International Journal of Remote Sensing

KEY, J. R.; MASLANIK, J. A.; BARRY, R. G.

doi: 10.1080/01431168908904014pmid: N/A

Abstract Where spatial boundaries between phenomena are diffuse, classification methods which construct mutually exclusive clusters seem inappropriate. The fuzzy c-means (FCM) algorithm assigns each observation to all clusters, with membership values as a function of distance to the cluster centre. The FCM algorithm is applied to Advanced Very High Resolution Radiometer (AVHRR) data for the purpose of classifying polar clouds and surfaces. Careful analysis of the fuzzy sets can provide information on which spectral channels are best suited to the classification of particular features and can help determine likely areas of misclassification. General agreement in the resulting classes and cloud fraction was found between the FCM algorithm, a manual classification and an unsupervised maximum likelihood classifier.

BOUMAN, B. A. M.; GOUDRIAAN, J.

doi: 10.1080/01431168908904015pmid: N/A

Abstract Direct derivation of biomass from radar backscattering gives erratic results so this paper discusses another method in which biomass was not estimated directly but was found as the accumulated value of the estimated crop growth rate. The estimation was based on soil crop cover and global radiation. The relationship between soil cover in the optical and microwave regions was investigated. Analysis of the methodology showed that improvement is obtained in comparison with the direct estimation method. Despite variation in parameters for different years, a remarkable consistency in estimated biomass was observed. Nevertheless, measurements of radar backscattering still suffer from too much variation to be reliable for biomass estimation.

VYGODSKAYA, N. N.; GORSHKOVA, I. I.; FADEYEVA, Ye. V.

doi: 10.1080/01431168908904016pmid: N/A

Abstract Based on the Goudriaan (1977) reflectance model, the choice of vegetation indices (VI) was substantiated. The Vis are optimal to reconstruct the key parameter of plant canopy, i.e. the phytoelements' relative surface. The numerical experiment simulated the changes within a wide spectrum of plant colour, spatial orientation of phytoelements and soil brightness. Under analysis were nine combinations of canopy spectral reflectances in the visible and near-infrared wavelength intervals. Among these are indices widely used in practical remote sensing of vegetative targets: the simple ratio (VI1]), normalized difference (VI8) and the perpendicular vegetative index (VI9). The optimal VIs was chosen by three criteria; (1) stability in tendencies in the VIs changes as a function of the phytoelements' relative surface, (2) sensitivity of the VIs to variations in the phytoelements' relative surface and (3) the impact of other parameters on the interdependence between the VIs and the phytoelements' relative surface. It was discovered that the advantages of VIs as compared with the canopy spectral reflectance are in the VIs' ability to improve the informative significance of remote sensing for solving inverse tasks. Using VI1 VI8 and VI9 for defining the relative surface of phytoelements permits one to minimize the impact of soil brightness. It is feasible to employ VI1 and VI8 for conditions of space-time variation in the orientation of phytoelements and it is better to use VI9 for conditions of variation in plant colour.

ALI, ANWAR; QUADIR, DEWAN A.; HUH, OSCAR K.

doi: 10.1080/01431168908904017pmid: N/A

Abstract This paper examines the applicability of NOAA satellite AVHRR imagery to monitoring and studying the river floods and associated hydrological conditions in Bangladesh and the adjoining regions. The flood period considered is that of the last 10 days of September 1984, when Bangladesh experienced one of the worst floods in recent years. Imagery from dry winter conditions have also been used for comparison. The analysis involved the calibration of the NOAA-7 Advanced Very High Resolution Radiometer (AVHRR) IB data tapes, linearization of pixel sizes and geometric rectification of the image to ground reference points. The imagery showed that a vast area of Bangladesh was flooded and rivers were highly turbid. Comparisons have been made between the major rivers, particularly the Ganges and the Brahmaputra, during the flood season. It has been found that the Ganges has higher albedo (is more turbid) and is warmer than the Brahmaputra during the flood season. The Brahmaputra shows a gradual increase in albedo level downstream, the reverse of which is observed in the Ganges. The turbidity and temperature distribution in the coastal area have also been studied. The results of this study are based on the physical interpretation of the high-quality AVHRR radiometric imagery. It has not been possible here to supplement and test the results of this study with conventional surface measurement data, which is intended to be done in the future.

JOHANNESSEN, JOHNNY A.; JOHANNESSEN, OLA M.; HAUGAN, PETER M.

doi: 10.1080/01431168908904018pmid: N/A

Abstract The Chrysochromulina polylepis bloom in Skagerrak in May 1988 had severe consequences for the sea farming industry. Following the unexpected occurrence of the bloom combined remote-sensing and model studies of the mesoscale circulation of the Norwegian coastal current were performed in support of the daily forecasts of the advance of the algal front. A passive tracer included in the model simulated the drift path of the algae demonstrating the impact of meanders and eddies. Qualitatively the results of the simulation were in good agreement with the direct observations until the bloom culminated, suggesting dominance of the physical processes. In contrast, biological processes lead to the algal mortality during the retreat.

THOMSON, A. G.; MILNER, C.

doi: 10.1080/01431168908904019pmid: N/A

Abstract The mean annual population densities of free-ranging sheep in upland North Wales were significantly correlated with near-infrared (0.76–0.90 μm) radiance recorded by the Landsat Thematic Mapper. The relationship between sheep density and the normalized difference vegetation index (NDVI) indicated that high sheep densities always occurred on vegetation with high NDVI values, but low sheep densities may occur on vegetation with either low or high NDVI values.

EDWARDS, GEORGE J.; CURE, WILLIAM W.

doi: 10.1080/01431168908904020pmid: N/A

Abstract Abstract. A scanning spectroradiometer was used to measure the optical densities of an aerial photograph of an experimental field in which soybeans were growing in response to different concentrations of ozone, an air pollutant. The plants were growing in 3 m diameter, 2.4 m high open-top exposure chambers. Correlation coefficients among the film densities, plant yield and visual estimates of non-green leaf area for the 16 test plots were highly significant (p<001); those with ozone treatment concentrations were significant (p<0-05). The early senescence induced by this form of environmental stress can thus be detected by film density differences, and these differences are well correlated with ground indicators of crop condition.