Science Service System
You are here : Home : Proposals_Summary

Summary of Proposal HYD0354

TitleEvaluation of TerraSAR for high resolution flood, landuse and infrastructure mapping in the Mekong River Delta, Vietnam
Investigator Künzer, Claudia - DFD DLR, LA
Team Member
Dr. Renaud, Fabrice - United, EHS
Prof. Dr. Gerke, Solvay - ZEF - Center for Development Research, University of Bonn, ZEF
Dr. Clemens, Joachim - INRES - Plant Nutrition, University of Bonn, INRES
Dr. Apel, Heiko - GeoForschungsZentrum Potsdam, GFZ
Dr. Künzer, Claudia - University of Würzburg, Remote sensing
Dr. Heege, Thomas - EOMAP GmbH & Co.KG, Eomap
Nguyen, Dao - GIRS-VAST, GIS and Remoste Sensing Research Center
Vo Khac, Tri - SIWRR, Southern Institute of Water Resources Research
Dr. Precht, Elimar - DHI-WASY, Niederlassung Syke
Ehlers, Manfred - University of Osnabrück, Institut für Geoinformatik und Fernerkundung
Tang, Ping - Chinese Academy of Sciences, Institute of Remote Sensing Applications

Within the WISDOM project "Water related Information System for the Mekong River Delta" it is envisaged to set up an online information system (IS) for the German and Vietnamese project partners and last but not least for Vietnamese end-users such as planning institutions (also ministries) in Vietnam.

Objectives for TerraSAR analyses are i) the derivation of hydrology related products such as water masks and moisture products, ii) the support of landcover classificatione specially for crop type discrimination and rice mapping, iii) the derivation of infracstructure (roads, dams, and especially small settlements (most houses with tin roofs). Products shall be compared and validated with optical data and data collected during field surveys.

From a mehtodologic point of view, three different algorithms will be compared for water mask generation - these algorithms are already available in house at DFD and are a) a threshold based IDL routine for water extraction from radar data, b) a segement based Matlab algorithm and c) detailed studies with ecognition. The characteristics of TerraSAR for flood mask derivation shall be compared not only based on these algorithms but also with respect to other radar sensors (ASAR, Palsar) and optical data (Spot, Quickbird). In the field of landuse discrimination the data shall be integrated into joint optical-radar fusion analyses and especially support rice mapping (see extension below).

Data requirements are as following:

Monthly coverage of study areas Can Tho and Tam Nong (as specified under study areas) during the rainy and dry season in strip map mode (3m), and of inividual hot spots to be defined in these areas in 1m spotlight mode (e.g. mainly for setteled areas). Wet season roughly July to December, dry season: January to June. Data take should be synchronized with other data takes in the project (Palsar, ASAR, optical, e.g. Quickbird) and with planned field campaigns. Furthermore full coverage ScanSar mode 16m data for the two mentioned study areas during the dry and wet season during one period (e.g. starting with wet season in summer 2008) . Furthermore, it would be crucial for hydrological modeling to cover one study area in dense time steps during one flood season (e.g. every overpass, for a period of 3 months). It is expected that overall below 100 scenes over the course of the next 2.5 years would be acquired.

Deliverables will include water masks and wetness products for the dry and the rainy season, landcover and infrastructure information, detailed quantitative studies on spatial data accuracy, radiometric data stability, influence of the differen polarizations and comparisons with other sensors.

Proposal-Extension: RICEMAN as subproject of WISDOM project

Objectives for TerraSAR analyses: The objectives of the study were to understand the relationship between radar backscatter coefficients and selected parameters (e.g. plant age and biomass) of rice crops over the Mekong Delta, and to develop a rice crop inventory system using time-series Envisat ASAR and TerraSAR-X imagery.

From a mehtodologic point of view: This research examines the methods for rice identification and mapping in the study area by using ASAR APP and TerraSAR-X datasets. Based on the discovered relationships between rice parameters and radar backscattering, various methods will be applied.

Data requirements are as following: Additionally to unmodified requirements of WISDOM data acquisitions dual polarized TSX stripmap mode data over 2 areas are required to derive repition rate of every area of 22 days. Repeated acquisition from January 2010 on until end of 2011, to achieve acquisition of several rice growing cycles.

: Extension of HYD_0354 proposal with 100 additional scenes for 2010-2013 (end of expected 2nd WISDOM project phase).

Back to list of proposals

© DLR 2004-2016