• 10:00 Potentials of Bistatic Polarimetric SAR for Subsurface Imaging in Arid Regions

  Paillou, Philippe University of Bordeaux, France

  Show abstract

  The Argentinian space agency – CONAE – is developing a polarimetric L-band SAR platform for applications related to agriculture, hydrology and forestry: the SAOCOM mission (1). The selected Falcon-9 launcher has spare capacity and CONAE then offered to ESA the possibility to contribute to a passive system for mission enhancement. ESA is exploring the technical feasibility and the scientific value of a bistatic extension to SAOCOM, by means of a receive-only L-band companion satellite, the SAOCOM-TANGO instrument. TANGO is targeted to fly in formation with SAOCOM-1B, with flying geometries designed to allow the exploration of three acquisition modes: a tomographic mode, to build up a 3D view of the radar scattering surfaces, a bistatic mode, subdivided into two acquisitions geometries, and a specular mode, where the companion satellite views the illuminated scene from the opposite side (2). Science data will be acquired for boreal and tropical forest structure retrieval, subsurface mapping of ice and arid zones, monitoring of urban environments, and retrieval of soil moisture. Low frequency SAR has the capability to probe several meters of dry sediments, allowing to map buried geological features (faults, past river beds, paleo-lakes) in arid regions (3). The access to such information can be particularly useful to assess fossil water resources in desert areas. Based on the Integral Equation Model approach proposed by Fung et al. (4), we developed a two-layers bistatic and polarimetric scattering model, taking into account several terms: a surface scattering term of a superficial layer (aeolian sediments), a subsurface scattering term of a buried layer (bedrock), and a volume scattering term in the superficial layer. This model is particularly well suited to the study of the potentials of bistatic SAR for subsurface imaging in arid regions. We were able to simulate bistatic scattering patterns for typical two-layers scenarii, and to explore the effects of surface parameters, such as roughness and dielectric constant, and acquisition parameters, such as polar and azimuth angles (5). We derived some optimal bistatic and specular acquisition parameters, which allow to minimize the contribution of the superficial layer, and then enhance the scattering of the subsurface layer. This will contribute to the definition of bistatic and specular acquisitions for the SAOCOM-TANGO system, in order to better image hidden subsurface geological features in desert environments. (1) D’Aria D., Giudici A., Monti-Guarnieri P., Rizzoli J., “A wide swath, full polarimetric, L-band spaceborne SAR”, Proc. of 8th Radar Conference, Rome, Italy, 2008. (2) Davidson M., Gebert N., Dominguez B.C., Fois F., Silvestrin P., “SAOCOM-CS: A passive companion to SAOCOM for single pass L-band bistatic interferometry and tomography”, IGARSS'14, Quebec, Canada, 2014. (3) Paillou Ph., S. Lopez, T. Farr, A. Rosenqvist, “Mapping Subsurface Geology in Sahara using L-band SAR: First Results from the ALOS/PALSAR Imaging Radar”, IEEE J. of Selected Topics in Earth Observations and Remote Sensing, vol. 3, no 4, pp. 632-636, 2010. (4) Fung. A.K, Liu W.Y., Chen K.S., Tsay M.K., “An improved IEM model for bistatic scattering from rough surface”, J. of Electromagn. Waves and Appl., vol. 16, no. 5, pp. 689-702, 2002. (5) Johnson J.T, Ouellette J.D., “Polarizations features in bistatic scattering from rough surfaces”, IEEE Trans. On Geoscience and Remote Sensing, vol. 52, no. 3, pp. 1616-1626, 2014.

  
   

• 09:40 The performance of Sentinel-1 for mapping built-up areas

  Voormansik, Kaupo (1); Zalite, Karlis (1); Praks, Jaan (2) 1: Tartu Observatory, Estonia; 2: Aalto University, Finland

  Show abstract

  This study assesses Sentinel-1 performance for built-up area detection. The algorithms, tested for built-up area detection, are based on utilization of speckle divergence local area statistics and interferometric coherence. The classification accuracies and final built-up area map spatial resolutions are analysed and discussed for different imaging modes (stripmap and interferometric wide swath) of the instrument. The significance of using different polarisations and their combinations (HH/HV, VV/VH) is assessed and discussed. The work is based on the original idea to apply Sentinel-1 SAR imagery and anonymous mobile positioning data for urban growth monitoring, presented for the Copernicus Masters DLR Environmental Challenge 2013, where it won the first price in Earth Observation applications development. According to United Nations analysis global urban population is expected to grow from 3.6 billion in 2011 to 6.3 billion in 2050 [1]. Moreover 83% of the governments have expressed their concern about the population geographical distribution in their country [1]. The increase of urban population means also an increase in built-up areas. SAR as an instrument sensitive to object’s structure is inherently sensitive to buildings which contain a lot of dihedral and trihedral shape scatterers. Still a very limited number of studies have been carried out about built-up area retrieval with SAR [2] [3] [4]. Thanks to the Copernicus programme free and open data policy and high revisit (global land surface coverage in every 12 days) Sentinel-1 will be a very important SAR data supplier for operational applications and also for the international research community for the next decade. In this study Sentinel-1 dual pol (HH/HV and VV/VH) stripmap (5 m by 5 m spatial res.) and interferometric wide swath (20 m by 5 m res.) mode imagery is used. Based on preliminary analysis interferometric wide swath mode allowed 85% overall classification accuracy with 0.5 ha resolution whereas stripmap mode 83% overall classification accuracy with 0.1 ha resolution in the Estonian test-site. Works Cited [1] United Nations, Department of Economic and Social Affairs, Population Division, „World Urbanization Prospects, the 2011 Revision. Final Report with Annex Tables.,“ New York, 2011. [2] T. Esch, M. Thiel, A. Schenk, A. Roth, A. Muller ja S. Dech, „Delineation of Urban Footprints From TerraSAR-X Data by Analyzing Speckle Characteristics and Intensity Information,“ IEEE Transactions on Geoscience and Remote Sensing, kd. 48, nr 2, pp. 905-916, 2010. [3] H. Taubenbock, T. Esch, A. Felbier, A. Roth ja S. Dech, „Pattern-Based Accuracy Assessment of an Urban Footprint Classification Using TerraSAR-X Data,“ IEEE Geoscience and Remote Sensing Letters, kd. 8, nr 2, pp. 278-282, 2011. [4] T. Esch, H. Taubenböck, A. Roth, W. Heldens, A. Felbier, M. Thiel, M. Schmidt, A. Müller ja S. Dech, „TanDEM-X mission—new perspectives for the inventory and monitoring of global settlement patterns,“ Journal of Applied Remote Sensing, kd. 6, nr 1, 2012.

  
   

• 10:20 Four-Step OVoG Inversion for the Retrieval of Agricultural Crop Structure by Means of Multi-Baseline Polarimetric SAR Interferometry

  Pichierri, Manuele (1); Hajnsek, Irena (1,2) 1: ETH Zurich, Switzerland; 2: Microwaves and Radar Institute, DLR Oberpfaffenhofen, Germany

  Show abstract

  ABSTRACT Nowadays, world is experiencing a dramatic increase in population and the economy of some of the most populated countries is still strongly agriculture-based. An improved management of agricultural resources is therefore required and Pol-InSAR [1], through the combination of polarimetric information with interferometric observations, presents a valuable support to accomplish the task of monitoring the vegetation height and the crop canopy structure. To ensure a quantitative analysis of the Pol-InSAR complex coherences, the electromagnetic scattering from an agricultural crop is modeled via the Oriented Volume over Ground model (OVoG) [2], a 2-layer structure encompassing a cloud of spheroidal particles with a preferred orientation and an underlying impenetrable ground layer. As a result, the propagation through the canopy becomes polarisation-dependent (i.e. non-zero differential extinction and refractivity between volume eigenpolarisations) due to the anisotropic effects introduced by the scatterers’ shape and orientation. In this work, for the first time an OVoG inversion scheme (4-step inversion, first presented in [3]) has been developed and tested on a dataset of multi-frequency (L-, C- and X-Band), multi-temporal and multi-baseline airborne DLR’s F-SAR acquisitions over a rural area in Germany, in 2014. The use of a multi-baseline extended observation space (the OVoG inversion problem is undetermined if only one baseline is employed), allows one to potentially retrieve the whole set of the OVoG structural parameters (e.g. crop height, differential extinction between eigenpolarisations and ground-to-volume ratios) without any a priori information (e.g. the knowledge of the topography) or further assumptions (e.g. pure volume coherence, or nearly-zero differential extinction [4][5]). Unlike black-box optimizations [6], each intermediate step of the proposed methodology is not hidden and its partial outcomes are easily accessible, so that a quantitative assessment of the overall performance is evidently enhanced. The availability of a rich dataset of SAR acquisitions throughout the whole plant growth cycle as well as the diversity of crop types represent an outstanding opportunity for the early assessment of the potential of multi-baseline Pol-InSAR for the retrieval of the agricultural crop structure. REFERENCES [1] S.R. Cloude and K.P. Papathanassiou: Polarimetric SAR Interferometry, IEEE Transactions on Geoscience and Remote Sensing, 36(5):1551-1565, 1998. [2] R. N. Treuhaft and S.R. Cloude: The structure of oriented vegetation from Polarimetric Interferometry, IEEE Transactions on Geoscience and Remote Sensing, 37(5 II):2620-2624, 1999. [3] M. Pichierri and I. Hajnsek: First results towards the retrieval of agricultural crop structure by means of Polarimetric SAR Interferometry, Proceedings of EUSAR, Berlin, 2014. [4] S.R. Cloude and K.P. Papathanassiou: Three-stage inversion process for polarimetric SAR interferometry. IEE Proceedings - Radar, Sonar and Navigation, 150(3):125 – 134, 2003. [5] J.M. Lopez-Sanchez, I. Hajnsek, and J.D. Ballester-Berman: First Demonstration of Agriculture Height Retrieval with PolInSAR Airborne Data, IEEE Geoscience and Remote Sensing Letters, 9(2): 242-246, 2012. [6] J.M. Lopez-Sanchez, J.D. Ballester-Berman, and Y. Marquez-Moreno: Model limitations and parameter-estimation methods for agricultural applications of polarimetric SAR interferometry, IEEE Transactions on Geoscience and Remote Sensing, 45(11):3481-3493, 2007.

  
   

• 09:00 The role of polarimetry in a dynamical approach for agricultural crop monitoring

  Vicente-Guijalba, Fernando; Lopez-Sanchez, Juan M.; Martinez-Marin, Tomas University of Alicante, Spain

  Show abstract

  Recently a new methodology has been proposed for agricultural monitoring exploiting time series of SAR images [1, 2]. This new methodology relies on a dynamic perspective that provides a suitable context for temporal analysis. The approach is aimed to extract and employ the temporal information, contained in the scene itself, in order to build an evolution model for the crop and then to combine the model with new observations for estimation purposes. In particular, an application of this approach consists in estimating the phenological stage of a crop by combining both the dynamics of the process and the observed information provided by polarimetric SAR images. The approach is defined in a very general form basically consisting in two main stages. The first one consists of a modeling or learning algorithm which will define the nominal behavior of the growing process. This model is obtained employing a reference set of parcels for which the phenological information is previously known. The model is used afterwards in the second stage to provide an estimation each time a new acquisition is supplied. The estimation is defined employing an adapted Extended Kalman Filter. Current and future satellite SAR systems will provide in a near future numerous polarimetric and temporal configurations. For instance Sentinel-1 products consist in dual polarisation (HH+VH or VV+HV) with a 12-day repeat cycle. Moreover, this sampling rate will decrease to only 6 days with the scheduled two-satellite constellation. In the case of the future Radarsat Constellation Mission (RCM) a compact polarimetric product will be delivered in a very low sampling rate. This diverse scenario forces us to evaluate how the different possibilities will affect the proposed dynamic approach. In this work we perform an analysis focused on evaluating the actual role of the polarimetry and what is the impact of considering alternative configurations. The flexibility of the methodology allows us to define the problem dimension, i.e. the state space, using different sets of polarimetric features. This characteristic can be exploited directly to evaluate how the estimation methodology is affected by different configurations. The evaluation is carried out from the application point of view, evaluating the results' quality through a confusion matrix analysis employing a set of fields with different cereal crops. The dataset gathered over different crop types (cereals, field pea and canola) during the ESA funded AgriSAR2009 campaign is employed in the study. This consists of 24 images acquired by Radarsat-2 in Fine-Quad polarimetric mode. References: [1] F. Vicente-Guijalba, T. Martinez-Marin, and J. M. Lopez-Sanchez, “Crop phenology estimation using a multitemporal model and a Kalman filtering strategy,” IEEE Geoscience and Remote Sensing Letters, vol. 11, no. 6, pp. 1081–1085, June 2014. [2] F. Vicente-Guijalba, J. M. Lopez-Sanchez, and T. Martinez-Marin, "A dynamical system approach for agricultural crop phenology tracking using PolSAR time-series: cereals case study at C-band," Proceedings of 10th European Conference on Synthetic Aperture Radar (EUSAR 2014), Berlin, Germany, 3-5 June 2014.

  
   

• 09:20 POLARIMETRIC L-BAND ALOS FOR ACCURATE ASSESSMENT OF FIRE DAMAGES IN BOREAL PEATLANDS

  Touzi, Ridha (1); Omari, Khalid (1); Sleep, Bob (2) 1: Canada Centre for Remote Sensing, Canada; 2: Alberta Environment and Sustainable Resource Development

  Show abstract

  Fire is an important process in boreal and sub-arctic ecosystems and will continue to influence the structure and dynamics of these peatland dominated systems. As a result of climate change, it is predicted that by 2050 there will be a 50% increase in forest fire activity in northern Canada and that the fire season will be longer with increased ignitions (both lighting and human caused) and an overall larger area burned [9]. Accurate and up to date detailed mapping of annual burn coverage will provide important insights into trends in fires and an understanding of the relative frequency of burns within different habitat types. Cost effective wetland characterization and monitoring are now possible due to advances in the technology of earth observation satellites that provide the possibility of efficient monitoring of wetland status over large and remote areas. Recently, the visible near-infared (VNIR) satellites (and Landsat in particular) have become the most popular source of information for wetland mapping. In Canada, Landsat 7 data combined with Radarsat1 HH-polarization SAR data have been used as the primary source of information for building the Canadian Wetland Inventory [1]. The use of the additional all-weather single polarization (HH) Radarsat-1 information permits better delineation and monitoring of wetland water extent [2]. Unfortunately, the combination with VNIR optical and SAR satellite information cannot clearly discriminate bogs from poor (shrub and sedge) fens, and treed bogs may be confused with upland forests [1, 2]. As such there is an immediate need for new tools that permits a clear identification of bogs, and their discrimination from poor fens and upland forests. This will permit an efficient monitoring of bog to fen transformations, related to climate change or anthropogenic activities, and accurate assessment of fire damages in peatland areas. Recently, it has been shown that earth observation satellites equipped with L-band polarimetric SAR could provide the required information for cost effective peatland mapping and monitoring in the boreal and subarctic peatlands [4, 5]. The unique capability of polarimetric SAR for the detection of peatland subsurface water flow, first demonstrated in [3], could permit an enhanced discrimination of bog from fen, and this should result in better monitoring of bog-fen transformations related to the various (natural and anthropogenic) sources of stress. This was demonstrated using polarimetric L-band ALOS collected over a boreal peatland, La Baie des Mines, and subarctic peatlands in the Wapusk National Park in Canada [4, 5]. The Touzi decomposition [6, 7] that describes the scattering type in terms of magnitude and phase, provides the required information for separation of bogs from poor fens using the long penetrating capabilities of the L-band polarimetric ALOS data The present study is conducted, jointly with ESRD, using ALOS data collected over boreal petalands in the Athabasca oil sand exploration region (near Fort McMurray, Alberta). An improved and updated wetland inventory in Alberta’s northern boreal forest would support advances in wildfire management planning and strategic wildfire suppression. A series of polarimetric ALOS acquisitions collected over peatlands are analyzed with reference to field data collected, and to a Wetland Inventory generated many years ago using aerial photos and many field sampling. The sensitivity of the scattering type phase [6, 7] to peatland subsurface water flow is confirmed using the L-band ALOS data. This permits an enhanced discrimination of bog from poor fens, and accurate separation of treed bog from upland forests. The added value of polarimetric L-band SAR information to Optics (Landsat TM) sensors for assessment of fire damages over forests and peatlands is well demonstrated. While optics sensors can identify well burned forest and peatlands areas, accurate assessment of peat health in burned treed bogs can only be obtained using the peat subsurface information provided by the long penetrating L-band polarimetric ALOS. These very promising results are obtained using an optimized version of the Touzi decomposition, which interprets the incoherent target scattering decomposition in terms of nine independent target parameters. The added value of scattered wave rotation independent parameter (degree of polarization, and total scattered intensity) optimization [8] is also assessed with reference to the information provided by the incoherent target scattering decomposition. [1] Fournier R., M. Grenier, A. Lavoie, and R. Helie, “Towards a strategy to implement the Canadian wetland inventory using satellite remote sensing,” Can. J. Remote Sens., vol. 33, pp. S1–S16, Nov. 2007. [2] Grenier M., A.M. Demers, S. Labrecque, M. Benoit, R. Fournier, and B. Drolet, “An object-based method to map wetland using RADARSAT-1 and Landsat ETM images: test case on two sites in Quebec, Canada” Can. J. Remote Sens., vol. 33, pp. S28–S45, Nov. 2007. [3] Touzi R., A. Deschamps and G. Rother, “Wetland characterization using polarimetric Radarsat-2 capability”, Can. J. Rem. Sens., Vol. 33, No. 1:S56–S67, 2007. [4] R. Touzi, G. Gosselin, and R. Brooks, “Investigation of L-band ALOS and C-band Radarsat2 for peatland subsurface water flow monitoring”, IGARSS’12, Munich, Germany, July 2012 [5] R. Touzi, G. Gosselin, and and R. Brooks, “Peatland subsurface water flow monitoring using polarimetric L-band”, ESA Book on POLSAR_Applications, in press [6] Touzi, R., “Target Scattering Decomposition in Terms of Roll-Invariant Target Parameters ”, IEEE Trans. Geosci. Remote Sens., Vol. 45, No. 1, Jan. 2007 [7] Touzi, A. Deschamps and G. Rother, “Phase of target scattering for wetland characterization using polarimetric C-band SAR”, IEEE TGRS, Vol. 47, No. 9, pp. 3241-3261, Sep. 2009 [8] Touzi R., S. Goze, T. Le Toan, A. Lopes, and E. Mougin , “Polarimetric discriminators for SAR images” , IEEE Trans. Geoscience Rem. Sens., Vol.30, No. 5, pp 973 980, Sep. 1992 [9] R.K. Brook, “Forest and tundra fires in the Hudson Bay Lowlands of Manitoba”. IN: Climate Change: Linking Traditional and Scientific Knowledge (eds R. Riewe & J. Oakes). p. 257-266. Aboriginal Issues

  
   

Highlights

In May 2013, the Biomass mission concept was selected to become the next in the series of satellites developed to further our understanding of Earth.

Released: 21/06/2013

REPLAY EARTH EXPLORER 7 USER CONSULTATION MEETING (PART 1)

Released: 11/03/2013

POLinSAR 2013 Workshop

Released: 22/02/2013