Sown Biodiverse Pastures (SBP) are the basis of a high-yield grazing system tailored for Mediterranean ecosystems and widely implemented in Southern Portugal. The application of precision farming methods in SBP requires cost-effective monitoring using remote sensing (RS). The main hurdle for the remote monitoring of SBP is the fact that the bulk of the pastures are installed in open Montado agroforestry systems. Sparsely distributed trees cast shadows that hinder the identification of the underlaying pasture using Unmanned Aerial Vehicles (UAV) imagery. Image acquisition in the Spring is made difficult by the presence of flowers that mislead the classification algorithms. Here, we tested multiple procedures for the geographical, object-based image classification (GEOBIA) of SBP, aiming to reduce the effects of tree shadows and flowers in open Montado systems. We used remotely sensed data acquired between November 2017 and May 2018 in three Portuguese farms. We used three machine learning supervised classification algorithms: Random Forests (RF), Support Vector Machine (SVM) and Artificial Neural Networks (ANN). We classified SBP based on: (1) a single-period image for the maximum Normalized Difference Vegetation Index (NDVI) epoch in each of the three farms, and (2) multi-temporal image stacking. RF, SVM and ANN were trained using some visible (red, green and blue bands) and near-infrared (NIR) reflectance bands, plus NDVI and a Digital Surface Model (DSM). We obtained high overall accuracy and kappa index (higher than 79% and 0.60, respectively). The RF algorithm had the highest overall accuracy (more than 92%) for all farms. Multitemporal image classification increased the accuracy of the algorithms. as it helped to correctly identify as SBP the areas covered by tree shadows and flower patches, which would be misclassified using single image classification. This study thus established the first workflow for SBP monitoring based on remotely sensed data, suggesting an operational approach for SBP identification. The workflow can be applied to other types of pastures in agroforestry regions to reduce the effects of shadows and flowering in classification problems.

Accurate crop data are essential for reliable irrigation water requirements (IWR) calculations, which can be acquired during the crop growth season for a given region using earth observation (EO) satellite time series. In addition, a relationship between crop coefficients and the NDVI can be established to allow for crop evapotranspiration estimation. The main objective of the present work was to develop a methodology, and its implementation in an application software, to extract crop parameters from EO image time series for a set of parcels of different types of crops, to be used as input data for a soil water balance model to compute IWR. The methodology was tested at two distinct test sites, the Vila Franca de Xira (site I) and Vila Velha de Ródão (site II) municipalities, Portugal. Landsat-7 and −8 images acquired from April to October 2013 were used for site I, while SPOT-5 Take-5 images from April to September 2015 were considered for site II. EO data were used to estimate the basal crop coefficients, planting dates, and crops growth stage lengths. Based on crop, soil and meteorological data, the IWR for the main crops of both test regions were estimated using the IrrigRotation model. The crop coefficient curves obtained from the EO data proved to be reliable for IWR estimation.

In this study an optical empirical algorithm was used to evaluate the robustness of some experimental processes to derive Satellite-Derived Bathymetric (SDB) models for shallow waters in the Minho river mouth, Caminha, Portugal. Multiple procedures to calibrate and modelling SDB calculation were tested. Regarding to these procedures, the following approaches were studied: i) Based on TOA and WL reflectance, a linear regression, as proposed by Stumpf et al.[3], and a quadratic regression to modelling the SDB algorithm were tested; (ii) several in situ data sets acquired at different epochs were used in order to investigate the dependency of the temporal and the spatial density of the calibration sample; (iii) The tidal level was considered in order to provide a tide height at the time of image acquisition to depth models calibration. Numerous SDB models from Sentinel-2 imagery were derived and compared with a reference hydrographic survey data. The results show that SDB models present a good level of reliability regardless of the acquisition date or source of the in situ datasets. Coefficients of determination (R²) higher than 50% were obtained for the majority of the tested procedures. The quadratic modelling approach also appear to retrieve SDB models in agreement with hydrographic surveys data. In addition, the results demonstrate that the SDB information is influenced by the spatial density variation of the calibration datasets. Furthermore, the operational capabilities of the synergy of optical and synthetic aperture radar (SAR) data to derive SDB information in shallow waters will be briefly discussed.