The olive tree (Olea europaea L.) is an emblematic, long-living fruit tree species of profound economic and environmental importance. This study is a literature review of articles published during the last 10 years about the role of beneficial microbes [Arbuscular Mycorrhizal Fungi (AMF), Plant Growth Promoting Rhizobacteria (PGPR), Plant Growth Promoting Fungi (PGPF), and Endophytes] on olive tree plant growth and productivity, pathogen control, and alleviation from abiotic stress. The majority of the studies examined the AMF effect using mostly Rhizophagus irregularis and Glomus mosseae species. These AMF species stimulate the root growth improving the resistance of olive plants to environmental and transplantation stresses. Among the PGPR, the nitrogen-fixing bacteria Azospirillum sp. and potassium- and phosphorous-solubilizing Bacillus sp. species were studied extensively. These PGPR species were combined with proper cultural practices and improved considerably olive plant’s growth. The endophytic bacterial species Pseudomonas fluorescens and Bacillus sp., as well as the fungal species Trichoderma sp. were identified as the most effective biocontrol agents against olive tree diseases (e.g., Verticillium wilt, root rot, and anthracnose).

Organic fertilization has been proposed as an alternative approach to supply nutrients for crops, in the frame of organic and sustainable agriculture, with the aim to decrease high inorganic fertilization rates, protect the environment and decrease production costs for farmers. Since different types of organic fertilizers, such as manures, olive mill wastewater (OMW), sewage sludge (SS), crushed pruning wastes, composts and cover crops, exist as soil amendments to improve soil fertility, enhance plant nutrition and sustain the productivity of tree crops, their role as biofertilizers has been fully analyzed under the most important published papers. In addition, the benefits and drawbacks of organic fertilization, in a comparative approach with inorganic fertilization, are presented and discussed. Within the most important advantages of organic fertilizers, the enhancement of beneficial soil microorganisms and the improvement in soil physical properties and fertility should be included, while their most important disadvantage is their inability to directly satisfy the prompt N nutritional needs of tree crops, due to slow N mineralization rates. Finally, some novel aspects on the interrelation among innovative organic fertilizers for tree crops, sustainable field management, crop productivity and fruit quality are also included in this review, under the light of the most important and recent research data existing in the literature, with the aim to provide recommendations and future directions for organic fertilizers by tree growers.

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.