No abstract

No abstract

In the current paper we assess different machine learning (ML) models and hybrid geostatistical methods in the prediction of soil pH using digital elevation model derivates (environmental covariates) and co-located soil parameters (soil covariates). The study was located in the area of Grevena, Greece, where 266 disturbed soil samples were collected from randomly selected locations and analyzed in the laboratory of the Soil and Water Resources Institute. The different models that were assessed were random forests (RF), random forests kriging (RFK), gradient boosting (GB), gradient boosting kriging (GBK), neural networks (NN), and neural networks kriging (NNK) and finally, multiple linear regression (MLR), ordinary kriging (OK), and regression kriging (RK) that although they are not ML models, they were used for comparison reasons. Both the GB and RF models presented the best results in the study, with NN a close second. The introduction of OK to the ML models’ residuals did not have a major impact. Classical geostatistical or hybrid geostatistical methods without ML (OK, MLR, and RK) exhibited worse prediction accuracy compared to the models that included ML. Furthermore, different implementations (methods and packages) of the same ML models were also assessed. Regarding RF and GB, the different implementations that were applied (ranger-ranger, randomForest-rf, xgboost-xgbTree, xgboost-xgbDART) led to similar results, whereas in NN, the differences between the implementations used (nnet-nnet and nnet-avNNet) were more distinct. Finally, ML models tuned through a random search optimization method were compared with the same ML models with their default values. The results showed that the predictions were improved by the optimization process only where the ML algorithms demanded a large number of hyperparameters that needed tuning and there was a significant difference between the default values and the optimized ones, like in the case of GB and NN, but not in RF. In general, the current study concluded that although RF and GB presented approximately the same prediction accuracy, RF had more consistent results, regardless of different packages, different hyperparameter selection methods, or even the inclusion of OK in the ML models’ residuals.

In this paper, the development of a Web-based GIS system for the monitoring and assessment of the Black Sea is presented. The integrated multilevel system is based on the combination of terrestrial and satellite Earth observation data through the technological assets provided by innovative information tools and facilities. The key component of the system is a unified, easy to update geodatabase including a wide range of appropriately selected environmental parameters. The collection procedure of current and historical data along with the methods employed for their processing in three test areas of the current study are extensively discussed, and special attention is given to the overall design and structure of the developed geodatabase. Furthermore, the information system includes a decision support component (DSC) which allows assessment and effective management of a wide range of heterogeneous data and environmental parameters within an appropriately designed and well-tested methodology. The DSC provides simplified and straightforward results based on a classification procedure, thus contributing to a monitoring system not only for experts but for auxiliary staff as well. The examples of the system's functionality that are presented highlight its usability as well as the assistance that is provided to the decision maker. The given examples emphasize on the Danube Delta area; however, the information layers of the integrated system can be expanded in the future to cover other regions, thus contributing to the development of an environmental monitoring system for the entire Black Sea.

Many of the old geodetic reference frames which realized in the previous decades using classical observations carry biases. These biases are mainly caused due to the problematic observations and/or the tectonic motion. That is the case of the official Greek geodetic reference frame which consists of classical and satellite observations. Herein, we present a rigorous approach of the reconstruction of the Greek official reference frame based on the modern geodetic reference frames and their ability to express the spatial position and the dynamic change of the stations. We applied the rigorous approach to ninety stations located in Greece and we compare it with the officially accepted procedure. We found a consistency at 59.4cm between the rigorous and the officially accepted approaches, respectively. The associated mean bias estimation was estimated at 51.4 cm, indicating the resistance of a rather large amount of systematic effects. In addition, the observed discrepancies between the two approaches show great inhomogeneity all over the country. Keywords: Geodetic Reference System (GRS), GRS Datum, transformation between two GRS, ETRS89Resumo: Muitos dos antigos sistemas geodésicos de referência que foram realizados nas últimas décadas por meio de observações clássicas apresentam erros. Estes erros são principalmente oriundos de problemas nas observações e/ou movimentos tectônicos. Este é o caso do sistema geodésico de referência Grego, que consiste de observações clássicas e por satélites. Neste trabalho apresenta-se uma abordagem rigorosa para a reconstrução deste sistema geodésico de referência baseada nos sistemas de referência modernos e na sua habilidade de expressar a posição espacial e as mudanças dinâmicas das estações de referência. Foi aplicada a abordagem rigorosa a noventa estações localizadas na Grécia e foi realizada uma comparação com o procedimento oficialmente aceito. Encontrou-se uma consistência de 59,4 cm entre os procedimentos rigoroso e oficial. A estimativa de erro associada foi de 51,4 cm, indicando grande efeito sistemático. Além disso, as discrepâncias observadas entre as duas abordagens demonstram falta de homogeneidade ao longo do país.Palavras-chave: Datum Clássico; Transformação; Campo de velocidade; ETRS89; Sistema de referência.