discussion for the emissivity problem, with regards to the studied region.
1. INTRODUCTION
The land surface temperature (LST) is an important factor
controlling most physical, chemical and biological processes in
the Earth. Knowledge of the LST is necessary for many
environmental studies and management activities of the Earth
surface resources (Li & Becker 1993).
It is possible to obtain measures of the radiance data from
space-borne devices by means records of the emitted energy
directly from the earth surface. In order to obtain this parameter
from space radiometry in the thermal infrared part of the
eletromagnetic spectrum, it is necessary to take into account
emissivity and to correct the recorded signal for the
perturbations created by the atmosphere along the path between
the Earth's surface and the sensor (Becker & Li 1990).
McMillin (1975) proposed a method called split-window based
on the differential absorption in two adjacent spectral windows
in the thermal infrared for correct the atmospheric effects.
The LST results from energy exchange at the surface and
satellite-based monitoring can be regarded as an important
prerequisite of regional or global observations of surface water,
energy and radiation budgets (Andersen, 1997). This is because
accurate estimation of LST implies understanding of the
energy/matter interactions and physical parameters that change
in space and time in the biosphere. In the last 10 years, very
* Corresponding author.
complex models have been extensively applied to the accurate
of LST estimation by means space-borne devices for monitoring
some bio-geo-chemical processes over the Earth. However, a
' detailed knowledge of the processes that leads to errors in the
LST estimations are still wainting to be addressed for.
The parameter temperature is classically referred as the kinetic
movement of the particles inside the matter. On other hand,
LST retrieved from satellite generally is defined as the skin
temperature of the ground surface due to the longwave thermal
infrared radiation that escapes from the surface. However, the
ground surface is far from a skin or homogeneous surface with
two dimensions (Vogt 1996 apud Qin & Karnieli 1999).
Usually it is compared to various objects on the surface and
some of them such as vegetation may be best described in three
dimensions, even so, several times they are not conservative on
time scales. The difference Tskin — Tair varies particularly with
the surface water status, the roughness leght and the wind
speed. Such physical conditions often makes the relation
between Tair and LST into the biosphere thermodynamics
understanding a lot obscure.
In this paper the authors compare a short-term monitoring of
LST and Tair variations by means of two different night-time
satellite overpasses. This informations can be very useful for
agrometeorology in the growing crop fields. The objective is to
156
International /
MIC RUE
observe the r
regard to the a
The data wer
orbital scann
Radiometer” (
and Atmosp
Operational
Currently, in
which provide
polar orbiting
able to collect
ocean, and atn
Many resear
emissivity anc
1993) or theo
1996). In thi:
(1996) algorit
from measure:
This study ;
distributed ov
Km”, at the S
Longitudes
27°04’49’’S a
The study are
and crop field
shrub and gras
The satellite «
angle overpa:
received, anal
files at the €
Remoto e Me
located at coo
The Table 1 p
based-stations
surface in the
Tabl
Statio!
Bajé
Bom Je
Caxias
E. do $
Farrou
Iraí
Porto /
Quaraí
Santa I
S. Vitó
Sào Lu
E c Oo) Q ui ££ uh ox
It were select
16 with two «
High Resoluti
in Table 2.
Table 2. Selec
