PROJECT TITLE :                        

Development of methods for in-season monitoring of sugarcane crop in Peninsular India using Earth Observations for precision water and nitrogen management.

Sl.NoTeam DetailsTeam Details
Dr. V.C. Patil, Director
K.J. Somaiya Institute of Applied Agricultural
Research, Sameerwadi – 587 316
Taluka: Mudhol, Dist: Bagalkot,
Karnataka, India,
Phone: (91-08350) 260046 / 47 /48
Fax: (91 – 08350 – 260037)
Mobile No. 07022260486
Email: patil.vc@somaiya.com
Dr. Sergey Bartalev,
Head of Terrestrial Ecosystems
Space Research Institute (IKI),
Academy of Sciences, 117997, 84/32
Profsoyuznaya str., Moscow, Russia
Fax:  +7 495 913 30 40

STARTING DATE                                                    :January, 2019

DURATION                                                               :Two years

OBJECTIVES                                                                         :

1. To identify spectral signatures of sugarcane to discriminate different sugarcane varieties and

     growth stages.

2. To develop methods for developing spatial crop water stress index maps.

3. To develop methods for spatial nitrogen nutrition index maps.

4. To develop methods for preparing precision irrigation water and nitrogen application prescription maps.


Discrimination of different sugarcane varieties and growth stages:

Time series Landsat 8 / Sentinel 2 / Hyperion images (need based)will be evaluated for discrimination of varieties and growth stages of sugarcane and for assessing nitrogen and water status of sugarcane crop. The study area covers Bagalkot and Belgaum districts of Karnataka. Time series satellite data will be evaluated for the discrimination of ruling sugarcane varieties cultivated in the study area such as Co 86032, CoC 671 and Co 91010 as well as the four distinct growth stages of sugarcane crop such as:germination phase, tillering (formative) phase, grand growth phase and maturity &ripening phase.

Crop water Stress index and Precision irrigation water prescription maps:

Assessment of ET at different spatial and temporal scales is useful for irrigation water management at different growth stages of sugarcane. Mapping EvapoTranspiration at high Resolution using with Internalized Calibration (METRIC) (Allen et al 2007) model will be used for estimating ET. Surface properties derived from Landsat 8 images such as Albedo, surface emissivity, NDVI, Leaf Area Index and surface temperature will be used as inputs for the METRIC model to estimate ET. ET is estimated as a residual of surface energy equation.

LE= Rn– G – H

Where LE= Latent Energy consumed by ET, Rn= heat radiation and G is sensible heat flux conducted in to the ground and H is sensible heat flux convected to the air. Rn , H and G are calculated for each pixel of the satellite image. ET at the instant of the satellite image (ETinst) will be calculated for each pixel by following equation

3600 is a converting factor from second to hour, ρω is water density and λ is water latent heat of vaporisation. ETdaily is calculated using ETinst.weather data from nearby weather station near the Area of Interest. Reference ET (ETrdaily) is calculated using the weather data from nearby weather station. Using ET reference daily (ETrdaily) and ET instantanious  (ETinst), ET actual daily (ETdaily)is calculated.

MODIS data will be used for estimating crop water status by assessing the various Vegetation Indices such as Normalized Difference Water Index (NDWI) (Zarco-Tejada et.al., 2003) and Shortwave Infrared Water Stress Index (SIWSI) (Fensholt and Sandholt, 2003).

Using spectral signature analysis data from the timeline multispectral satellite imagery, estimated ET for a day, soil characteristics, etc., a decision support system will be developed using soft computing techniques such as supervised neural network, fuzzy systems and genetic algorithm.The developed decision support system will be utilized for generating prescription maps for precision irrigation water management.

Nitrogen nutrition status and precision nitrogen prescription maps:

Sensitive spectral wavelengths for quantifying nitrogen content existed mainly in the visible, red edge and far near-infrared regions of the electromagnetic spectrum. Several discriminative indices such as Chlorophyll Index (CI) (Bausch and Khosla , 2010), Modified Soil Adjusted Vegetation Index (MSAVI), (Bagheri, et.al., 2012), Optimized Soil Adjusted Vegetation Index(OSAVI), (Jia, et.al., 2011), and Triangular Greenness Index (TGI) (Hunt, et.al., 2013) and such other indices will be compared for finding the best index suitable for accessing nitrogen status in sugarcane crop at different growth stages in different varieties.

Nitrogen Nutrition Index (NNI) which is a ratio between the actual leaf nitrogen concentration (%Na) in the field at a specific growth stage and Critical Nitrogen (%Nc), will be worked out from the ground truth data. Regression models will be developed from the ground truth data based NNI and the various vegetation indices derived from the satellite data. Using crop and soil nitrogen status, a decision support system will be developed by soft computing techniques. Nitrogen prescription maps will be generated by using decision support system.

Farmers of the study area will be given access to the developed prescription maps for precision water and nitrogen management in sugarcane through the Russian VEGA-GEOGLAM – a Web-based GIS platform for crop monitoring and decision support in agriculture.

NATIONAL & INTERNATIONAL                      :



Remote sensing techniques provide timely, up-to-date and relatively accurate information for the management of sugarcane crop. Discrimination of sugarcane varieties using Landsat 7 ETM+ spectral data, in-situ hyper-spectral measurements and Hyperion data has shown promising results. Large scale water productivity (WP) and Actual evapo-transpiration (ET) of sugarcane in Brazil, was quantified using MODIS images, gridded weather dataand SAFER (Simple Algorithm for Evapo-transpiration Retrieving) algorithm. Sugarcane leaf N concentration was predicted from hyperspectral data from hyperion images, using both non-linear random forest (RF) regression and SML regression models derived from the first-order derivative of reflectance.

In India, the applications of remote sensing techniques in sugarcane have been undertaken with particular emphasis on crop acreage and production estimation. However, there are very few studies in India, on the use of remote sensing,for the discrimination of varieties and growth stages; and in-season (real time) crop health, water and nutritional status monitoring, for smart sugarcane farming.However, the advent of Multi-spectral and hyper-spectral sensors has provided new opportunities. Remotely sensed data should find use in sugarcane agriculture with satisfactory results, by selecting appropriate spatial and spectral resolution as well as suitable processing techniques for extracting sugarcane spectral information.




Russian VEGA-GEOGLAM – a Web-based GIS platform for crop monitoring and decision support in agriculture, would be of immense help in developing either a separate platform or adopting the same platform for Peninsular India, with suitable modifications to suit the local ecosystem requirements and farmers needs. Real time monitoring and providing advisory services for in-season crop management would help the farmers to increase production and profitability



 1st year2nd year
India to RussiaOne YearOne Year
Russia to IndiaOne YearOne Year

Launch and first meeting of collaborators (Indo-Russian Project on 18th March, 2019)

Training on Vega Geoglam web platform


No comments.