August 27th - Advances in Crop Sensing

Precision agriculture depends on information and that is provided largely by remote and proximal sensors. This webinar will include presentations on remote sensing for crop monitoring, yield prediction and irrigation management, as well as proximal soil sensors for better soil management.
Presented on August 27th, 2020 at 13:30 CDT (UTC -5).

Carrot Yield Forecasting Using Different Multispectral Satellite Sensors

Corresponding Author and Presenter:

Angelica Suarez Cadavid

University of New England
Research Fellow, Applied Agricultural Remote Sensing Center
This study explored the accuracies of Sentinel-2 (S2) and Worldview-3 (WV3) satellite sensors to forecast carrot root yield at the field level in three principal vegetable regions in Australia located in Western Australia, Queensland and Tasmania. The growing regions featured different cropping configurations, seasons and soil conditions. Above ground biomass (AGB) and yield measures were collected between 2017 and 2019 from more than 400 sample locations in 24 fields across the regions. Several reflectance-based vegetation indices were calculated and calibrated with sample data to estimate carrot root yield (t/ha). Validated accuracies ranged from 82% to 100% at the field level confirming the significant applicability of this technology in vegetable crops where dynamics of the systems limit the implementation of other technologies such as yield monitors.

EO4Agri Analysis of Potential Earth Observation for Precision Agriculture

Corresponding Author and Presenter:

Karel Charvel

Project Manager
The main objective of EO4AGRI is to catalyze the evolution of the European capacity for improving operational agriculture monitoring from local to global levels based on information derived from Copernicus satellite observation data and through exploitation of associated geospatial and socio-economic information services. The main findings from this period are:
  • Climatic changes will be one from key drivers for future agriculture. Therefore it is necessary to provide more focused research in the area of analysis of dependencies of climate data with EO data, including historical analysis
  • Users require data with higher resolution
  • Willingness of farming sector to pay for new services is not very high and this could be a limitation for introducing new services
  • There is a need for deep integration of Earth Observation research with agriculture research. The same data could bring different results on the basis of different strategies
  • Green deal can be a good start for new Precision Agriculture Services, due to the fact that Green Deal addresses the environmental issues broadly, including Farm to Fork strategy, Biodiversity strategy, CAP strategy, Climate Change Strategy
  • European Data Strategy and Digital Twins ideas can open new possibilities for Public-Private-Partnership (PPP) for Agriculture Data. This should be also necessary to be reflected in the CAP reform
  • New intensive research in ICT like Artificial Intelligence, Big Data, Data fusion, HPC will be needed
  • In Public sector there are requirements on new services, not only on monitoring. Planning, water management and adaptation to climatic changes are also an important topic for the future.
  • Integration of EO and navigation data is necessary
  • Standardisation is key issue for building effective solutions for integration of Agriculture Data

Sugarcane Crop Classification and Varietal Discrimination from Sentinel-2 Using Supervised Classifier

Corresponding Author:

Virupakshagouda Patil

The presentation is about the identification and mapping of sugarcane crop and its varieties as it is important for farmers and sugar industry decision makers of the region, for scheduling harvest as well as for agricultural field Management. The objectives of this study were to create spectral library of crops for crop discrimination.  In this study, the reference spectra is prepared from the collected ground data for different crops i.e. Sugarcane, Corn, Wheat, Turmeric and different sugarcane varieties (Co86032, Co91010, Co M265, SNK9293, VSI08005 and  PI1110). Classification was performed by using Spectral Angle Mapper (SAM) which clearly ascertains its efficacy and reliability in retrieving the crop and its variety occurring in an area in an accurate yet quick manner. 

Evaluating Proximal Soil Sensors for Measurement of Soil Physical and Chemical Properties

Corresponding Author and Presenter:

Chase Maxton

Veris Technologies
Electronic Engineer
Multiple proximal soil sensor technologies (EC, Soil optical reflectance, sensor pH, Gamma) were evaluated on 17 fields across 8 states for their correlation to traditional lab analyses.

Intelligent Irrigation by Fusion of Spectral Vegetation Indices and SAR Imagery

Corresponding Author and Presenter:

Ofer Beeri

Manna Irrigation
Chief Scientist
Integration of different satellite sensors to allow irrigation decision making at small plots, regardless weather condition

Precision Agriculture Approach for Understanding Spatial Variability of Sugarcane Productivity: A Case Study from Karnataka

Corresponding Author and Presenter:

Virupakshagouda Patil

The presentation is about assessing of soil macronutrients (i.e., nitrogen, phosphorus, and potassium) for site-specific crop management (SSCM) and enhancing agricultural production. Geospatial technology has been utilized to assess the impact of soil properties on sugarcane productivity and optimize the decision making management for site-specific fertilization and precision farming. Kriging technique is used on 643 geo-referenced point soil samples during 2014-15 to interpolate lab analyzed macro nutrient such as pH, Electrical conductivity (EC), and soil available nutrients (N, P2O5, K2O). Spatial variability was quantified through Semi-variogram analysis using Geostatistics. Kriged maps were generated in Geographic Information System (GIS) using best fit exponential model. Sentinel-2 imageries are utilized for correlation between soil physical properties and the vegetation indices such as NDVI, NDWI, GNDVI, NDRE and EVI to develop prescription maps for variable rate application of fertilizers and soil amendments.

Estimating Daily 3 M Leaf Area Index from Space Using CubeSats Images

Corresponding Author and Presenter:

Yuval Sadeh

School of Earth, Atmosphere and Environment, Monash University, Melbourne, Australia
PhD Candidate
Remote-sensing applications are limited by the trade-off between spatial and temporal resolutions. Monitoring the dynamics of Leaf Area Index (LAI) from space is a key attribute to estimate crop types and their phenology over large areas, and in characterising spatial variations within growers’ fields. This presentation proposes a new method to fuse a time-series of Sentinel-2 and CubeSat imagery into daily RGB-NIR surface reflectance and subsequently LAI datasets at 3 m resolution. The results of this study, which focused on monitoring wheat LAI, show that this method is effective for high spatio-temporal monitoring of field-crops.

Presented on

August 27th, 2020
13:30 CDT (UTC -5).

Moderated by:

James Taylor

James Taylor

Watch the Webinar from August 27th

Advances in Crop Sensing

Were you unable to attend the Virtual ICPA 2020 Webinar Series? Do you want to watch the webinar again to view the keynote presentation and award presentation? Use the video controls on the preview image to play the video or if you prefer, you can download the file using the link below.
Download the File (395mb)