Naslov
Competence, perspective and potential role of agronomists in agriculture digitalization in Croatia

Autori
Tačković, Dominik ; Lončarić, Ružica ; Jelić Milković, Sanja ; Lončarić, Zdenko

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
1st International Symposium on Digital Technologies in Agriculture / Lončarić, Zdenko ; Jović ; Jurica - Osijek : Fakultet agrobiotehničkih znanosti Sveučilišta Josipa Jurja Strossmayera u Osijeku, 2022, 46-46

ISBN
978-953-8421-03-7

Skup
1st International Symposium on Digital Technologies in Agriculture

Mjesto i datum
Osijek, Hrvatska, 06.12.2022. - 08.12.2022

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
availability modeling, regression model, EDTA, Fe

Sažetak
Intensive agricultural production, without proper management, will probably cause soil degradation. Therefore, monitoring soil quality in agricultural regions is essential for soil protection. Analytical results of soil properties can be successfully combined with mathematical and/or computer models to predict values of soil indicators that have not been analytically determined. In doing so, it is very important to determine as accurately as possible which properties determine the intensity of a certain soil indicator to the greatest extent. Iron is an essential element whose availability in the soil depends significantly on the total concentration of Fe in the soil, soil pH value and SOM content, and the lack of plant available Fe can limit production on poor, especially carbonate soils. The aim of this research was to determine the possibility of using available data to predict Fe availability with regression models. The regression model was created using a data set with the results of analysis of Fe concentration in more than two hundred soil samples (total Fe extracted with aqua regia in the range 2-3.9% and plant available Fe in the range 10-45 mg kg-1). The model was used to predict the available Fe based on the analysed soil properties (pHH2O, pHKCl and SOM content) of 1, 000 soil samples with the total analysed production area 5, 194 ha. Modeling available Fe on a new set with 1, 000 soil samples predicted that the low concentration of available Fe will be only on 1.83 % of analysed areas, which is just 95 ha. The moderate range of plant available Fe was predicted at 245 ha (4, 72 % area). High level of Fe availability was predicted at 93, 45 % area, i.e. on 4.854 ha. The model predicts a very small proportion of samples with a low Fe available to the plant, which leads to the assumption of low accuracy of the model, even though SOM and two pH values were used. If that’s the case, a data set with basic soil indicators is not sufficient to predict Fe availability. The demonstrated use of the model indicates that modeling can be effectively used to predict certain soil indicators without actual laboratory analysis, saving time and resources. This can be especially significant in cases of a large number of soil samples from different and heterogeneous production areas. However, the accuracy of the model must be validated, which is possible only by validation with a reasonably selected new set of samples and analytical results. Thereby, it is more likely that analyzes of additional soil properties (like percentage of clay particles, total amount and different fractions of Fe) will be required in addition to the basic soil analyses (pHH2O, pHKCl and SOM content).

Izvorni jezik
Engleski

Znanstvena područja
Poljoprivreda (agronomija), Biotehnologija

POVEZANOST RADA