Future Farm project

The Future Farm Project was established to re-examine and improve the way in which soil and crop sensors, supplemented by other sources of useful/available data, are used to inform decisions about input management and to provide a way of automating the process from data acquisition, through analysis, to the formulation and implementation of decision options. In particular, using nitrogen (N) fertilizer management as a ‘use-case’, the project sought to enable enhanced grower confidence in N decision making through the adaptive generation of site-specific management models. A key element of these is their increased and improved use of in-season field monitored data (soil, crop, climatic), historic on-farm data, external public and private data and automation of decision rules in software that may potentially be linked to real-time application equipment. This was considered important given the pre-project perception that a lack of farmer confidence in precision agriculture-based decision making was constraining adoption of precision agriculture (PA) approaches to management of grains-based farming systems. This lack of adoption was in spite of the potential of PA approaches as a counter to farm labour shortages, the need to optimise resource use efficiency as a means of maintaining or enhancing farm profitability and the finding through an exhaustive modelling exercise, that the error associated with prediction of N fertilizer requirement based on expected yield was of the order of 50 kg N/ha. Future Farm was co-funded by GRDC and involved CSIRO (as lead research organisation) along with the University of Sydney, University of Southern Queensland, Queensland University of Technology and Agriculture Victoria. The project made us of both 'core' and 'satellite' field sites across the major grain-growing regions; core sites were the major project resource, whereas satellite sites were those where we collaborated opportunistically with farmers running their own strip trials. The dataset comprises data collected in-field at these various sites (using soil and crop analysis or through the use of proximal crop or soil sensors) or acquired through remote sensing or from publicly available sources (eg weather data, soil information systems); historical data were also acquired. It also includes data gathered through the use of yield monitors and protein sensors on the farmers' harvesters. For the latter reason along with other privacy issues, access to the dataset is restricted. Further information about Future Farm is available at https://grdc.com.au/resources-and-publications/grdc-update-papers/tab-content/grdc-update-papers/2022/02/better-targeted,-more-precise-fertiliser-decisions-as-a-counter-to-rising-fertiliser-prices-focussing-on-3-of-the-6-rs and on other relevant GRDC webpages. Code is available at: https://bitbucket.csiro.au/projects/FUTUREFARM

未来农场项目（Future Farm Project）的设立旨在重新审视并优化土壤与作物传感器的应用方式——这些传感器辅以其他有用/可用数据源——以指导投入管理决策，并提供一套从数据采集、分析到决策方案制定与实施的自动化流程。具体而言，该项目以氮肥（nitrogen, N）管理为‘用例’，通过自适应生成特定地块管理模型，旨在提升种植者对氮肥决策的信心。这些模型的核心要素包括：增加并优化季内田间监测数据（土壤、作物、气候）、农场历史数据、外部公共与私有数据的使用，以及将决策规则在软件中自动化——该软件或可与实时应用设备相连。考虑到项目启动前的认知——种植者对基于精准农业（precision agriculture）的决策缺乏信心，正制约着谷物种植系统管理中精准农业（PA）方法的采用——这一点尤为重要。尽管精准农业方法具有应对农场劳动力短缺的潜力、优化资源利用效率以维持或提升农场盈利能力的必要性，且通过详尽建模发现，基于预期产量预测氮肥需求的误差约为50千克氮/公顷，但该方法的采用率仍较低。未来农场项目由GRDC联合资助，参与方包括CSIRO（牵头研究机构）、悉尼大学、南昆士兰大学、昆士兰科技大学及维多利亚农业局。该项目在主要谷物种植区设置了‘核心’与‘卫星’两类试验田；核心试验田是项目的主要资源，而卫星试验田则是与开展自主条带试验的种植者进行机会性合作的场所。本数据集包含以下数据：在各试验田现场采集的数据（通过土壤与作物分析，或近地作物/土壤传感器）、通过遥感获取的数据、来自公开数据源的数据（如气象数据、土壤信息系统），以及历史数据。此外，数据集还包含通过种植者收割机上的产量监测仪和蛋白质传感器收集的数据。由于上述原因及其他隐私问题，本数据集的访问权限受限。关于未来农场项目的更多信息，请访问：https://grdc.com.au/resources-and-publications/grdc-update-papers/tab-content/grdc-update-papers/2022/02/better-targeted,-more-precise-fertiliser-decisions-as-a-counter-to-rising-fertiliser-prices-focussing-on-3-of-the-6-rs 及其他相关GRDC网页。代码可访问：https://bitbucket.csiro.au/projects/FUTUREFARM