Alexander Gikov

This data descriptor presents a dataset comprising crop and soil parameters measured in winter wheat fields near the town of Knezha, Bulgaria. The data were collected as part of a project evaluating the potential of vegetation indices derived from Sentinel-2 satellite imagery to predict biophysical and biochemical crop parameters. The core dataset consists of measurements obtained from 20 m × 20 m field plots and includes a broad range of parameters: leaf area index, fraction of absorbed photosynthetically active radiation, vegetation cover fraction, chlorophyll content, above-ground biomass, plant nitrogen content, biological yield, surface soil moisture, spectral reflectance, plant density, crop height, visual assessments of disease or pest damage, and data on weed occurrence. The dataset is complemented by unmanned aerial vehicle imagery, crop calendars, and field management information. The main soil types in the study area were characterized through soil profiles, while meteorological data were obtained from an automated weather station. The data were collected during the 2016–2017 and 2017–2018 agricultural seasons. The dataset is freely available for download and serves as a valuable resource for researchers in remote sensing—particularly for validating satellite-derived products—as well as for specialists involved in winter wheat monitoring, modeling, and agronomic studies.

Unmanned aerial vehicles (UAVs) are extensively used to gather remote sensing data, offering high image resolution and swift data acquisition despite being labor-intensive. In contrast, satellite-based remote sensing, providing sub-meter spatial resolution and frequent revisit times, could serve as an alternative data source for phenotyping. In this study, we separately evaluated pan-sharpened Pléiades satellite imagery (50 cm) and UAV imagery (2.5 cm) to phenotype durum wheat in small-plot (12 m × 1.10 m) breeding trials. The Gaussian process regression (GPR) algorithm, which provides predictions with uncertainty estimates, was trained with spectral bands and а selected set of vegetation indexes (VIs) as independent variables. Grain protein content (GPC) was better predicted with Pléiades data at the growth stage of 20% of inflorescence emerged but with only moderate accuracy (validation R²: 0.58). The grain yield (GY) and protein yield (PY) were better predicted using UAV data at the late milk and watery ripe growth stages, respectively (validation: R² 0.67 and 0.62, respectively). The cumulative VIs (the sum of VIs over the available images within the growing season) did not increase the accuracy of the models for either sensor. When mapping the estimated parameters, the spatial resolution of Pléiades revealed certain limitations. Nevertheless, our findings regarding GPC suggested that the usefulness of pan-sharpened Pléiades images for phenotyping should not be dismissed and warrants further exploration, particularly for breeding experiments with larger plot sizes.

This study tested the potential of parametric and nonparametric regression modeling utilizing multispectral data from two different unoccupied aerial vehicles (UAVs) as a tool for the prediction of and indirect selection of grain yield (GY) in barley breeding experiments. The coefficient of determination (R²) of the nonparametric models for GY prediction ranged between 0.33 and 0.61 depending on the UAV and flight date, where the highest value was achieved with the DJI Phantom 4 Multispectral (P4M) image from 26 May (milk ripening). The parametric models performed worse than the nonparametric ones for GY prediction. Independent of the retrieval method and UAV, GY retrieval was more accurate in milk ripening than dough ripening. The leaf area index (LAI), fraction of absorbed photosynthetically active radiation (fAPAR), fraction vegetation cover (fCover), and leaf chlorophyll content (LCC) were modeled at milk ripening using nonparametric models with the P4M images. A significant effect of the genotype was found for the estimated biophysical variables, which was referred to as remotely sensed phenotypic traits (RSPTs). Measured GY heritability was lower, with a few exceptions, compared to the RSPTs, indicating that GY was more environmentally influenced than the RSPTs. The moderate to strong genetic correlation of the RSPTs to GY in the present study indicated their potential utility as an indirect selection approach to identify high-yield genotypes of winter barley.

The utility of unmanned aerial vehicles (UAV) imagery in retrieving phenotypic data to support plant breeding research has been a topic of increasing interest in recent years. The advantages of image-based phenotyping are related to the high spatial and temporal resolution of the retrieved data and the non-destructive and rapid method of data acquisition. This study trains parametric and nonparametric regression models to retrieve leaf area index (LAI), fraction of absorbed photosynthetically active radiation (fAPAR), fractional vegetation cover (fCover), leaf chlorophyll content (LCC), canopy chlorophyll content (CCC), and grain yield (GY) of winter durum wheat breeding experiment from four-bands UAV images. A ground dataset, collected during two field campaigns and complemented with data from a previous study, is used for model development. The dataset is split at random into two parts, one for training and one for testing the models. The tested parametric models use the vegetation index formula and parametric functions. The tested nonparametric models are partial least square regression (PLSR), random forest regression (RFR), support vector regression (SVR), kernel ridge regression (KRR), and Gaussian processes regression (GPR). The retrieved biophysical variables along with traditional phenotypic traits (plant height, yield, and tillering) are analysed for detection of genetic diversity, proximity, and similarity in the studied genotypes. Analysis of variance (ANOVA), Duncan’s multiple range test, correlation analysis, and principal component analysis (PCA) are performed with the phenotypic traits. The parametric and nonparametric models show close results for GY retrieval, with parametric models indicating slightly higher accuracy (R² = 0.49; RMSE = 0.58 kg/plot; rRMSE = 6.1%). However, the nonparametric model, GPR, computes per pixel uncertainty estimation, making it more appealing for operational use. Furthermore, our results demonstrate that grain filling was better than flowering phenological stage to predict GY. The nonparametric models show better results for biophysical variables retrieval, with GPR presenting the highest prediction performance. Nonetheless, robust models are found only for LAI (R² = 0.48; RMSE = 0.64; rRMSE = 13.5%) and LCC (R² = 0.49; RMSE = 31.57 mg m⁻²; rRMSE = 6.4%) and therefore these are the only remotely sensed phenotypic traits included in the statistical analysis for preliminary assessment of wheat productivity. The results from ANOVA and PCA illustrate that the retrieved remotely sensed phenotypic traits are a valuable addition to the traditional phenotypic traits for plant breeding studies. We believe that these preliminary results could speed up crop improvement programs; however, stronger interdisciplinary research is still needed, as well as uncertainty estimation of the remotely sensed phenotypic traits.

At the place where Banderitsa and Demyanitsa rivers and their tributaries ceased to be confined to their narrow valleys one of the most impressive alluvial fans in South Bulgaria is formed – the one of Glazne river. The river valley morphology, as well as the evolution and the position of the Glazne fan, are controlled by the active normal faulting in the NE slopes and foot of Pirin Mountain. Тhe Quaternary glaciations produced large volumes of debris in the river valleys. There is an agreement that the processes of sediment transfer from the mountain to the Razlog graben have been highly active at the time and immediately after the Pleistocene glaciations. As a result, in the Bulgarian geological and geomorphological reports and scientific papers, the age of the alluvial fans at the NE foot of Pirin Mountain is assumed to be Late Pleistocene–Early Holocene. The results of our studies require a re-evaluation of these ideas. Using widely accepted methods for natural hazard assessment, our field and historical research and analysis allow us to claim that the Glazne fan is active. Parts of Bansko, that are situated in the upper-middle parts of the alluvial fan, have been affected by at least two significant debris floods during the XX century. The conducted measures to control river behavior have an important effect on risk reduction, yet they lead to significant modification of the zones of active aggradation and this means that new actions against future events must be taken.

At the end of the summer of 2017, the weather in Bulgaria was very dry and hot and this contributed for the occurrence of a series of wild fires. The largest and most prolonged was the wild fire that raged in Kresna gorge. The fire broke out on August 24 and lasted for nearly a week. The aim of the paper is to make an assessment of the damage caused by the August fire and by a smaller wild fire in the same region, which broke out on July 13, 2017. The assessment is based on images from the satellites Sentinel-2A and -2B. For the identification and mapping of burned areas the spectral indices NBR (Normalized Burn Ratio) and dNBR (difference NBR) were used. Map of burn severity within the fire perimeter was compiled by reclassifying the dNBR values into three classes. The total burned area was 2199.1 ha; of this, the August fire covered 96%. The greatest part of the fire scar, over 50% of the area, is characterized with medium burn severity (dNBR between 0.27 and 0.66). Both, the lowest (dNBR between 0.1 and 0.27) and highest (dNBR>0.66) burn severity levels covered about ¼ of the total burned area. A map of the major land cover types in the study area was prepared in order to assess the most severely affected land cover types. The map is derived from supervised SVM (Support Vector Machine) classification of a Sentinel-2B image acquired before the fires. It was found that the coniferous forest class is dominated by the highest burn severity level (43.7% of the area). In the broadleaf forest class the areas with the lowest level of burn severity prevail (53.5%). The main reason is that coniferous forests in the study area represent artificial plantations created at altitude lower than the usual altitude of coniferous species in the region. Therefore they are much more susceptible to fire than broadleaf species which are native to the oak belt.

This paper presents the results of a sub-pixel classification of crop types in Bulgaria from PROBA-V 100 m normalized difference vegetation index (NDVI) time series. Two sub-pixel classification methods, artificial neural network (ANN) and support vector regression (SVR) were used where the output was a set of area fraction images (AFIs) at 100 m resolution with pixels containing estimated area fractions of each class. High-resolution maps of two test sites derived from Sentinel-2 classifications were used to obtain training data for the sub-pixel classifications. The estimated area fractions have a good correspondence with the true area fractions when aggregated to regions of 10 × 10 km², especially when the SVR method was used. For the five dominant classes in the test sites the R² obtained after the aggregation was 86% (winter cereals), 81% (sunflower), 92% (broad-leaved forest), 89% (maize), and 67% (grasslands) when the SVR method was used.