Instructions: Please click on the title of the Research Highlight Presentation you would like to view. The Pre-Recorded Research Highlights will be viewed on-your-own, discussed during the workshop, and questions will be directed to the authors during the discussions on DAY 1 and 2.
Note: The Research Highlight presentations below are hosted privately on YouTube’s video platform and are not publicly accessible beyond the links provided on this webpage.
Registration Link: https://forms.gle/ipvkYWHd66pFix5T6
Zoom Meeting Link: https://us02web.zoom.us/j/84447564710?pwd=QWxvS1plemlkc3JQUnZESkljdzNtUT09
1. Long-term assessment of drought in Romania agricultural areas based on MODIS derived DSI
Authors: Claudiu Angearu, Irina Ontel, George Boldeanu, Denis Mihailescu, Argentina Nertan, Vasile Craciunescu, Simona Catana, Anisoara Irimescu
Institution: National Meteorological Administration of Romania
Abstract: Drought Severity Index (DSI) is a very versatile index, a standardized one, that uses Evapotranspiration (ET) and Normalized Difference Vegetation Index (NDVI), to assess drought based on plants stress. The main purpose of the study was to analyze the performance of the DSI and its validation based on different data sources (meteorological data, soil moisture content, agricultural production). Also, the multi-temporal and trends analysis represented an important section of the present study. The DSI was computed based on Terra MODIS satellite image, mostly due to the temporal resolution and spatial resolution which was the best to assess such phenomenon. The study was concentrated on three major agricultural areas: 2 sub divisions of Romanian Plain (Baragan and Oltenia Plain, the most fertile land in Romania) and Banat Plain located in western Romania. DSI was computed using a 19 years archive worth of MODIS data over Romania during the vegetation season, accounted as the period from April through September. As main results it identified that Baragan and Oltenia Plain are more prone to drought than Banat Plain, as especially in the 2002, 2007, 2012 years confirmed also by the SPEI and SMA.
2. Agricultural ecosystem drought analysis based on Vegetation Health Index in Baragan Plain, Romania
Author: Claudiu-Valeriu Angearu
Romanian National Meteorological Administration, 97 Bucharest-Ploieşti, Bucharest, Romania; Romanian Academy, Institute of Geography, 12 Dimitrie Racoviţă, Bucharest, Romania
Abstract:The drought occurrence and its severity from Baragan Plain has been identified based on Vegetation Health Index (VHI) computed from MODIS satellite products, from 30th of March to 29th of September, during 2000-2019.
According to the VHI mean and the weights of the drought types, the longest and the most severely affected periods by drought occurred in 2007, 2000, 2012, 2001 and 2003. The year 2007 is remarkable by the longest dry period, as well by the high intensity of the drought. The analysis has also highlighted the favorable years to vegetation (2004, 2005 and 2010).
The analysis of the VHI index indicates: i) drought occured at the beginning of the season in 2002 and 2003; (ii) a prolonged drought period during the vegetation season in 2000 and 2007; (iii) drought occurrence at the end of the vegetation season in 2012; iv) starting with 2008, the drought occurs more frequently in the last two months of the vegetation season; v) the drought prone areas are located in the Eastern and South-East of the Baragan Plain; vi) 36% of the agricultural area of the Baragan Plain was affected by drought over the analyzed period.
Keywords: Baragan Plain, drought, Vegetation Health Index, MODIS
3. Toward automated forest disturbance inventory using remote sensing data: Forest wildfire in Bulgaria
Authors: Dessislava Ganeva1, Lachezar Filchev1, Martin Schlerf2, Ulrich Leopold2, and Thomas Udelhoven2
1Space Research and Technology Institute – Bulgarian Academy of Sciences
2Luxemburg Institute of Science and Technology
Abstract: Forest disturbance inventory service and near real time mapping has emerged as a strong need for forestry decision making and management in Bulgaria where over one third of its area is covered by forests. The results of burned areas identification and mapping using the spectral indices NBR (Normalized Burn Ratio) and dNBR (difference NBR) are very similar to the results using time series trend analysis (linear regression model and seasonal Kendall test). The Earth Observation Time Series Analysis (EOTSA) Toolbox, one component of a larger ESA-funded project called PROBA-V MEP-TPS, implements this trend analysis with Proba-V data. EOTSA gives one possible technical solution for the prototype version of another ESA-funded project Forest Disturbance Inventory using Remote Sensing (FoReS) in Bulgaria.
4. Land cover changes inside a post-fire forest scars in Serbia based on satellite Sentinel-2 data
Authors: Olga Brovkina1, Marko Stojanović1, Slobodan Milanović2
1Global Change Research Institute of the Czech Academy of Science, Brno, Czech Republic; email@example.com, firstname.lastname@example.org
2Chair of Forest Protection, Faculty of Forestry, University of Belgrade, Belgrade Serbia; email@example.com
Abstract: The study proposes a method for monitoring fire impact using Sentinel-2 satellite data by combining spectral and textural features of land cover types inside a post-fire study sites. The optimal feature combination for mapping land covers inside study sites were investigated.
Dynamic in land covers of study sites were analysed. Burnt area index for Sentinel-2 (BAIS2) was shown independent on date acquisition of satellite images to distinguish forest burn from other land covers over the analysed May–September vegetation period. Texture of study site improved the classification results. The most accurate classification method for identification of study sites land covers (with 0.84 Kappa coefficient and 0.86 overall accuracy) was based on combination of Sentinel-2 bands, BAIS2, and texture by Fourier transform. Analysis of vegetation recovery within the study sites demonstrated different recovery rates. Natural regeneration of pine was not observed, during three to six years of observations following fire events. The proposed method improves long-term post-fire environment assessment. Its findings can support planning of forest management measures needed to effectively restore forest cover.
5. Discrimination of conifer species with the use of Sentinel-2 time series and Google Earth Engine platform
Authors: Stefanos Papaiordanidis1, Georgopoulos Nikos, Alexandra Stefanidou1, Ioannis Z. Gitas1
1Laboratory of Forest Management and Remote Sensing, Aristotle University of Thessaloniki, Greece
Abstract: Forest species classification is crucial for the sustainable management of forest ecosystems in terms of resource planning. Remote sensing technology has been widely used both as an alternative and in conjunction with field measurements in vegetation species classification. In this study, Sentinel-2 multispectral imagery time series were used within the Google Earth Engine cloud-based environment, for fir (Abies borisii-regis) and spruce (Pinus nigra) discrimination in Pertouli University Forest, Greece. More specifically, field measurements provided 31 samples for each of spruce and fir homogenous stands, and Sentinel-2 multi-temporal images (2017-2020) was employed for the calculation of various spectral indices (e.g., NDVI, MCARI). Furthermore, several statistics (mean, standard deviation, median, maximum value) were calculated for each class, alongside with the difference between them, to use as thresholds for spruce and fir species discrimination. A visual comparison of the mean time series for each class revealed a period (mid- June) where fir and spruce differ consistently, especially when the MCARI index is considered. To evaluate those thresholds, 37 new points for each class were extracted by photointerpretation. Similar processing steps were executed (time series building, cloud mask application, outlier removal, spectral index calculation) and each of the statistics derived thresholds were applied. Finally, the accuracy of each statistic was calculated, and maximum index value outperformed mean, standard deviation, and median by achieving a 97.3% accuracy.
Keywords: Sentinel-2, forest species classification, time-series analysis, spectral indices, Google Earth Engine
6. Rapid assessment of post-fire forest regeneration in Coniferous ecosystems using the Google Earth Engine
Authors: Maria Prodromou1,2, Chris Danezis1,2, Ioannis Gitas3, Diofantos Hadjimitsis1,2
Corresponding Author: (firstname.lastname@example.org
1 Department of Civil Engineering and Geomatics, Cyprus University of Technology, Limassol, Cyprus
2 ERATOSTHENES Centre of Excellence, Limassol, Cyprus
3 Department of Forestry, Laboratory of Forest Management and Remote Sensing, Aristotle
University of Thessaloniki, Greece
Abstract: Forest ecosystems are among the most important natural resources on the planet and play a key role in the global carbon budget. Cyprus is in Eastern Mediterranean, which is an area where forest fires occur frequently, especially during the summer period. Using satellite data, we can derive information such as indicators related to vegetation conditions, detection of active fires, and assessment of burned areas. This study aims to investigate the post-fire forest regeneration for two of the biggest fire events in coniferous forests in Cyprus. Two of the biggest fire events, which occurred in the State Forests of Cyprus during the past years were selected as study areas. The selected fire events have specifically occurred in Moniatis (June 2007) and Solea (June 2016). In
order to assess the post-fire vegetation regeneration dynamics for the case studies the vegetation indices NDVI, EVI, and NBR were computed for time series of Landsat-7, Landsat-8, and Sentinel-2 imagery using Google Earth Engine. The results of this study show how the proposed methodology allows the rapid assessment of post-fire coniferous forest regeneration on a medium-high scale based on free access satellite data in Google Earth Engine. This is very useful for the systematic monitoring of wildfires on a national level. This work has been supported by the project ‘ERATOSTHENES: Excellence Research Centre for Earth Surveillance and Space-Based Monitoring of the Environment-EXCELSIOR’ (https://excelsior2020.eu/) that has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 857510 (Call: WIDESPREAD-01-2018-2019 Teaming Phase 2) and the Government of the Republic of Cyprus through the Directorate General for European Programmes, Coordination and Development.
Keywords: Earth observation, post-fire, forest regeneration, Google Earth Engine
7. Seasonal Changes in an Ecologically Important Wetland, Imbros Lagoon between 2016 and 2020: Evidence from High Resolution Satellite Imageries
Authors: Levent GENC and Melis INALPULAT
Study Group, Canakkale Onsekiz Mart University
Abstract: The Imbros Lagoon is known to be one of the tree coastal salty lagoons in Turkey. Due to its geographic location, vegetation and water related properties, the lagoon serves as a precise zone especially for migrative bird species, in different times of the year. Increasing of evaporation in summer season manipulates the status of this specific zone and as result of the process the salt became visible whereas it becomes a hotspot for tourism activities in dry season. However, environmental issues that threat the wetland are reported whereby the most highlighted one is the impacts of climate change which would affect not only water but also biodiversity in return. In this context the study aimed to investigate the alternations in surface water area of the lagoon in respect to different seasons between 2016 and 2020 using Sentinel 2 imageries. Different water related indices and their combinations were used to identify the most discriminative one. Findings revealed that there are obvious changes in water area of the lagoon in all seasons in coherency with meteorological data. The study is still ongoing for determination of relations between water level and number of individual birds and number of species considering multi-years, and monthly.
8. Detecting changes in vegetation and climate that serve as early warning signal on land degradation using remote sensing in the Eastern Mediterranean region: Review Findings & Next Steps
Authors: Filippos Eliades1,2, Diofantos Hadjimitsis1,2, Chris Danezis1,2 and Felix Bachofer3
1 Department of Civil Engineering and Geomatics, Cyprus University of Technology, Limassol, Cyprus
2 ERATOSTHENES Centre of Excellence, Limassol, Cyprus
3 German Aerospace Center (DLR)—Earth Observation Center (EOC), Wessling, Germany
Abstract: Desertification and land degradation have severe negative effects on land-use, water resources, soil stability, agriculture and biodiversity in the Mediterranean. Drylands cover 33.8% of northern Mediterranean countries: approximately 69% of Spain and 66% of Cyprus. The European Environment Agency (EEA) indicated that 8% of the territory of the European Union (mostly in Bulgaria, Cyprus, Greece, Italy, Romania, Spain and Portugal) experience a ‘very high’ or ‘high sensitivity’ to desertification. For Cyprus Island, 9.68% of the land area was found to be susceptible to land degradation.
The objective of this literature review is to provide a detailed synthesis of the main contributions of the global vegetation changes research to the development of environmental knowledge based on land degradation/ desertification and EO-based science and technology, identifying the current fields of research and possible research gaps. We selected screened more than 1000 scientific papers from which we reviewed approximately 300 papers, identifying the objectives and remote sensing data used to characterize vegetation changes.
Considering that the land degradation process is actually the degradation of vegetation and the decline in productivity, it is reasonable to use vegetation indicators such as NDVI, EVI: enhanced vegetation index, LAI: leaf area index etc or productivity parameters such as GPP: gross primary productivity, NPP etc to characterize the evolution of desertification.
Overall, a detailed EO-based time-series monitoring and analysis of un-altered natural vegetation could provide indicators that may serve as early warning signals for the scale and level of climate change induced effects on vegetation and ecosystems that might lead to land degradation and even to desertification.
Keywords: phenology, remote sensing, desertification, land degradation, early warning signal, Cyprus
9. Integrated, timely relevant application of Sentinel data fusion in agricultural sector
Authors: L. Ronczyk1, B. Keller2, G. Farkas1, Gy. Harka1, F. Collivinarelli3, F. Holecz3
1 University of Pécs
2 Dalmand Corporation
3 sarmap SA
Abstract: The ultimate objective of this research project was to develop and implement an operational and customized service and generate 42 dedicated products for four major crop types (winter wheat, winter barley, rapeseed, and corn) based on Sentinel-1/-2 time-series for Dalmand corporation. The Sentinel based Complex Agri-Industrial Application (SCAIA) - running on an Unix-like operating system including HPCs and cloud computing server infrastructures – is the joint effort of the University of Pécs, sarmap SA, and Dalmand Corporation supported by the European Space Agency (ESA) and the National Information Infrastructure Development Institute Supercomputing Infrastructure (NIIF HPC). The prototype service has proved its feasibility and made the first step to its practical implementation. To reach this goal we had to complete a very complex task.
10. Multidimensional model for monitoring status of aquatic environment
Authors: Gordana Jakovljević1 and Miro Govedaricas2
1Faculty of Architecture, Civil Engineering and Geodesy, University of Banja Luka, Bosnia and Herzegovina
2Faculty of Technical Science, University of Novi Sad, Serbia
Abstract: Aquatic ecosystems are among most sensitive ecological environments. The 2030 Agenda for Sustainable development emphasizes the water-related issues by setting SDG 6. The comprehensive and efficient monitoring of water quality and quantity need to be established in order to understand current status, polluters and to prevent feature degradation.
Remote sensing data and new cloud technologies enables development of new approach for management of water resources. Based on those facts the multidimensional model was proposed. The multidimensional model covering all phases from acquisition to distribution of data by providing clearly defined methodologies for automatic extraction of water body geometry, topology, and attributes. The model is based on AI technology allowing development of full automation processing procedures and usage of remote sensing data in near-real or real time. The implementation framework based on Google Colab, Python, and Jupyter enabled the development of a ready-to-use. The multidimensional model improves several aspects of monitoring results. It significantly increases the frequency of water body geometry and water quality monitoring. In addition, it provide monitoring of water quality spatial variations. The resulting information can be used for monitoring of process towards the achievement of SDG, including Indicator 6.3.2., Indicator 6.4.2., Indicator 6.6.1., and Indicator 14.1.1.
11. Intercomparison of Ground-based Radar Data for Precipitation Monitoring in the Area of Cyprus
Authors: Eleni Loulli1, Johannes Bühl2, Silas Michaelides1, Athanasios Loukas3, Diofantos G. Hadjimitsis1
1 Cyprus University Of Technology & ERATOSTHENES Centre Of Excellence, Cyprus,
2 Leibniz Institute For Tropospheric Research (TROPOS), Germany
3 Aristotle University Of Thessaloniki, Greece
Abstract: Drought, being a multidimensional phenomenon, starts imperceptibly, advances slowly and cumulatively, and its consequences show up gradually. Cyprus has an excellent location for studying meteorology and climatology. This study uses observations from NASA’s Global Precipitation Measurement (GPM) Mission to calibrate the data from the two ground-based radars of the Department of Meteorology (DoM). The DPR (Dual-frequency Precipitation Radar) aboard of GPM is applied in order to derive the reflectivity and the respective rain rate at the ground with a spatial resolution of 5-25km for 120km wide swath. The ground-based radars scan in PPI mode with the radar holding an elevation angle constant and varying its azimuth angle, and provide raw information with a spatial resolution of 0.1° and a radius of 150km. The two datasets are interpolated on a universal grid in order to enable the calibration of the raw data and their validation with the GPM data. The results will contribute to the development of an automated method for the estimation of the precipitation budget over the area of Cyprus and thus, drought monitoring in the region of the eastern Mediterranean.
The presented work is under the EXCELSIOR project that received funding from the European Union [H2020-WIDESPREAD-04-2017: Teaming Phase2] project under grant agreement no. 857510, and from the Republic of Cyprus.
Keywords: remote sensing, precipitation, radar, GPM, eastern Mediterranean
12. Status of built-up areas and artificial impervious surfaces of Bulgarian coastal municipalities based on GHSL and GAIA data for the last four decades
Authors: Assoc. Prof. Ph.D., Lachezar Filchev* and Assoc. Prof. Dr. Eng. Lyubka Pashova+
*Remote Sensing and GIS Department, Space Research and Technology Institute, Bulgarian Academy of Sciences (SRTI-BAS), Acad. Georgi Bonchev Str., Block 1, 1113 Sofia, Bulgaria;
+Department Geodesy; National Institute of Geophysics, Geodesy and Geography, Bulgarian Academy of Sciences (NIGGG-BAS), Acad. G. Bonchev Str., Bock 3, 1113 Sofia, Bulgaria
Abstract: The built-up areas in the Bulgarian coastal municipalities have grown significantly since the democratic changes in the country in the 1990s. The rapid pace of construction negatively affected the environment and led to imbalances in vulnerable coastal ecosystems. The main factors contributing to the increase of Artificial Impervious Surfaces (AIS) are urbanization, industrialization and tourism development. A brief overview of the dynamics of the impervious surface in the municipalities along the Bulgarian Black Sea coast in the period 1975-2018 was made. Data from the Global Artificial Impermeable Surface (GMIS) through the Google Earth Engine and GHSL platform (JRC-EC) and GAIA were used. The comparison results show a trend of increase and spread of AIS over the last four decades as the minimal was in the period 1990–2000 and were maximum in 2000–2006 for almost all 14 Bulgarian Black Sea municipalities. The general changes in the southern part of the Black Sea coastal zone are significantly more than those in the northern. The dynamics of changes in urban areas with a maximum area of changes are in Nessebar, Kavarna, Balchik and Sozopol. The last decade dynamic of the Urban Heat Islands (UHI) using Global Surface UHI Explorer (https://yceo.users.earthengine.app/view/uhimap) confirm the findings for Burgas and Varna metropolitan areas. Finally, recommendations have been made for the use of the result for national maritime spatial planning.
Keywords: Impervious surface, Built-up, UHI, GHSL, GAIA, Copernicus Program, Landsat, Bulgarian Black Sea coast
13. Imaging Spectroscopy Of Shallow Coastal Waters
Authors: Despina Makri1,2, Dimosthenis Traganos3, Athos Agapiou1,2, Diofantos Hadjimitsis1,2
1 Cyprus University of Technology, Cyprus
2 Eratosthenes Centre of Excellence, Limassol, Cyprus
3 German Aerospace Center, Helmholtz Association of German Research Centers (HZ) Cologne, Germany
Abstract: Imaging spectroscopy in shallow coastal waters is a newly developed technology advance that promises to fill the gap between the traditional remote sensing techniques and the land imaging sensors (Landsat-8 and Sentinel-2) with the airborne hyperspectral imaging systems. The monitoring of coastal waters was implemented with multispectral satellite data with medium or higher spatial resolution, providing a large coverage area. The extracted information gave good results for monospecific, continuous, and large seagrass meadows. With hyperspectral remote sensing, we can differentiate the submerged vegetation and coral reef taxa, enabling applications for fine tracking of biodiversity or the identification of resident or invasive species. Challenges and limitations in coastal water observations still exist, as those aquatic ecosystems are characterized by high complexity. Another critical issue is the atmospheric correction over the shallow coastal waters, as it is essential to implement different approaches compared to land or ocean applications. In the near future, image spectroscopy will be paired with cloud computing, artificial intelligence, and spaceborne optical and lidar data (e.g., Sentinel and Landsat series, and ICESat-2) to scale up to Earth Observation of shallow coastal waters at multiple levels (e.g., water quality, bathymetry, and seabed composition) across national to global scales.