Abstract
The combination of high spatial resolution and multi-date satellite imagery offers new opportunities for mapping and monitoring crop types of different agricultural field sizes. However, mapping of crop types at high spatial resolution requires high-quality crop type reference data typically collected from the ground-based surveys to create the maps and/or to assess the map accuracy. The availability of sufficient crop type reference data is limited over large geographic regions because of the time, effort, cost, and accessibility in different parts of the world. To generate large area crop type maps, any existing, but limited reference data must be spatially extended to other regions using appropriate and available non-ground-based sources. There is the potential to classify High Resolution Imagery (HRI) using a phenology-based approach to generate additional reference data within similar agriculture ecological zones (AEZs) based on the crop characteristics, their types, and their growing season. Therefore, the objective of this study was to evaluate if existing, limited crop type reference data could be extended using this approach. Multi-date, high spatial resolution satellite images were used to spatially extend the limited crop type reference data from one region (called the training region (TR)) to another region (called the test region (TE)) within the same AEZ using a phenology-based Decision Tree (DT) classifier for three different field sizes. The results demonstrate that this phenology-based classification approach can efficiently and effectively extend the limited crop type reference data to other regions in same AEZ for different field sizes.
Highlights
Food security is one of the major challenges that human beings are facing (Zhong et al, 2014)
The crop/no-crop maps of the six regions were subsequently classified into the crop types using the phenology-based classification algorithm and training data collected from the 2015 Cropland Data Layer (CDL) of the TR regions
The crop type maps of the TE regions were developed from the multi-dates of satellite imagery and training data derived from the TR regions using a Decision Trees (DT) approach since the goal was to extend the classification into similar regions without collecting training data in those regions
Summary
Food security is one of the major challenges that human beings are facing (Zhong et al, 2014). By 2050, the global population of 9.8 billion will demand 70% more food than is consumed today (Schwab et al, 2014). To meet this demand, cropland areas are increasing using current agriculture practices causing greenhouse gas (GHG) emissions and environmental degradation (Adams and Eswaran, 2000; Beach et al, 2008). The required knowledge to improve these current agriculture practices and model GHG variability in different agriculture systems demands the identification of different crop types (Ramankutty et al, 2008; Peña-Barragán et al, 2011; Gong et al, 2013). Acquiring crop type information over large geographic regions is extremely relevant for decision making and policy actions (Yang et al, 2011; Foerster et al, 2012).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
