2004 International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004
0,7 > 40 correlatio 1
Æ 3 1 CA en rte |
T. E after rain ] SA
E , A |
» 5 0,6 4 |
"the 705 —>-10 |
lyses 3 —» 45 041
nsity d 4} jum
NWS, Temporal 04 T r r r
phase lag 0 45 60
a, of (days) -0,2 +
ients 0,4 +
The er
-0,6 4 rainfall
been | (days)
-0.8 4
1, the =
from ; se
4 rrr te rem rca
best 10 20 30 40 SO 60 70 80 90 100 110 120 130 140 150 160 170 180 correlation T >
accumulated rainfall (day) p nes (NDVI x ranfall)
the | before rain
ghest 3 : : m ™ Nir 0,3
a; f Figure 4. Correlation coeficients between Terra-MODIS NDVI
€..0 2 TA : ; 0,5 4
images and total rainfall preciptation (mm). f
0,4 4
— 0,2 + |
0 at in|
AF Tod The values of Terra-MODIS IVDN had been correlated with the 425177 wr oss |
A amount of rains, the data used were supplied by the |
5 i NN : 0,4 + |
metcorological station of Santa Rita do Passa Quatro |
: municipalty (C4-107, DAEE). Each point of the image was ey |
„74 ‘ i . . " . . . ... - e |
s correlated with the precipitation accumulated in milliliters of 03 1 |
1 . S . 4 |
Musis rain, of 50 up to 360 previous days. Figure 4 are shown the Load
' graphically the averages of the correlation coefficient of each
| point of the image correlated with the pluvial precipitation. Figure 5. Correlation coeficients between Terra-MODIS NDVI
MP images and total rainfall preciptation (mm).
i For the conditions without temporal phase lag (equal zero), the
curves had been gotten with approximately 120 days (4 months)
v of accumulated precipitation. However, if consider the temporal 4. CONCLUSIONS
RE phase lag, the curve had been gotten with 360 and a positive
ni displacement of the IVDN of 95 days. The variations of In this work it was possible to establish, of quantitative form.
un temporal phase are not equal during all the vear. The the relations between the more important climatic factors in the
B coefficients of correlation and the respective displacements of seasonal variation of the IVDN gotten for orbital images. It was
phase in the period of daily before rain period and after-rainfall possible to define the maximum and minimum intervals of
period are represented in figure 5. IVDN in conditions where probably the hidric availability and
es of fotoperíodo had been more important to the sazonal variation.
es ( - : rs :
had I'he cases of low extreme temperatures had a effect of reduction
€ ^ . " . oy jc TE re Snir
ge of The gotten results indicate that the relation between the field of the values of IVDN of the grass physiognomies. The effect
üclure data and of the orbital sensors of the vegetation still are of the low temperatures will be sharper in the beginning of the
/ with dependent of the integration between the sampling methods. In dry period and diminish gradually with elapsing of this period.
| the case of the measures of IAF, an inverse relation between the Some difficulties had been identified to correlate parameters of
space resolution and the coefficients of correlation exists, the the vegetation and the spectral indices of vegetation. This duc
det best result had been gotten for the images of the Ikonos- to, mainly, to the characteristics of the methods and the absence
E( : s ed as ; "+ ; : : ; ; s
Santos Multispectral. This inverse relation already was noticed by of investigation about the characterization of the structure of
Se | i ^ di. : . c J . ’ ; - . . ; ;
all for other authors (White er al., 1998). Still related to the sampling, leaves in the interior of the cronw that compose the vegetal
| been another problem is that it deserves to be investigated the covering. The results of this work can serve to the global
. . . ^ »efimatfoes T ar: stere > > ze 0etatı M > I J
ontrol depthness of the crown and its structure in the interior of the estimates of parameters of the vegetation Using the IV DN,
Wil crown of the trees being of basic importance to consider the fenologia of each
LI vr. 4 . . : . . .
| thev Cerrado physiognomies and as well as its seasonal variation.
ations
Beyond the structural problems, would be necessary methods of a
sampling of the vegetative fenologia of the arboreal species, 5. REFERENCES
including its relations with lianas and epifts. The effect of the
epifts already had been studied in barren vegetations (Karnieli
et al., 1996). Beyond these effect, the burlap can influence the
IVDN in the dry periods (vanLeeuwen & Huete, 1996).
ANDERSON, G.L.; HANSON, J.D.; Haas, R.H. Evaluating
Landsat Thematic Mapper Derived Vegetation Indices for
Estimating Above-Ground Biomass on Semiarid
Rangelands. Remote Sensing of Environment, New York,
45:165-175. 1993.
BATISTA, G.T.; SHIMABUKURO, Y.E.; LAWRENCE, W.T. The
long-term monitoring of vegetation cover in the
397
