In this study, we propose integrating unmanned aerial systems (UASs) and machine learning classification for suitability prediction of expanding habitats for endangered flora species to prevent ...further extinction. Remote sensing imaging of the protected steppe-like grassland in Bilje using the DJI P4 Multispectral UAS ensured non-invasive data collection. A total of 129 individual flora units of five endangered flora species, including small pasque flower (Pulsatilla pratensis (L.) Miller ssp. nigricans (Störck) Zämelis), green-winged orchid (Orchis morio (L.)), Hungarian false leopardbane (Doronicum hungaricum Rchb.f.), bloody cranesbill (Geranium sanguineum (L.)) and Hungarian iris (Iris variegate (L.)) were detected and georeferenced. Habitat suitability in the projected area, designated for the expansion of the current area of steppe-like grassland in Bilje, was predicted using the binomial machine learning classification algorithm based on three groups of environmental abiotic criteria: vegetation, soil, and topography. Four machine learning classification methods were evaluated: random forest, XGBoost, neural network, and generalized linear model. The random forest method outperformed the other classification methods for all five flora species and achieved the highest receiver operating characteristic (ROC) values, ranging from 0.809 to 0.999. Soil compaction was the least favorable criterion for the habitat suitability of all five flora species, indicating the need to perform soil tillage operations to potentially enable the expansion of their coverage in the projected area. However, potential habitat suitability was detected for the critically endangered flora species of Hungarian false leopardbane, indicating its habitat-related potential for expanding and preventing further extinction. In addition to the current methods of predicting current coverage and population count of endangered species using UASs, the proposed method could serve as a basis for decision making in nature conservation and land management.
Leaf Soil-Plant Analysis Development (SPAD) prediction is a crucial measure of plant health and is essential for optimizing indoor plant management. The deep learning methods offer advanced tools for ...precise evaluations but their adaptation to the heterogeneous indoor plant ecosystem presents distinct challenges. This study assesses how accurately deep neural network (DNN) predicts SPAD values in leaves on indoor plants when compared to well-established machine learning techniques, including Random Forest (RF) and Extreme Gradient Boosting (XGB). The covariates for prediction were based on low-cost multispectral and soil electro-conductivity (EC) sensors, enabling a non-destructive sensing approach. The study also strongly emphasized multicollinearity analysis quantified by the Variance Inflation Factor (VIF) and two independent indices, as well as its effect on prediction accuracy using deep and machine learning methods. DNN resulted in higher accuracy to RF and XGB, also performing better using filtered data after multicollinearity analysis based on the coefficient of determination (R2), root mean square error (RMSE) and mean absolute error (MAE) (R2 = 0.589, RMSE = 11.68, MAE = 9.52) in comparison to using all input covariates (R2 = 0.476, RMSE = 12.90, MAE = 10.94). Overall, DNN was proven as a more accurate prediction method than the conventional machine learning approach for the prediction of leaf SPAD values in indoor plants, despite using heterogenous plant types and input covariates.
While SoilGrids is an important source of soil property data for a wide range of environmental studies worldwide, there is currently an extreme lack of studies evaluating its accuracy against ...independent ground truth soil sampling data. This study aimed to provide a comprehensive insight into the accuracy of SoilGrids layers for three physical soil properties representing soil texture components (clay, silt, and sand soil contents) using ground truth data in the heterogeneous landscape of Croatia. These ground truth data consisted of 686 soil samples collected within the national project at a 0–30 cm soil depth, representing the most recent official national data available. The main specificity of this study was that SoilGrids was created based on zero soil samples in the study area, according to the ISRIC WoSIS Soil Profile Database, which is very sparse for the wider surroundings of the study area. The accuracy assessment metrics indicated an overall low accuracy of the SoilGrids data compared with the ground truth data in Croatia, with the average coefficient of determination (R2) ranging from 0.039 for silt and sand to 0.267 for clay, while the normalized root-mean-square error (NRMSE) ranged from 0.362 to 2.553. Despite the great value of SoilGrids in a vast range of environmental studies, this study proved that the accuracy of its products is highly dependent on the presence of ground truth data in the study area.
With the emergence of machine learning methods during the past decade, alternatives to conventional geostatistical methods for soil mapping are becoming increasingly more sophisticated. To provide a ...complete overview of their performance, this study performed cost–benefit analysis of four soil mapping methods based on five criteria: accuracy, processing time, robustness, scalability and applicability. The evaluated methods were ordinary kriging (OK), regression kriging (RK), random forest (RF) and ensemble machine learning (EML) for the prediction of total soil carbon and nitrogen. The results of these mechanisms were objectively standardized using the linear scaling method, and their relative importance was quantified using the analytic hierarchy process (AHP). EML resulted in the highest cost–benefit score of the tested methods, with maximum values of accuracy, robustness and scalability, achieving a 55.6% higher score than the second-ranked RF method. The two geostatistical methods ranked last in the cost–benefit analysis. Despite that, OK could retain its place as the most frequent method for soil mapping in recent studies due to its widespread, user-friendly implementation in GIS software and its univariate character. Further improvement of machine learning methods with regards to computational efficiency could additionally improve their cost–benefit advantage and establish them as the universal standard for soil mapping.
This research was based on the exploitational analysis of the navigational devices by a family farm from eastern part of Croatia - location Bilje. The research developed into the assessment of the ...total overlapping surface which ocurred during with and without navigation. The difference between estimated surfaces of overlaps formulates total reduction of the surface of an overlap. Calculations confirm that this farm managed to decrease the surface of the overlaps from 364 ha to much smaller number of 152 ha just by using navigation, which totals to a difference of 211,74 ha (58,24 %). The farm was able to process an area of 5 364,06 ha in all operations, where 364,06 ha was lost on overlapping without the use of navigation. Larger manufacturers have better possibility and the need to install new technologies to a greater extent, where savings are in much bigger numbers. In order to make proper use of a navigational system, it is necessary to invest additional resources so that the system can be equipped with an automatic section control. The automatic section control demands additional investments in devices that control the work of the section.
The paper presents the result of a triennial field experiment (2013‒15), aiming to determine the influence of irrigation, nitrogen fertilization, and cultivars, as well as their interactions on the ...yield and chemical properties of the soybean seeds. Four soybean cultivars (Lucija, Vita, Ika and Tena) of different maturity groups were investigate as a sub‐subplot factor (C). The main plot factor (A - irrigation) resulted in a statistically very significant (P≤0.01) seed yield in all three years, and it was found out by an analysis of variance. The subplot factor (B - nitrogen fertilization) had an impact on the grain yield depending on the research year, while sub‐subplot factor (C-cultivar) significantly affected all examined traits. The factor interactions and their significance varied by the research years. The seed yield achieved in 2013 (3883 kg ha-1) indicated a great importance of all factors’ interaction. The correlations between a seed yield and a protein and oil concentration were determined during the research.
Duckweed is a widespread type of tiny free-floating plants of the flowering class. A typical representative of the family of the cowhide (Lemnaceae) is a large duckweed (Spirodela polyrhiza) and it ...is very common on Croatian inland waters. Like all other species of duckweeds, it is characterized by the possibility of vegetative and sexual reproduction and very rapid growth. It has the ability to remove nitrogen, phosphorus and some heavy metals from the substrate and is considered a very desirable raw material for biogas production for several reasons. The necessity of reducing environmental pollution of nitrates from agricultural production and the ability to grow water lenses under eutrophic conditions have sparked this research. The primary objective of the study is to determine the possibilities of continuously growing large duckweed at different concentrations of digestates for the purpose of biogas production. The secondary goal is to determine the dependence between the different digestate concentrations used for the duckweed green mass production and the quantity and quality of the biogas obtained through the anaerobic digestion process at thermophilic conditions.
Vodena leća je široko rasprostranjena vrsta sitnih slobodno plivajućih biljaka iz razreda cvjetnica. Tipičan predstavnik porodice kozlačevki (Lemnaceae) je velika vodena leća (Spirodela polyrhiza) i vrlo je česta na vodenim površinama kontinentalne Hrvatske. Kao i sve ostale vrste vodenih leća, karakterizira ju mogućnost vegetativnoga i spolnog razmnožavanja te vrlo brz rast. Smatra se vrlo poželjnom sirovinom za proizvodnju bioplina iz nekoliko razloga. Nužnost smanjenja zagađenja okoliša nitratima iz poljoprivredne proizvodnje te sposobnost rasta vodenih leća u eutrofnim uvjetima potaknuli su ovo istraživanje. Primarni cilj istraživanja jest utvrditi mogućnosti kontinuiranoga uzgoja velike vodene leće na različitim koncentracijama digestata, sa ciljem proizvodnje bioplina. Sekundarni je cilj utvrditi zavisnost između različitih koncentracija digestata korištenih za proizvodnju zelene mase vodene leće te kvantitete i kvalitete bioplina dobivenog postupkom anaerobne digestije pri termofilnim uvjetima.
U trogodišnjim poljskim ispitivanjima utvrđen je utjecaj godine, tretmana navodnjavanja i sorte na visinu biljaka i urod zrna soje u uvjetima istočne Hrvatske. Statistički vrlo značajan utjecaj ...(P=0,01) na visinu biljaka soje pokazuju sva tri ispitivana faktora, kao i njihove interakcije, izuzev interakcije navodnjavanja i sorte koja je bila značajna na razini P=0,05. Na urod zrna soje ispitivani faktori, kao i sve njihove interakcije, utječu na razini značajnosti P=0,01. Dobivene vrijednosti uroda zrna soje tijekom istraživanja (uglavnom veće od 3000 kg ha-1), ukazuju na važnost odabira sorte i tretmana navodnjavanja u prilagodbi proizvodnje soje nepovoljnim vremenskim utjecajima godine.
Biological nitrogen fixation represents the subject of numerous investigations of the scientific community. The advantages of this process, as already well known while scientific research trials ...attempts to clarify the interactions of symbionts in this type of nitrogen fixation as well as the influence of abiotic factors on its efficacy. A field experiment was conducted to investigate the influence of garden pea cultivars ('Alicia' and 'Miracle of America'), seed inoculation with nodule bacteria (Rhizobium pisi DSM 30132, and Rhizobium leguminosarum bv. viciae OS–103), and nitrogen fertilization (0, 30 and 60 kg N*ha-1). The observed parameters are: stand density, number of pods, mass of 1,000 grains, mass of pods and grains and grain yield. It was established that all the investigated factors significantly influenced the traits. The seeds inoculated with the indigenous strain R. leguminosarum bv. viciae OS–103 had significantly increased numbers of pods per unit area, grain yield, and the weight of 1,000 grains, while a higher amount of applied nitrogen resulted only in an increase of grain yield. Cultivar 'Alicia' achieved a significantly higher grain yield compared to the 'Miracle of America' cultivars, while application of 60 kg N*ha-1 achieved statistically higher grain yield compared to control. Inoculation with R. leguminosarum bv. viciae OS-103 produced a significantly higher grain yield compared to inoculation with the reference strain R. pisi DSM 30132. The observed parameters were significantly influenced by the garden pea cultivar, seed inoculation with nodule bacteria, and nitrogen fertilization therefore further investigation are needed with new inoculant strains and new cultivars under different agroecological conditions.