Article Preview
TopIntroduction
The world’s growing demand for food in the long term (Baudron & Giller, 2004) has raised the concern of our ability to meet this need without putting enormous pressure on the world’s natural resources and causing environmental damage. Climate change will also greatly impact food supply and demand and tougher environmental conditions, while anticipated resource limitations and increased production costs are putting constantly pressure on crop production systems. The challenge of the adoption of precision agriculture technologies seems to be a ‘One-way road’ to increase farming efficiency while minimizing environmental impacts (Awan, 2016; Foley et al., 2011).
For the last two decades, technological innovations have been tested to improve farming efficiency and reduce environmental impact (Daberkow & McBride 2003; Robertson et al. 2012; Tey & Brindal 2012). However, in the beginning, increased implementation costs had limited or uncertain benefits that lead more farmers to be unwilling to adopt available PA technologies on their farms (Castle et al., 2016). Recent studies (Liu et al., 2017; Nawar et al., 2017) on PA technologies indicated that the adoption of this technology can offer increased yields and productivity and also economic returns from reduced agricultural inputs limiting the excessive use of agro-chemicals in accordance with the latest environmental legislation. Individual studies (Calegari et al., 2013; Jayakumar et al., 2017; West and Kovacs, 2017) also focused and demonstrated the economic (monetary), agronomic (yield increase) and environmental benefits (reduction of negative impacts) of adopting PA technologies. These research findings on how data derived from soil characteristics, plant populations and environment can be organized to deliver targeted input applications to crop production systems encourage farmers to step into the new era of digital agriculture (Panagopoulos et al., 2014; West and Kovacs, 2017; Nawar et al., 2017).
In the traditional farm management model each field is treated as a homogeneous area (Srinivasan, 2006), where soil, topographic and environmental conditions are considered to be similar and the inputs are applied uniformly regardless any potential variability or heterogeneity. This approach leads to unwanted explicit economic costs due to inefficient application of inputs, causing also environmental damage due to the surplus of the unused nutrients (up to 30% of total N) that end up to ecosystems and the environment through leaching of water-soluble nitrates (Meisinger & Delgado, 2002), or runoff and gaseous emissions that increase the contamination risk (Follett & Delgado 2002; Hyytiainen et al., 2011; Rodriguez et al., 2011). In this case, the adoption of PA technologies can deliver a more efficient application of inputs under different conditions (Pierpaoli et al., 2013) or apply a single rate of a specific crop input to attain maximum efficiency (Vrindts et al., 2015) to sub-regions of broad similarity, defined as management zones, which regularly provide low or high yields (Fleming et al., 2004).