Sustainable agriculture is currently being challenged under climate change scenarios since extreme environmental processes disrupt and diminish global food production. For example, drought-induced increases in plant diseases and rainfall caused a decrease in food production. Machine Learning and Agricultural Big Data are high-performance computing technologies that allow analyzing a large amount of data to understand agricultural production. Machine Learning and Agricultural Big Data are high-performance computing technologies that allow the processing and analysis of large amounts of heterogeneous data for which intelligent IT and high-resolution remote sensing techniques are required. However, the selection of ML algorithms depends on the types of data to be used. Therefore, agricultural scientists need to understand the data and the sources from which they are derived. These data can be structured, such as temperature and humidity data, which are usually numerical (e.g., float); semi-structured, such as those from spreadsheets and information repositories, since these data types are not previously defined and are stored in No-SQL databases; and unstructured, such as those from files such as PDF, TIFF, and satellite images, since they have not been processed and therefore are not stored in any database but in repositories (e.g., Hadoop). This study provides insight into the data types used in Agricultural Big Data along with their main challenges and trends. It analyzes 43 papers selected through the protocol proposed by Kitchenham and Charters and validated with the PRISMA criteria. It was found that the primary data sources are Databases, Sensors, Cameras, GPS, and Remote Sensing, which capture data stored in Platforms such as Hadoop, Cloud Computing, and Google Earth Engine. In the future, Data Lakes will allow for data integration across different platforms, as they provide representation models of other data types and the relationships between them, improving the quality of the data to be integrated.

Climate change is currently one of agriculture’s main problems in achieving sustainability. It causes drought, increased rainfall, and increased diseases, causing a decrease in food production. In order to combat these problems, Agricultural Big Data contributes with tools that improve the understanding of complex, multivariate, and unpredictable agricultural ecosystems through the collection, storage, processing, and analysis of vast amounts of data from diverse heterogeneous sources. This research aims to discuss the advancement of technologies used in Agricultural Big Data architectures in the context of climate change. The study aims to highlight the tools used to process, analyze, and visualize the data, to discuss the use of the architectures in crop, water, climate, and soil management, and especially to analyze the context, whether it is in Resilience Mitigation or Adaptation. The PRISMA protocol guided the study, finding 33 relevant papers. However, despite advances in this line of research, few papers were found that mention architecture components, in addition to a lack of standards and the use of reference architectures that allow the proper development of Agricultural Big Data in the context of climate change.

This article proposes a monitoring system that allows to track transitions between different stages in the berry harvesting process (berry picking, waiting for transport, transport and arrival at the packing site) solely using information from temperature and vibration sensors located in the basket. The monitoring system assumes a characterization of the process based on hidden Markov models and uses the Viterbi algorithm to perform inferences and estimate the most likely state trajectory. The obtained state trajectory estimate is then used to compute a potential damage indicator in real time. The proposed methodology does not require information about the weight of the basket to identify each of the different stages, which makes it effective and more efficient than other alternatives available in the industry.

In some important berry-producing countries, such as Chile, the fruit is harvested manually. The markets for these products are generally very distant, and any damage caused to the fruit during harvesting will be expressed in its shelf life. The first step to understanding the harvesting process is to identify what happens to the harvest baskets in each stage (picking, wait-full, transport-full, freezing tunnel, emptying and transport-empty), allowing variables that can affect the shelf life to be identified. This article proposes the use of Smartbins, intelligent harvest baskets with sensors to collect weight, temperature, and vibration data. Combined analysis of the variables collected, using machine learning algorithms, allows the system to estimate which stage the basket is at with an accuracy of 80%, and to assess whether the fruit has been exposed to situations that could affect its shelf life. Due to imbalance characteristics of the data collected, the best results were obtained in longer stages (picking and wait-full stages with 89% and 86% respectively).

Climate change is currently one of the main problems facing agriculture to achieve sustainability. It causes situations such as drought, increased rainfall, and increased diseases, causing a decrease in food production. In order to combat these problems, Agricultural Big Data contributes with tools that allow improving the understanding of complex, multivariate, and unpredictable agricultural ecosystems through the collection, storage, processing, and analysis of vast amounts of data from diverse heterogeneous sources. This research aims to discuss the advancement of technologies used in Agricultural Big Data architectures in the context of climate change. The study aims to highlight the tools used to process, analyze, and visualize the data and discuss the use of the architectures in the crop, water, climate, and soil management, especially to analyze the context, whether it is in Resilience Mitigation or Adaptation. The PRISMA protocol guided the study, finding 33 relevant papers. Despite the advances in this line of research, few papers were found that mention the components of the architectures, in addition to the lack of standards and the use of reference architectures, which allow the proper development of Agricultural Big Data in the context of climate change.