Classification and Prediction of Acid Sulfate Soils

Background and goals

Artificial Intelligence (AI) has become a subject of an extensive society-wide discussion, with its adepts and detractors. While some people consider that AI can be very dangerous, other affirm that it will save the world. Independently of these extremist positions, AI is an interdisciplinary and cross-sectional field, which is integrated in our every-day life. Currently, data is already gathered, analyzed and used for intelligent services in several industries such as retail, health care, or environmental technology. Due to its great potential to solve complex problems, the development and use of AI is going to be ever more important in the future. Nowadays, one of the major environmental problems of Finland is the Acid Sulfate soils. This type of soils generates serious ecological damages. To reduce this problem, it is essential to locate the areas where this type of soils appears. In this work we focus on the classification and prediction of acid sulfate soils as well as in the creation of digital soil maps by means of machine learning techniques. This project is a collaboration between Arcada University of Applied Sciences, Åbo Akademi University and Geological Survey of Finland (GTK).

Objectives and benefits

One of the main objectives of this project is the study of different machine learning techniques in the prediction and classification of AS soils in Finland. The goal is to establish a set of models capturing the main features of AS soils in order to classify actual soils and predict soils to be potentially problematic. This project is anticipated to produce several high-impact publications. The project results are i) reported in 3 articles in high-level scientific forums, ii) formation of a Nordic consortium to exchange knowledge, data and models, iii) submission of an application for external funding with Åbo Akademi and GTK.

Results

One of the main objectives of this project is the study of different machine learning techniques in the prediction and classification of AS soils in Finland. The goal is to establish a set of models capturing the main features of AS soils in order to classify actual soils and predict soils to be potentially problematic. This project is anticipated to produce several high-impact publications. The project results are i) reported in 3 articles in high-level scientific forums, ii) formation of a Nordic consortium to exchange knowledge, data and models, iii) submission of an application for external funding with Åbo Akademi and GTK.

Societal impact

It is of utmost importance to attract young very promising researchers to Arcada. Arcada has succeeded to do so in the past but need to continue this activity to remain on the global AI map. This new project (if granted) enables Virginia Estévez Nuño to establish herself as a prominent researcher in a new field, and Kaj-Mikael Björk to establish a new research consortium in a very interesting field. This consortium may put Arcada on the map, not only as a AI enabling research organization, but also as an AI solution organization for environmentally important areas, which are one of the cornerstones for a suistanable future. Thus, we will not only gain recognition in AI related fields, but also in fields that are vital for a sustainable future. To summarize the impact, the project is at the core of the Arcada strategy in three ways; we will create a nordic consortium, a new track of research in sustainability in Arcada, and further develop our AI capabilities.