Sushma Panigrahy

Discrimination of rice crop grown under different cultural practices using temporal ERS-1 synthetic aperture radar data

Radar remote sensing has a significant role to play in remote sensing based crop inventory programmes due to its independence from cloud cover. In this study, an attempt has been made to evaluate the utility of temporal ERS-1 SAR data to classify rice crop grown in different growing environments. The sites represent four major types of lowland cultivation practice prevailing in India. Results showed more than 90% classification accuracy for all types of wetland rice using three-date SAR data. Data acquired during the early vegetative stage were found essential for high accuracy. The accuracy was mainly affected by the presence of rivers/streams in the scene. High accuracy was obtained for lowland intermediate and irrigated rice areas. A significant effect of wind was observed on the radar backscatter from stagnant water bodies but not on the rice fields during early growth stages. The study indicates the feasibility of operational use of ERS SAR data for estimation of areas of rice crop grown under lowland cultural practice.

Evaluation of RADARSAT Standard Beam data for identification of potato and rice crops in India

Abstract The Canadian satellite RADARSAT launched in November 1995 acquires C-band HH polarisation Synthetic Aperture Radar (SAR) data in various incident angles and spatial resolutions. In this study, the Standard Beam S7 SAR data with 45°–49° incidence angle has been used to discriminate rice and potato crops grown in the Gangetic plains of West Bengal state. Four-date data acquired in the 24-day repeat cycle between January 2 and March 15, 1997 was used to study the temporal backscatter characteristics of these crops in relation to the growth stages. Two, three and four-date data were used to classify the crops. The results show that the backscatter was the lowest during puddling of rice fields and increased as the crop growth progressed. The backscatter during this period changed from −18 dB to −8 dB. This temporal behaviour was similar to that observed in case of ERS-SAR data. The classification accuracy of rice areas was 94% using four-date data. Two-date data, one corresponding to pre-field preparation and the other corresponding to transplantation stage, resulted in 92% accuracy. The last observation is of particular interest as one may estimate the crop area as early as within 20–30 days of transplantation. Such an early estimate is not feasible using optical remote sensing data or ERS-SAR data. The backscatter of potato crop varied from −9 dB to −6 dB during the growth phase and showed large variations during early vegetative stage. Two-date data, one acquired during 40–45 days of planting and another at maturing stage, resulted in 93% classification accuracy for potato. All other combinations of two-date data resulted in less than 90% classification accuracy for potato.

Evaluation of classifiers for processing Hyperion (EO-1) data of tropical vegetation

There is an urgent necessity to monitor changes in the natural surface features of earth. Compared to broadband multispectral data, hyperspectral data provides a better option with high spectral resolution. Classification of vegetation with the use of hyperspectral remote sensing generates a classical problem of high dimensional inputs. Complexity gets compounded as we move from airborne hyperspectral to Spaceborne technology. It is unclear how different classification algorithms will perform on a complex scene of tropical forests collected by spaceborne hyperspectral sensor. The present study was carried out to evaluate the performance of three different classifiers (Artificial Neural Network, Spectral Angle Mapper, Support Vector Machine) over highly diverse tropical forest vegetation utilizing hyperspectral (EO-1) data. Appropriate band selection was done by Stepwise Discriminant Analysis. The Stepwise Discriminant Analysis resulted in identifying 22 best bands to discriminate the eight identified tropical vegetation classes. Maximum numbers of bands came from SWIR region. ANN classifier gave highest OAA values of 81% with the help of 22 selected bands from SDA. The image classified with the help SVM showed OAA of 71%, whereas the SAM showed the lowest OAA of 66%. All the three classifiers were also tested to check their efficiency in classifying spectra coming from 165 processed bands. SVM showed highest OAA of 80%. Classified subset images coming from ANN (from 22 bands) and SVM (from 165 bands) are quite similar in showing the distribution of eight vegetation classes. Both the images appeared close to the actual distribution of vegetation seen in the study area. OAA levels obtained in this study by ANN and SVM classifiers identify the suitability of these classifiers for tropical vegetation discrimination.

Comparative evaluation of the sensitivity of multi‐polarized multi‐frequency SAR backscatter to plant density

Interaction of synthetic aperture radar (SAR) with vegetation is volumetric in nature, hence SAR is sensitive to the variation in vegetation density. At the same time SAR is also sensitive to other target properties such as canopy structure, canopy moisture, soil moisture and surface roughness of the underlying soil. However, the sensitivity of SAR backscatter to the vegetation density depends upon the frequency, polarization and angle of incidence at which the SAR is operated. This paper provides comparative evaluation of the sensitivity of multi‐frequency and multi‐polarized SAR backscatter to the plant density of Prosopis juliflora, a thorny plant. Monitoring of P. juliflora is of importance as the state forest department introduced it to arrest the spread of desert. In carrying out this study, data from the SIR‐C/X‐SAR mission over parts of Gujarat, India, have been used. In the present study, the variation of multi‐frequency (L and C) and multi‐polarized (HH, VV and VH) SAR backscatter with plant density has been studied. The results clearly indicate that cross‐polarized SAR backscatter at longer wavelength is the appropriate choice for the quantitative retrieval of plant density.

A processing and software system for rice crop inventory using multi-date RADARSAT ScanSAR data

Abstract An operational crop survey program requires standardised procedures and software packages to meet the specified targets of timeliness and accuracy of estimates. Currently, the focus is to include Synthetic Aperture Radar (SAR) data in such a program, as these data are available from a number of sensors. A procedure has been developed to use multi-date SAR data for rice crop inventory. The steps were packaged together for ease-of-use and with minimal user interaction. The package, SARCROPS, is built around the EASI/PACE software. It is, at present, tuned for RADARSAT ScanSAR data. The package was used during the 1998–1999 and 1999–2000 seasons to estimate the rice area at state level in India with the participation of a number of interdisciplinary users. Around 45 and 89 scenes of ScanSAR data were used during these two seasons, respectively. This paper reports the details of the SARCROPS processing chain.

Early estimation of rice area using temporal ERS-1 synthetic aperture radar data a case study for the Howrah and Hughly districts of West Bengal, India

Abstract The characteristic temporal backscattering signature of rice crop grown under flooded condition was used to estimate rice acreage for a region in West Bengal, India. To date ERS-1 Synthetic Aperture Radar (SAR) data, one acquired within 30 days of transplantation and another after 30-40 days was found to be optimum for early estimation of rice acreage. The rice crop was found to be distinctly separable from forest, tree vegetation, village/urban areas. Misclassification of rice was observed mainly with water, waterlogged areas and fallow fields.

Deriving cropping system performance indices using remote sensing data and GIS

The cropping system approach is a holistic management of variant and invariant resources to optimize the food production. Various indices are used to assess and evaluate the efficiency and sustainability of the systems. These indices are generally computed from the data collected by traditional survey methods that are time consuming and non‐spatial. An attempt has been made to derive such indices using satellite remote sensing data for the state of West Bengal, India. Three indices—Multiple Cropping Index (MCI), Area Diversity Index (ADI) and Cultivated Land Utilization Index (CLUI)—were attempted. Multi‐date, multisensor data from Indian Remote Sensing Satellite (IRS) and Radarsat Synthetic Aperture Radar (SAR) were used to derive cropping pattern, crop rotation, and crop calendar. Crop type, acreage, rotation and crop duration were used as inputs to compute the indices at district and state level. The indices were categorized as high, medium and low to evaluate the performance of each of the 16 districts. The average MCI of the state derived was 140. At district level it varied from 104 to 177. The average ADI of state was 2.5 and varied from 1.5 to 5.0.

Role of middle infrared bands of Landsat thematic mapper in determining the classification accuracy of rice

Abstract Accuracy of discriminating rice crop from other vegetation classes investigated using different band combinations of Landsat Thematic Mapper (TM) data over an area in Orissa state of India. Colour-infrared (CIR)aerial photographs of 1:18,000 scale were used to prepare base maps at 1:5,000 scale to identify training areas. A maximum likelihood classification of the training class pixels were done using three band combinations viz. TM 1234, TM 2345, TM 2347, Classification accuracy using different band combinations was computed from the error matrices of classified training class pixels. The results showed that the classification accuracy of rice was significantly greater in TM 2345 and TM 2347 band combinations than in TM 1234.

Evaluation of hyperspectral indices for LAI estimation and discrimination of potato crop under different irrigation treatments

In this study, various hyperspectral indices were evaluated for estimation of leaf area index (LAI) and crop discrimination under different irrigation treatments. The study was conducted for potato crop using the spectral reflectance values measured by a hand‐held spectro‐radiometer. Three categories of hyperspectral indices, such as ratio/difference indices, multivariate indices and derivative based indices were computed. It was found that, among various band combinations for NDVI (normalized difference vegetation index) and SAVI (soil adjusted vegetation index), the band combination of the 780∼680, produced highest correlation coefficient with LAI. Among all the forms of LAI and VI empirical relationships, the power and exponential equations had highest R 2 and F values. Analysis of variance showed that, hyperspectral indices were found to be more efficient than the LAI to detect the differences among crops under different irrigation treatments. The discriminant analysis produced a set of five most optimum bands to discriminate the crops under three irrigation treatments.

Analysis of temporal backscattering of cotton crops using a semiempirical model

To develop an operational methodology for estimating soil moisture and crop biophysical parameters and to generate a crop cover map, backscattering signatures of vegetation canopies are investigated using multitemporal Radarsat synthetic aperture radar (SAR) data over a predominantly cotton-growing area in India during low to peak crop growth stage. A simple parameterization of the water-cloud model with volumetric soil moisture content (m/sub v/) and leaf area idex (LAI) is used to simulate the microwave backscattering coefficient (/spl sigma//sup 0/), as it is found to be a good candidate for operational purposes as demonstrated by several workers in past. The influence of crop height (H), LAI, and m/sub v/ on /spl sigma//sup 0/ is investigated during peak crop growth stage. A linear relationship between LAI and crop height is derived semiempirically, and a linear zone is chosen for analysis during the peak crop-growing stage. Estimation of average volume fraction of leaves (V~/sub l/) and attenuation factor (L) by two different approaches is discussed: 1) using linear relationship between LAI versus crop height and 2) from the water-cloud model parameter (/spl kappa/) estimation by iterative minimum least square error approach. It is observed that model-estimated parameters agree well with the measured values within an acceptable error limit. At lower soil moisture, m/sub v//spl cong/0.02(cm/sup 3//spl middot/cm/sup -3/), the dynamic range of /spl sigma//sup 0/ is found to be about +5 dB for 0-70 cm of crop height but monotonously decreases to null at a transition point, having m/sub v//spl ap/0.38(cm/sup 3//spl middot/cm/sup -3/). A positive correlation is found between backscattering coefficient and crop height till this transition point but shows a negative correlation beyond that, signifying the predominant attenuation by vegetation over soil. Differential moisture sensitivity (d/spl sigma//sup 0//dm/sub v/) of the backscattering coefficient decreases by half from 20.55 dB/(cm/sup 3//spl middot/cm/sup -3/) for dry and bare-field conditions to 10.68 dB/(cm/sup 3//spl middot/cm/sup -3/) for wet and crop-covered fields (m/sub v/=0.38cm/sup 3//spl middot/cm/sup -3/, H=70cm), whereas differential crop height sensitivity (d/spl sigma//sup 0//dH) varies from 0.22-0.03 dB/cm for bare-field conditions to crop-covered fields with crop height 70 cm. It is found that the percentage of relative error is smallest (2.27%) for LAI and attenuation factor estimation using the value of V~/sub l/, from LAI models, whereas it is 4.25% when estimating from the attenuation coefficient (/spl kappa/) from the model.