Search

Crop maps are highly desired information in modern agriculture as they enable possessors to manage their business in the most optimal way. Usually in remote sensing, crop mapping is performed using satellite images and within-season ground truth samples that are collected in extensive survey campaigns every year, neglecting information and knowledge that past seasons’ classification models provided. This paper assessed different temporal transferring approaches, including transfer learning, together with traditional crop mapping approach to provide an exhaustive comparison. Transferring approaches differed in portion of knowledge utilized from a historical model and that coming from a target season dataset. Three distinct algorithms, Random Forest, Convolutional Neural Network and Transformer, were employed and evaluated using highly dense time series of Sentinel-1 data. Source and target domain were respectively represented by two sets, 2017–2020 and 2021 season data, and 9 different crop types were classified. Results showcased that transferring a model has a great potential in crop mapping when little to no ground truth data is available for the target season. However, traditional approach catches up rather quickly and even surpasses transfer learning approach in terms of performance after a certain portion of target domain data is collected. Without target season ground truth data, model transferring can yield modest crop mapping performance of 78% for F1 score, between 84% and 86% F1 score with transfer learning employed in conjunction with limited target season ground truth (i.e. between 120 and 720 parcels), and 88% F1 score at best with traditional approach (ca. 720 parcels). Even though a good discriminatory is found between different crop types, there is still a room for improvement regarding the least represented classes in the dataset. The study significantly contributes to the area of agricultural monitoring and management by demonstrating the effectiveness of transfer learning while lessening the necessity for extensive and labor-intensive data collection, thereby fostering cost and time efficiency. Utilizing Sentinel-1 data, it provides a practical and efficient solution for agricultural analysis worldwide regardless of cloudiness.

Abstract This article describes a teaching strategy that synergizes computing and management, aimed at the running of complex projects in industry and academia, in the areas of civil engineering, physics, geosciences, and a number of other related fields. The course derived from this strategy includes four parts: (a) Computing with a selected set of modern paradigms—the stress is on Control Flow and Data Flow computing paradigms, but paradigms conditionally referred to as Energy Flow and Diffusion Flow are also covered; (b) Project management that is holistic—the stress is on the wide plethora of issues spanning from the preparation of project proposals, all the way to incorporation activities to follow after the completion of a successful project; (c) Examples from past research and development experiences—the stress is on experiences of leading experts from academia and industry; (d) Student projects that stimulate creativity—the stress is on methods that educators could use to induce and accelerate the creativity of students in general. Finally, the article ends with selected pearls of wisdom that could be treated as suggestions for further elaboration.

Abstract This paper presents a transfer learning approach to the crop classification problem based on time series of images from the Sentinel-2 dataset labeled for two regions: Brittany (France) and Vojvodina (Serbia). During preprocessing, cloudy images are removed from the input data, the time series are interpolated over the time dimension, and additional remote sensing indices are calculated. We chose TransformerEncoder as the base model for knowledge transfer from source to target domain with French and Serbian data, respectively. Even more, the accuracy of the base model with the preprocessing step is improved by 2% when trained and evaluated on the French dataset. The transfer learning approach with fine-tuning of the pre-trained weights on the French dataset outperformed all other methods in terms of overall accuracy 0.94 and mean class recall 0.907 on the Serbian dataset. Our partially fine-tuned model improved recall of crop types that were poorly classified by the base model. In the case of sugar beet, class recall is improved by 85.71%.

Open pit coal mining affects surrounding populated areas, resulting in terrain surface deformation. Surface deformation should be monitored as often as possible to control deformations and prevent potential incidents. This paper analyzes time series deformation estimated from the Sentinel-1 satellite images using the Persistent Scatterer Interferometry method to monitor subsidence rates caused by open pit mining activities. It is possible to measure deformations using classical geodetic methods, but those are rarely used in practice because they are time-consuming and expensive for application in large areas. Using the open access radar images from the Sentinel-1 mission, 513 images from the repository were downloaded between October 2016 and the end of December 2020. We present the processing steps in detail in order to establish a workflow for the automated processing of vertical displacement estimation using open source tools; a total of 402 images were processed: 215 images belonged to the ascending satellite orbit, 187 images belonged to the descending orbit, and 111 images were rejected because of adverse weather conditions. The PS InSAR technique has never been used for the mines of the Republic of Serbia or for land surveying practices related to deformation monitoring. The results based on the Sentinel-1 images were compared with results from geodetic leveling and with neotectonic uplift trends. The trend lines of vertical displacement obtained from PS and corresponding leveling are significantly similar (a Pearson correlation of 85% with a p-value of 0.015). The final evaluation reported results of vertical displacements at the leveling benchmark of −3.4 mm/year with the PS InSAR method and −2.7 mm/year with the leveling method. A comparison of the PS vertical displacements with a settlement model fits reasonably, suggesting that the measurements are valid. As four years of PS time series data is insufficient to establish undisputable conclusions on the neotectonics uplift, extending the time series (covering at least a decade) implies that this approach will become attractive in future neotectonic uplift trend estimations. This study illustrates not only the ability of Sentinel-1 data in mapping vertical deformations, but the obtained results could also be used for geohazard monitoring and land monitoring in general for the area of interest.

A spatiotemporal machine learning framework for automated prediction and analysis of long-term Land Use/Land Cover dynamics is presented. The framework includes: (1) harmonization and preprocessing of spatial and spatiotemporal input datasets (GLAD Landsat, NPP/VIIRS) including five million harmonized LUCAS and CORINE Land Cover-derived training samples, (2) model building based on spatial k-fold cross-validation and hyper-parameter optimization, (3) prediction of the most probable class, class probabilities and model variance of predicted probabilities per pixel, (4) LULC change analysis on time-series of produced maps. The spatiotemporal ensemble model consists of a random forest, gradient boosted tree classifier, and an artificial neural network, with a logistic regressor as meta-learner. The results show that the most important variables for mapping LULC in Europe are: seasonal aggregates of Landsat green and near-infrared bands, multiple Landsat-derived spectral indices, long-term surface water probability, and elevation. Spatial cross-validation of the model indicates consistent performance across multiple years with overall accuracy (a weighted F1-score) of 0.49, 0.63, and 0.83 when predicting 43 (level-3), 14 (level-2), and five classes (level-1). Additional experiments show that spatiotemporal models generalize better to unknown years, outperforming single-year models on known-year classification by 2.7% and unknown-year classification by 3.5%. Results of the accuracy assessment using 48,365 independent test samples shows 87% match with the validation points. Results of time-series analysis (time-series of LULC probabilities and NDVI images) suggest forest loss in large parts of Sweden, the Alps, and Scotland. Positive and negative trends in NDVI in general match the land degradation and land restoration classes, with “urbanization” showing the most negative NDVI trend. An advantage of using spatiotemporal ML is that the fitted model can be used to predict LULC in years that were not included in its training dataset, allowing generalization to past and future periods, e.g. to predict LULC for years prior to 2000 and beyond 2020. The generated LULC time-series data stack (ODSE-LULC), including the training points, is publicly available via the ODSE Viewer. Functions used to prepare data and run modeling are available via the eumap library for Python.

Measurement(s) temperature of air • pressure • volume of hydrological precipitation Technology Type(s) weather station Factor Type(s) digital elevation model (DEM) • topographic wetness index (TWI) • Tropical Rainfall Measuring Mission/Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (TRMM/IMERG) Sample Characteristic - Environment climate Sample Characteristic - Location Serbia Machine-accessible metadata file describing the reported data: https://doi.org/10.6084/m9.figshare.14102393

Abstract Soil property and class maps for the continent of Africa were so far only available at very generalised scales, with many countries not mapped at all. Thanks to an increasing quantity and availability of soil samples collected at field point locations by various government and/or NGO funded projects, it is now possible to produce detailed pan-African maps of soil nutrients, including micro-nutrients at fine spatial resolutions. In this paper we describe production of a 30 m resolution Soil Information System of the African continent using, to date, the most comprehensive compilation of soil samples ( $$N \approx 150,000$$ N ≈ 150 , 000 ) and Earth Observation data. We produced predictions for soil pH, organic carbon (C) and total nitrogen (N), total carbon, effective Cation Exchange Capacity (eCEC), extractable—phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), sulfur (S), sodium (Na), iron (Fe), zinc (Zn)—silt, clay and sand, stone content, bulk density and depth to bedrock, at three depths (0, 20 and 50 cm) and using 2-scale 3D Ensemble Machine Learning framework implemented in the mlr (Machine Learning in R) package. As covariate layers we used 250 m resolution (MODIS, PROBA-V and SM2RAIN products), and 30 m resolution (Sentinel-2, Landsat and DTM derivatives) images. Our fivefold spatial Cross-Validation results showed varying accuracy levels ranging from the best performing soil pH (CCC = 0.900) to more poorly predictable extractable phosphorus (CCC = 0.654) and sulphur (CCC = 0.708) and depth to bedrock. Sentinel-2 bands SWIR (B11, B12), NIR (B09, B8A), Landsat SWIR bands, and vertical depth derived from 30 m resolution DTM, were the overall most important 30 m resolution covariates. Climatic data images—SM2RAIN, bioclimatic variables and MODIS Land Surface Temperature—however, remained as the overall most important variables for predicting soil chemical variables at continental scale. This publicly available 30-m Soil Information System of Africa aims at supporting numerous applications, including soil and fertilizer policies and investments, agronomic advice to close yield gaps, environmental programs, or targeting of nutrition interventions.

There is an increasing need for pan-European spatial datasets, mainly to support the common European Union policies. This has inevitably raised demands for adopting pan-European cartographic projections to visualize the spatial data. The Map Projection Workshop organized by EuroGeographics in 2001 provided a recommendation to the European Commission to adopt the Lambert conformal conic projection for conformal pan-European mapping at the scales smaller or equal to 1:500,000. This paper discusses if the projection is an optimal solution in terms of linear deformations over the mapping region. An optimized CAMPREL (conformal adaptive mapping projection of rotation ellipsoid) projection for the area of interest is proposed as an alternative solution. The projection quality criteria were calculated and compared with those of the Lambert conformal conic projection. The maximal possible absolute linear distortion for conformal mapping of the pan-European area is also given. It has been shown that the CAMPREL projection designed for pan-European mapping better meets the projection selection criteria.

For many decades, kriging and deterministic interpolation techniques, such as inverse distance weighting and nearest neighbour interpolation, have been the most popular spatial interpolation techniques. Kriging with external drift and regression kriging have become basic techniques that benefit both from spatial autocorrelation and covariate information. More recently, machine learning techniques, such as random forest and gradient boosting, have become increasingly popular and are now often used for spatial interpolation. Some attempts have been made to explicitly take the spatial component into account in machine learning, but so far, none of these approaches have taken the natural route of incorporating the nearest observations and their distances to the prediction location as covariates. In this research, we explored the value of including observations at the nearest locations and their distances from the prediction location by introducing Random Forest Spatial Interpolation (RFSI). We compared RFSI with deterministic interpolation methods, ordinary kriging, regression kriging, Random Forest and Random Forest for spatial prediction (RFsp) in three case studies. The first case study made use of synthetic data, i.e., simulations from normally distributed stationary random fields with a known semivariogram, for which ordinary kriging is known to be optimal. The second and third case studies evaluated the performance of the various interpolation methods using daily precipitation data for the 2016–2018 period in Catalonia, Spain, and mean daily temperature for the year 2008 in Croatia. Results of the synthetic case study showed that RFSI outperformed most simple deterministic interpolation techniques and had similar performance as inverse distance weighting and RFsp. As expected, kriging was the most accurate technique in the synthetic case study. In the precipitation and temperature case studies, RFSI mostly outperformed regression kriging, inverse distance weighting, random forest, and RFsp. Moreover, RFSI was substantially faster than RFsp, particularly when the training dataset was large and high-resolution prediction maps were made.

This paper describes the technical development and accuracy assessment of the most recent and improved version of the SoilGrids system at 250m resolution (June 2016 update). SoilGrids provides global predictions for standard numeric soil properties (organic carbon, bulk density, Cation Exchange Capacity (CEC), pH, soil texture fractions and coarse fragments) at seven standard depths (0, 5, 15, 30, 60, 100 and 200 cm), in addition to predictions of depth to bedrock and distribution of soil classes based on the World Reference Base (WRB) and USDA classification systems (ca. 280 raster layers in total). Predictions were based on ca. 150,000 soil profiles used for training and a stack of 158 remote sensing-based soil covariates (primarily derived from MODIS land products, SRTM DEM derivatives, climatic images and global landform and lithology maps), which were used to fit an ensemble of machine learning methods-random forest and gradient boosting and/or multinomial logistic regression-as implemented in the R packages ranger, xgboost, nnet and caret. The results of 10-fold cross-validation show that the ensemble models explain between 56% (coarse fragments) and 83% (pH) of variation with an overall average of 61%. Improvements in the relative accuracy considering the amount of variation explained, in comparison to the previous version of SoilGrids at 1 km spatial resolution, range from 60 to 230%. Improvements can be attributed to: (1) the use of machine learning instead of linear regression, (2) to considerable investments in preparing finer resolution covariate layers and (3) to insertion of additional soil profiles. Further development of SoilGrids could include refinement of methods to incorporate input uncertainties and derivation of posterior probability distributions (per pixel), and further automation of spatial modeling so that soil maps can be generated for potentially hundreds of soil variables. Another area of future research is the development of methods for multiscale merging of SoilGrids predictions with local and/or national gridded soil products (e.g. up to 50 m spatial resolution) so that increasingly more accurate, complete and consistent global soil information can be produced. SoilGrids are available under the Open Data Base License.

Solving the problem of publicly available census data disaggregation has preoccupied numerous researchers intensively. A noteworthy advance in the methodology was made thanks to the contemporary storage and presentation of spatial and socio-economic data in the GIS environment. It is also important that a large number of auxiliary databases (satellite images, theme layers pertaining to land use and land cover, etc.) are publicly available and are periodically supplemented at increasingly shorter time intervals. Soil sealing databases are another class of auxiliary databases that pertain to land areas which have, due to anthropogenic influences, become a water-impermeable layer and indicate the level of spatial development and spatial contents that correlate to the population distribution. The soil sealing database can be a useful tool for dasymetric mapping of population when combined with town planning documentation that describes land use and height of residential buildings. The results of such mapping can help monitor the spatio-temporal dynamics of population trends in periods between two censuses. This study presents a methodology in which a soil-sealing database is combined with auxiliary data in a test area covered by the Master Plan of the Belgrade City, with census data from the year 2002 and the results of the year 2011. The results of the model validation indicate application of the proposed methodology in highly urbanised areas. / Vprašanju razčlenjevanja javno dostopnih podatkov iz popisa so se posvečali že številni raziskovalci. Sodobni načini hranjenja ter predstavitve prostorskih in družbeno-ekonomskih podatkov v okolju GIS so prinesli pomemben napredek v metodologiji. Pri tem je pomembno tudi, da je javno dostopnih veliko pomožnih zbirk podatkov (satelitski posnetki, tematske plasti, ki se nanašajo na rabo zemljišč in pokrovnost tal, itd.), ki se vse hitreje dopolnjujejo. Zbirke podatkov o pozidanosti tal so eden od razredov pomožnih zbirk podatkov o zemljiščih, ki so zaradi antropogenih vplivov postala vodotesna, ter pričajo o stopnji prostorskega razvoja in prostorskih vsebinah, ki se povezujejo z razporeditvijo prebivalstva. Zbirka podatkov o pozidanosti tal je lahko v kombinaciji z dokumentacijo o načrtovanju mesta koristno orodje za dasimetrično kartiranje prebivalstva, ki opisuje rabo tal in višino stanovanjskih zgradb. Rezultati takšnega kartiranja so uporabni pri spremljanju prostorsko-časovne dinamike prebivalstva med dvema popisoma. V študiji je predstavljena metodologija, pri kateri je zbirka podatkov o pozidanosti tal kombinirana s pomožnimi podatki na testnem območju iz glavnega načrta Mesta Beograd, s podatki iz popisa iz leta 2002 in rezultati iz leta 2011. Preverjanje veljavnosti modela kaže, da je predlagana metodologija uporabna na močno urbaniziranih območjih.

The goal of this study is to analyse the predictive performance of the random forest machine learning technique in comparison to commonly used hedonic models based on multiple regression for the prediction of apartment prices. A data set that includes 7407 records of apartment transactions referring to real estate sales from 2008–2013 in the city of Ljubljana, the capital of Slovenia, was used in order to test and compare the predictive performances of both models. Apparent challenges faced during modelling included (1) the non-linear nature of the prediction assignment task; (2) input data being based on transactions occurring over a period of great price changes in Ljubljana whereby a 28% decline was noted in six consecutive testing years; and (3) the complex urban form of the case study area. Available explanatory variables, organised as a Geographic Information Systems (GIS) ready dataset, including the structural and age characteristics of the apartments as well as environmental and neighbourhood information were considered in the modelling procedure. All performance measures (R2 values, sales ratios, mean average percentage error (MAPE), coefficient of dispersion (COD)) revealed significantly better results for predictions obtained by the random forest method, which confirms the prospective of this machine learning technique on apartment price prediction.