Thomas Alexandridis

Vegetation Indices: Advances Made in Biomass Estimation and Vegetation Monitoring in the Last 30 Years

Abstract During the last 30 years Vegetation Indices (VI) have been extensively used for tracing and monitoring vegetation conditions, such as health, growth levels, production, water and nutrients stress, etc. In this paper the characteristics of over 20 VIs based on the VNIR spectrum are described in order to provide the reader with adequate material to form a picture of their nature and purpose. It is not, though, a review article due to the fact that a huge volume of work exists all over the world and a simple lining up of the related papers would not contribute to an understanding of the usefulness of VIs. A limited number of review work is included, together with research results from various operational and research applications of VI for wheat damage assessment in Northern Greece.

Decision Fusion of GA Self-Organizing Neuro-Fuzzy Multilayered Classifiers for Land Cover Classification Using Textural and Spectral Features

A novel Self-Organizing Neuro-Fuzzy Multilayered Classifier, the GA-SONeFMUC model, is proposed in this paper for land cover classification of multispectral images. The model is composed of generic fuzzy neuron classifiers (FNCs) arranged in layers, which are implemented by fuzzy rule-based systems. At each layer, parent FNCs are combined to generate a descendant FNC at the next layer with higher classification accuracy. To exploit the information acquired by the parent FNCs, their decision supports are combined using a fusion operator. As a result, a data splitting is devised within each FNC, distinguishing those pixels that are currently correctly classified to a high certainty grade from the ambiguous ones. The former are handled by the fuser, while the ambiguous pixels are further processed to enhance their classification confidence. The GA-SONeFMUC structure is determined in a self-constructing way via a structure-learning algorithm with feature selection capabilities. The parameters of the models obtained after structure learning are optimized using a real-coded genetic algorithm. For effective classification, we formulated three input sets containing spectral and textural feature types. To explore information coming from different feature sources, we apply a classifier fusion approach at the final stage. The outputs of individual classifiers constructed from each input set are combined to provide the final assignments. Our approach is tested on a lake-wetland ecosystem of international importance using an IKONOS image. A high-classification performance of 92.02% and of 75.55% for the wetland zone and the surrounding agricultural zone is achieved, respectively.

Using Earth Observation to update a Natura 2000 habitat map for a wetland in Greece.

The European Habitats Directive 92/43/EEC has defined the need for the conservation of habitats and species with the adoption of appropriate measures. Within the Natura 2000 ecological network of special areas of conservation, natural habitats will be monitored to ensure the maintenance or restoration of their composition, structure and extent. The European Space Agencys GlobWetland project has provided remotely sensed products for several Ramsar wetlands worldwide, such as detailed land cover-land use, water cycle and inundated vegetation maps. This paper presents the development and testing of an operational methodology for updating a wetlands habitat map using the GlobWetland products, and the evaluation of the extent to which GlobWetland products have contributed to the habitat map updating. The developed methodology incorporated both automated and analyst-supervised techniques to photo-interpret, delineate, refine, and evaluate the updated habitat polygons. The developed methodology was proven successful in its application to the wetland complex of the Axios-Loudias-Aliakmon delta (Greece). The resulting habitat map met the European and Greek national requirements. Results revealed that GlobWetland products were a valuable contribution, but source data (enhanced satellite images) were necessary to discriminate spectrally similar habitats. Finally, the developed methodology can be modified for original habitat mapping.

An estimation of the optimum temporal resolution for monitoring vegetation condition on a nationwide scale using MODIS/Terra data

Monitoring vegetation condition is an important issue in the Mediterranean region, in terms of both securing food and preventing fires. The recent abundance of remotely sensed data, such as the daily availability of MODIS imagery, raises the issue of appropriate temporal sampling when monitoring vegetation: under‐sampling may not accurately describe the phenomenon under consideration, whilst over‐sampling would increase the cost of the project without additional benefit. The aim of this work is to estimate the optimum temporal resolution for vegetation monitoring on a nationwide scale using 250 m MODIS/Terra daily images and composites. Specific objectives include: (i) an investigation into the optimum temporal resolution for monitoring vegetation condition during the dry season on a nationwide scale using time‐series analysis of Normalized Difference Vegetation Index, NDVI, datasets, (ii) an investigation into whether this temporal resolution differs between the two major vegetation categories of natural and managed vegetation, and (iii) a quality assessment of multi‐temporal NDVI composites following the proposed optimum temporal resolution. A time‐series of daily NDVI data is developed for Greece using MODIS/Terra 250 m imagery. After smoothing to remove noise and cloud influence, it is subjected to temporal autocorrelation analysis, and its level of significance is the adopted objective function. In addition, NDVI composites are created at various temporal resolutions and compared using qualitative criteria. Results indicate that the proposed optimum temporal resolution is different for managed and natural vegetation. Finally, quality assessment of the multi‐temporal NDVI composites reveals that compositing at the proposed optimum temporal resolution could derive products that are useful for operational monitoring of vegetation.

Wheat yield prediction using machine learning and advanced sensing techniques

A yield potential prediction model was developed and evaluated for winter wheat.The soil parameters were estimated through online soil spectroscopy with a prototype sensor.Main inputs for yield potential prediction were estimated soil parameters and remote sensing vegetation indices.The proposed architecture provided visual information about the factors affecting the yield potential. Understanding yield limiting factors requires high resolution multi-layer information about factors affecting crop growth and yield. Therefore, on-line proximal soil sensing for estimation of soil properties is required, due to the ability of these sensors to collect high resolution data (>1500 sample per ha), and subsequently reducing labor and time cost of soil sampling and analysis. The aim of this paper is to predict within field variation in wheat yield, based on on-line multi-layer soil data, and satellite imagery crop growth characteristics. Supervised self-organizing maps capable of handling existent information from different soil and crop sensors by utilizing an unsupervised learning algorithm were used. The performance of counter-propagation artificial neural networks (CP-ANNs), XY-fused Networks (XY-Fs) and Supervised Kohonen Networks (SKNs) for predicting wheat yield in a 22ha field in Bedfordshire, UK were compared for a single cropping season. The self organizing models consisted of input nodes corresponded to feature vectors formed from normalized values of on-line predicted soil parameters and the satellite normalized difference vegetation index (NDVI). The output nodes consisted of yield isofrequency classes, which were predicted from the three trained networks. Results showed that cross validation based yield prediction of the SKN model for the low yield class exceeded 91% which can be considered as highly accurate given the complex relationship between limiting factors and the yield. The medium and high yield class reached 70% and 83% respectively. The average overall accuracy for SKN was 81.65%, for CP-ANN 78.3% and for XY-F 80.92%, showing that the SKN model had the best overall performance.

Integrated Methodology for Estimating Water Use in Mediterranean Agricultural Areas

Agricultural use is by far the largest consumer of fresh water worldwide, especially in the Mediterranean, where it has reached unsustainable levels, thus posing a serious threat to water resources. Having a good estimate of the water used in an agricultural area would help water managers create incentives for water savings at the farmer and basin level, and meet the demands of the European Water Framework Directive. This work presents an integrated methodology for estimating water use in Mediterranean agricultural areas. It is based on well established methods of estimating the actual evapotranspiration through surface energy fluxes, customized for better performance under the Mediterranean conditions: small parcel sizes, detailed crop pattern, and lack of necessary data. The methodology has been tested and validated on the agricultural plain of the river Strimonas (Greece) using a time series of Terra MODIS and Landsat 5 TM satellite images, and used to produce a seasonal water use map at a high spatial resolution. Finally, a tool has been designed to implement the methodology with a user-friendly interface, in order to facilitate its operational use.

Evaluation of UAV imagery for mapping Silybum marianum weed patches

ABSTRACT Silybum marianum (L.) Gaertn weed has the tendency to grow in patches. In order to perform site-specific weed management, determining the spatial distribution of weeds is important for its eradication. Remote sensing has been used to perform species discrimination and it offers promising techniques for operational weed mapping. In the present study, the feasibility of high-resolution imaging for S. marianum detection and mapping is reported. A multispectral camera (green–red–near-infrared) mounted on a fixed wing unmanned aerial vehicle (UAV) was used for the acquisition of high-resolution images with pixel size of 0.1 m. The maximum likelihood (ML) classifier was used to classify the S. marianum among other weed species present in a field, with Avena sterilisL. being predominant. As input to the classifier, the three spectral bands and the texture were used. The scale of the mapping was varied by degrading the image resolution to evaluate classification performance, with 1 m resolution providing the highest classification accuracy. The classification rates obtained using ML reached an overall accuracy of 87.04% with a K-hat statistic of 74%. The results prove the feasibility of operational S. marianum mapping using UAV and multispectral imaging.

Rapid error assessment for quantitative estimations from Landsat 7 gap-filled images

The failure of the Scan Line Corrector (SLC) of the Landsat ETM+ (Enhanced Thematic Mapper Plus) instrument in 2003 had resulted in missing values for 22% of each scene. As the remaining pixels were of high quality, several procedures had been developed to fill the gaps and increase the usability of the SLC-off images. In this letter, a methodology is presented to assess the error when estimating quantitative parameters from gap-filled Landsat 7 images. The error from the gap-filling procedure was estimated using an external reference image. The methodology was applied in a Mediterranean river basin using two types of gap-filling methods and the error was estimated for leaf area index (LAI), actual evapotranspiration (ETa) and soil moisture in the rootzone (SMrz), three remotely sensed products which are commonly used in hydrological studies. The results suggest that the interpolation method had lower errors in all examined products. The proposed methodology is an imperative step that each user of gap-filled products could use to estimate the associated error before using the maps.

A novel self-organizing neuro-fuzzy multilayered classifier for land cover classification of a VHR image

A novel self‐organizing neuro‐fuzzy multilayered classifier (SONeFMUC) is introduced in this paper, with feature selection capabilities, for the classification of an IKONOS image. The structure of the proposed network is developed in a sequential fashion using the group method of data handling (GMDH) algorithm. The node models, regarded as generic classifiers, are represented by fuzzy rule‐based systems, combined with a fusion scheme. A data splitting mechanism is incorporated to discriminate between correctly classified and ambiguous pixels. The classifier was tested on the wetland of international importance of Lake Koronia, Greece, and the surrounding agricultural area. To achieve higher classification accuracy, the image was decomposed into two zones: the wetland and the agricultural zones. Apart from the initial bands, additional input features were considered: textural features, intensity–hue–saturation (IHS) and tasseled cap transformation. To assess the quality of the suggested model, the SONeFMUC was compared with a maximum likelihood classifier (MLC). The experimental results show that the SONeFMUC exhibited superior performance to the MLC, providing less confusion of the dominant classes in both zones. In the wetland zone, an overall accuracy of 89.5% was attained.

Combining remotely sensed surface energy fluxes and GIS analysis of groundwater parameters for irrigation system assessment

Abstract Despite being necessary for effective water management, the assessment of an irrigation system requires a large amount of input data for the estimation of related parameters and indicators, which are seldom measured in a regular and reliable manner. In this work, spatially distributed surface energy balance fluxes and geographical information systems analysis of multiple groundwater parameters were used to estimate water availability, supply, and demand, in order to calculate water-accounting indicators. This methodology was used to evaluate the performance of an irrigation system in the Pinios river basin (Greece) at two selected years of high and low water availability. Time series of archived satellite images and groundwater measurements have been used for past years to support comparative analyses, due to the limited availability of actual water measurements. The resulting maps from the proposed methodology show that the performance of the irrigation system varied across space and time due to differences in its characteristics and changes in its operation, driven by fluctuation of water availability and the response of stakeholders to water depletion. Irrigation districts with unsustainable water management were identified and, together with those with slow and/or limited groundwater recharge, were brought to the attention of water managers. The observed differences in the system operation between the wet and dry years were attributed not only to the hydrological conditions of each year, but also to the changing behaviour of farmers and the improvement actions of the water managers.