Precision Agriculture

In this article, Prof. Wouter Saeys introduces the domain of Precision Agriculture.

Wouter is a professor in Biosystems Engineering at the University of Leuven in Belgium, where he leads a group working on precision technology for the agrofood sector.

Throughout history, the main aim of agriculture has always been to produce sufficient food to feed mankind. The world population is expected to grow to over 9 billion by 2050. So, agriculture is challenged to increase its productivity by more than 50% to produce sufficient food to feed all.

Only 70% of the land on earth is habitable and 50% of this land is already cultivated. So, one option could be to increase the cultivated area. However, this would come at the expense of a further loss in biodiversity. Agriculture is already held responsible for threatening 24 000 of the 28 000 species evaluated to be threatened with extinction. So, it would be better if we could further increase crop yields on the arable land which is already in use.

Over the past century, a tremendous increase in agricultural productivity has already been achieved. If we look at the evolution of the corn grain yield in the US, we can see that it remained fairly stable between 1866 and 1936. However, from the late 1930’s on, we see a fairly constant increase in the yield thanks to the introduction of hybrid breeds. This annual increase more than doubled from the mid 1950’s with the introduction of mechanization, mineral fertilizers and chemical pesticides. Thanks to this tremendous increase in crop yield, it is possible to produce the same amount of crop as in 1961 on 30% of the farmland. However, not all regions are at the same yield level yet. In India and China, the increasing trend has started later, while in Africa the growth rate is still quite limited. So, one might argue that to reach the targeted increase in food production by 2050 we should apply this success recipe worldwide.

Unfortunately, it will not be this simple, because the strong increase in crop yield realized over the past century, has created a lot of collateral damage in terms of a huge negative impact on the environment. Due to the excessive use of fertilizers and pesticides, ground and surface waters are polluted, resulting in a negative impact on human health and biodiversity. Moreover, agriculture already uses 70% of the globally available fresh water and is held responsible for more than 20% of the greenhouse gas emissions. So, we will have to apply these resources more efficiently to minimize the negative impact on the environment and reach our productivity goals in a sustainable way.

One of the main reasons for this inefficiency can be found in the complex interplay of many factors influencing crop growth, which results in large uncertainty. Plants require sunlight, water and a sufficiently high temperature to grow. So, crop growth is largely determined by the weather conditions. These can vary from one season to another and many choices with respect to tillage, planting and fertilization have to be made without knowing which weather we will get. Plants have to take up water and nutrients from the soil in which they are rooted. The weather conditions and soil properties determine when these nutrients become available to the plant and to what extent these can be extracted from it. For example, if organic fertilizer is applied too late in the season, a large part of the nutrients may only become available when the crop has already been harvested and leach out to the environment during the wet winter period. So, optimal timing of the actions and adaptation to the actual crop dynamics are crucial to maximize the resource efficiency.

Another main source of losses is the scale at which operations are performed. Thanks to the advancements in mechanization a farmer can till large areas and produce large amounts of food. However, he or she can no longer judge the needs of individual plants like a smallholder farmer could do. So, all plants are treated in the same way and the same dose of fertilizer or pesticide is applied to the entire field including a large safety margin. As labor is the most expensive production factor in developed countries, this allows the farmer to obtain a high yield at the lowest cost. However, it does not result in optimal use of the resources, leading to a large impact on the environment. For example, the entire field would be sprayed with herbicide while less than 5% of it is covered with weed. Field trials have shown that by only spraying herbicide where weeds are present, the herbicide use can in some cases may be reduced by 90%. So, to reach a higher resource efficiency the variation within one field should be taken into account to define the right actions at the right place.

To be able to decide on the right actions, precision agriculture makes use of advanced technologies for sensing, data analysis and automation. Geo-referenced data is acquired during tillage, crop growth and harvest to determine the soil and crop characteristics for every place in the field. This information is then combined with crop growth models to determine the local yield potential and to decide on the most suitable action for every location in the field. This is then translated into task maps which can be loaded into the machinery to adapt the tillage, fertilization or crop protection to the local needs. In this way, the yield can be maximized while minimizing the impact on the environment. This is illustrated in figure 1.

Figure 1: Precision Agriculture. (Click to expand)

Why not have a look at Prof. Wouter Saeys complementary presentation in the See Also section below