S. S. Ray

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.

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.

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.

Mitigating future climate change effects by shifting planting dates of crops in rice-wheat cropping system

The present study focuses on (1) impact of climate change scenarios on crop yields in rice–wheat cropping system in central Indian Punjab and (2) assessment of shifting trans-/planting date as an adaptation strategy to mitigate that impact. Climate scenarios were derived from General Circulation Model’s simulations viz. Hadley Center Coupled Model Version 3, Australia’s Commonwealth Scientific and Industrial Research Organization Mk2 and Second Version of Canadian Center for Climate Modeling and Analysis Coupled Global Climate Model. Crop duration and yields were simulated with CropSyst model. Simulation analysis showed decline in crop yields depending upon changed levels of temperature and CO2 in different scenarios. The magnitude of yield decline was highest in 2080 under the A2 scenario of the CCCMA model. Under the changed climate, shifting trans-/planting date from 171st to 178th for rice and from 309th to 324th Julian day for wheat resulted in least reduction in crop yields and surfaced as a practical adaptation measure for sustaining yields in future.

DERIVATION OF INDICES USING REMOTE SENSING DATA TO EVALUATE CROPPING SYSTEMS

This paper presents the work done in Bathinda District of Punjab state of India for evaluating the cropping system efficiency using multi-date, multi-year and multi-sensor satellite based remote sensing data along with various spatial and non-spatial collateral data. Three efficiency indices, such as Multiple Cropping Index (MCI), Area Diversity Index (DI), Cultivated Land Utilization Index (CLUI), have been worked out to characterize the cropping systems. The salient findings point out that, the MCI has, increased remarkably. A further increase is possible by only taking a third crop. The ADI has increased in kharif (rainy) season, due to introduction of rice in the cotton belt, however in rabi (winter) season the ADI has reduced nearly to one, showing it to be a mono-cropped situation. The CLUI is low (> 0.5) in many blocks, showing there is a great scope to improve it. Since in summer the land is remaining unutilized, a summer crop can very well be taken up to improve it.

Forecasting wheat yield in Punjab state of India by combining crop simulation model WOFOST and remotely sensed inputs

An attempt has been made to assimilate remotely sensed input data in mechanistic crop simulation model World Food Studies (WOFOST) for in-season wheat yield forecasting in Punjab state of India. Spatial weather data at ‘5 km × 5 km’ grid were generated through interpolation of daily available weather data. Grid-wise sowing date was estimated from time-series normalized difference vegetation index (NDVI) data product from vegetation sensor of SPOT satellite (SPOT-VGT). The leaf area index (LAI) derived from remotely sensed data was used in the simulation model WOFOST for predicting spatial yield. The simulated wheat grain yield for each grid was aggregated to district level using the actual wheat fraction for each grid derived from remote sensing-based wheat crop map. A comparison was made between the estimated yield and that reported by Department of Agriculture. The procedure was repeated for three crop seasons to check the reliability. The results indicated that this technique could be used for spatial yield prediction at regional level with a root mean square error (RMSE) of <0.4 tonnes ha−1 at state level.

Analysis of cropping pattern changes in Bathinda district, Punjab

A study was conducted in the Bathinda district of Punjab state for mapping the cropping pattern and crop rotation, monitoring long term changes in cropping pattern by using the satellite based remote sensing data along other spatial and non-spatial collateral data. Multi-date IRS LISS I and IRS WiFS sensor data have been used for this study. Cropping pattern maps and crop rotation maps were generated for the years 1988-89 and 1998-99. The present study has shown the increase of cropping intensity significantly, mainly due to increase in rice area. However, crop diversity has decreased mainly due to decline in the area under the minor crops like pearl millet, gram, rapeseed/ mustard. There is increase in area coverage of cotton-wheat and rice-wheat rotation, at the expense of the minor crops.

Discrimination of wheat crop stage using CHRIS/PROBA multi-angle narrowband data

Multi-angular narrowband compact high-resolution imaging spectrometer (CHRIS) on-board the project for on-board autonomy (PROBA) data of 18 March 2008 were used in this study to discriminate three different growth stages of wheat crop grown in the Central State Farm of Suratgarh, Rajasthan, India. Results showed that the off-nadir view angles performed better than nadir viewing for crop stage discrimination. Among all the off-nadir viewing angles, −55.37° view angle (in the backward-scattering direction) had the highest normalized distance between the crop stage classes. Based on the analysis, the five best bands were identified as 630, 660, 674, 705 and 712 nm for separating wheat at different stages.

A GIS and remote sensing based approach for siting cold storage infrastructure for horticultural crops: A case study for potato crop in Bardhaman district, West Bengal, India

A pilot study was taken up to evolve an optimum plan to locate cold stores for potato in Bardhaman district of West Bengal, India, a leading potato growing area. Remote sensing data from IRS sensors was used to delineate the potato growing area. Road and village information was derived from high-resolution remote sensing data and 1:50,000 scale Survey of India topographic maps. The analysis showed that the present capacity of cold stores can cater to around 40 percent of production. A combination of buffering and location-allocation analysis was performed using Art/Info software. Total 57 cold storage sites with average capacity were identified. Further, analysis was carried out for phase wise development of sites according to priority.

Identification of indices for accurate estimation of anthocyanin and carotenoids in different species of flowers using hyperspectral data

ABSTRACT Quantification of accessory pigments is very useful tool for assessment of quality of flowers. Available techniques for the quantification of accessory pigments are either time-consuming or highly laborious. However, till date, very few researchers have managed to provide an accurate solution of this problem. In present study, we have developed a model for accurate estimation of anthocyanin and carotenoids in different species of flowers using hyperspectral data. Numbers of bands sensitive towards the variation in the flower pigments were identified using different techniques such as correlation and stepwise discriminant analysis. Regression model developed using published indices worked well and gave moderately good results with highest R2 of 0.60 for anthocyanin and 0.66 for carotenoids. Likewise, developed simple ratio (SR) index (SR 331/581) gave best results for anthocyanins (R2 = 0.67) and SR 331/631 gave best results for carotenoids with R2 = 0.68.