Evolution and application of digital technologies to predict crop type and crop phenology in agriculture

Abstract

The downside risk of crop production affects the entire supply chain of the agricultural industry nationally and globally. This also has a profound impact on food security, and thus livelihoods, in many parts of the world. The advent of high temporal, spatial and spectral resolution remote sensing platforms, specifically during the last five years, and the advancement in software pipelines and cloud computing have resulted in the collating, analysing and application of “BIG DATA” systems, especially in agriculture. Furthermore, the application of traditional and novel computational and machine learning approaches is assisting in resolving complex interactions, to reveal components of eco-physiological systems that were previously deemed either “too difficult” to solve or “unseen”. In this review, digital technologies encompass mathematical, computational, proximal- and remote sensing technologies. Here, we review the current state of digital technologies and their application in broad acre cropping systems globally and in Australia. More specifically, we discuss the advances in (i) remote sensing platforms, (ii) machine learning approaches to discriminate between crops, and (iii) the prediction of crop phenological stages from both sensing and crop simulation systems for major Australian winter crops. An integrated solution is proposed to allow accurate development, validation and scalability of predictive tools for crop phenology mapping at within-field scales, across extensive cropping areas.

Can We Harness “Enviromics” to Accelerate Crop Improvement by Integrating Breeding and Agronomy?

“…Complexity, cost, and the timeliness of detailed measurements of the water status of environments and genotypic variation for plant responses to water deficits under field conditions have limited adoption and application of many discoveries and methods to the scale of breeding programs. Recently, new proximal and remote sensor technologies and data modelling capabilities have become available to enhance characterization of environments and measure plant responses under field conditions at higher throughput and at greater scales to enhance applications for crop improvement and yield prediction (Pauli et al, 2016;Guan et al, 2017;Araus et al, 2018;Messina et al, 2018;Van Eeuwijk et al, 2019;Cooper et al, 2020;Messina et al, 2020;Peng et al, 2020;Schwalbert et al, 2020;Costa-Neto et al, 2021;Jain et al, 2021;Jin et al, 2021;Potgieter et al, 2021;Smith et al, 2021;Yang et al, 2021).…”

“…Further, for most crop breeding programs the relationships between the environments sampled in METs and the dominant environmental conditions of the TPE are neither well understood nor adequately quantified (Cooper and DeLacy, 1994;Cooper et al, 2021). Improved sensor technologies and prediction methodologies are urgently required to characterize and study environments within breeding and agronomy METs and to quantify the relationships between the environments sampled in METs for all stages of crop improvement programs and their importance for the TPE (Messina et al, 2020;Crespo-Herrera et al, 2021;Kusmec et al, 2021;Potgieter et al, 2021;Smith et al, 2021).…”

“…Farmers seek improved technology combinations based on genotypes and agronomic management that can consistently deliver yield productivity close to the potential of their on-farm environments, while managing the risk of crop failure (Hammer et al, 2014;Hunt et al, 2021). As a constructive step toward improving the predictability of on-farm crop productivity, there has been continual refinement of the definition of environments in METs and the agricultural TPE to recognize the important role of crop management and for investigation of the influences of genotype-by-environmentby-management (GxExM) interactions (Kirkegaard and Hunt, 2010;Hammer et al, 2014Hammer et al, , 2020Hatfield and Walthall, 2015;Beres et al, 2020;Peng et al, 2020;Cooper et al, 2021;Hunt et al, 2021;Potgieter et al, 2021;Smith et al, 2021). Thus, we can study genetic improvements from the perspective of the breeder, crop management improvements from the perspective of the agronomist, and improvement in genotypemanagement technology combinations from the perspective of the farmer.…”

Unsupervised Plot-Scale LAI Phenotyping via UAV-Based Imaging, Modelling, and Machine Learning

“…However, currently most satellite data have insufficient temporal and spatial resolution to be utilized in field plot trials. Precise positioning and unmanned aerial vehicle (UAV) technologies facilitate high-throughput phenotyping in agriculture and breeding programs [18][19][20]. Compared with satellite data, the use of UAV platforms in breeding programs has several advantages.…”

Bayesian multi-level calibration of a process-based maize phenology model

“…It is controlled by environmental variables and determines the timing of plant organ development and the distribution of assimilates to different parts of the plant. Thus, predictions of phenological development are essential for evaluating crop growth and yield, and may also support field management decisions such as the timing of fertilizer application [6]. These phenology predictions are made possible by using numerical models.…”