Cheng Chien Liu

Processing of FORMOSAT-2 Daily Revisit Imagery for Site Surveillance

The successful operation of FORMOSAT-2, which was launched on May 21, 2004, proved the concept that the temporal resolution of a remote sensing system can be much improved by deploying a high spatial resolution sensor in a daily revisit orbit, and each accessible scene can be systematically observed from the same angle under similar illumination conditions. These characteristics make FORMOSAT-2 an ideal satellite for site surveillance. The unique orbit and the arrangement of the charge-coupled device lines onboard FORMOSAT-2, however, also raise new challenges in image processing. This paper describes a fast and automatic system that is able to process a large amount of FORMOSAT-2 daily revisit imagery for the purpose of site surveillance. The system is comprised of several modules, including level-2 product generation, band-to-band coregistration, a spectral preserved pan-sharpening technique, and multitemporal imagery matching. Two examples processed by the system are given to demonstrate the applicability of FORMOSAT-2 daily revisit imagery for site surveillance. The experiences of operating FORMOSAT-2 for more than one and a half years are summarized, and the advantages and disadvantages of a daily revisit orbit are discussed. Experience obtained from this paper would benefit the system design and image processing of future satellite missions with similar specifications, such as the Pleacuteiades HR scheduled to be launched in 2008

Image processing of FORMOSAT-2 data for monitoring the South Asia tsunami

We report on the actions of the first daily revisit satellite, FORMOSAT‐2, in the recent Indian Ocean tsunami disaster. Starting from the first images of Banda Aceh and Phuket taken on 28 December 2004, FORMOSAT‐2 used its unique orbit and pointable sensor system to demonstrate the extent to which it is able to respond to emergencies. A total of 137 images throughout the Indian Ocean rim countries were taken within a month. The data were immediately analysed and turned into damage‐assessment maps and other information resources for humanitarian aid. This paper focuses on the image‐processing procedure followed for a fast response to the South Asia tsunami event. The imageodesy technique is used to coregister the level‐2 product of FORMOSAT‐2 image at high accuracy and speed. A novel approach for spectral reservation data fusion has also been proposed. With the advantages of accurate coregistration and reliable spectral property, the colour composites of FORMOSAT‐2 imagery have been used as the principle source of information for our tsunami hazard assessment. The potential of FORMOSAT‐2 for disaster monitoring is discussed. The technique developed in this research will be adapted to produce pan sharpened images as a standard value added product of FORMOSAT‐2.

Cyclone-driven deep sea injection of freshwater and heat by hyperpycnal flow in the subtropics

China (973 Program) [2009CB421200]; Program of Introducing Talents of Discipline to Universities [B07034]; Academia Sinica Thematic Program AFOBi, Taiwan [NSC 98-2116-M-001-005]

A multi-sensor approach to examining the distribution of total suspended matter (TSM) in the Albemarle-Pamlico Estuarine System, NC, USA

For many coastal waters, total suspended matter (TSM) plays a major role in key biological, chemical and geological processes. Effective mapping and monitoring technologies for TSM are therefore needed to support research investigations and environmental assessment and management efforts. Although several investigators have demonstrated that TSM or suspended sediments can be successfully mapped using MODIS 250 m data for relatively large water bodies, MODIS 250 m data is of more limited use for smaller estuaries and bays or aquatic systems with complex shoreline geometry. To adequately examine TSM in the Albemarle-Pamlico Estuarine System (APES) of North Carolina, the large-scale synoptic view of MODIS and the higher spatial resolution of other sensors are required. MODIS, Landsat 7 ETM+ and FORMOSAT-2 remote sensing instrument (RSI) data were collected on 8 November, 24 November and 10 December, 2010. Using TSM images (mg/L) derived from MODIS 250 m band 1 (620–670 nm) data, Landsat 7 ETM+ 30 m band 3 (630–690 nm) and FORMOSAT-2 RSI 8 m band 3 (630−690 nm) atmospherically corrected images were calibrated to TSM for select areas of the APES. There was a significant linear relationship between both Landsat 7 ETM+ (r2 = 0.87, n = 599, P < 0.001) and FORMOSAT-2 RSI (r2 = 0.95, n = 583, P < 0.001) reflectance images and MODIS-derived TSM concentrations, thus providing consistent estimates of TSM at 250, 30 and 8 m pixel resolutions. This multi-sensor approach will support a broad range of investigations on the water quality of the APES and help guide sampling schemes of future field campaigns.

Fast and accurate model of underwater scalar irradiance

A spectral model of scalar irradiance with depth is applied to calculations of photosynthetically available radiation for a vertically homogeneous water column. The model runs more than 14,000 times faster than the full Hydrolight code, while it limits the percentage error to 2.20% and the maximum error to less than 4.78%. The distribution of incident sky radiance and the effects of a wind-roughened surface are integrated into this model. It can be applied to case 1 waters as well as to case 2 waters that happen to be gelbstoff rich, and the volume-scattering phase function can be generated dynamically based on the backscatter fraction. This new model is both fast and accurate and is, therefore, suitable for use interactively in models of the oceanic system, such as biogeochemical models or the heat budget part of global circulation models. It can also be applied by use of remote-sensing data to improve light-field calculations as a function of depth, which is needed for the estimation of global ocean carbon production and the ocean heat budget.

Automatic extraction of ground control regions and orthorectification of remote sensing imagery

We develop a fast and accurate method that is able to automatically select and match a large amount of ground control regions (GCRs) for orthorectifying remote sensing imagery. This new method is comprised of four modules, namely automatic extraction of GCRs, fast image-to-image matching, iterating and filtering of GCRs, and rigorous orthorectification. We assess the accuracy of this new method by processing the high-temporal- and high-spatial-resolution Formosat-2 imagery. Results show that the accurate orthoimage with a root mean square error of less than 1.5 pixels can be automatically generated from one standard Formosat-2 image (covering 12 km x 12 km) in 55 minutes. This new method has been incorporated into the Formosat-2 automatic image processing system and has been used to produce orthoimages on a daily-basis.

Integrating semianalytical and genetic algorithms to retrieve the constituents of water bodies from remote sensing of ocean color

This work presents a novel GA-SA approach to retrieve the constituents of water bodies from remote sensing of ocean color. This approach is validated and compared to the existing algorithms using the same synthetic and in-situ datasets compiled by the International Ocean Color Coordinate Group. Comparing to the other methods, the GA-SA approach provides better retrievals for both the inherent optical properties and various water constituents. This novel approach is successfully applied in processing the images taken by MODerate resolution Imaging Spectroradiometer (MODIS) and generates regional maps of chlorophyll-a concentration, total suspended matter, and the absorption coefficient of color dissolved organic matter at 443nm.

Surface heat balance analysis of Tainan City on March 6, 2001 using ASTER and Formosat-2 data

The urban heat island phenomenon occurs as a mixed result of anthropogenic heat discharge, decreased vegetation, and increased artificial impervious surfaces. To clarify the contribution of each factor to the urban heat island, it is necessary to evaluate the surface heat balance. Satellite remote sensing data of Tainan City, Taiwan, obtained from Terra ASTER and Formosat-2 were used to estimate surface heat balance in this study. ASTER data is suitable for analyzing heat balance because of the wide spectral range. We used Formosat-2 multispectral data to classify the land surface, which was used to interpolate some surface parameters for estimating heat fluxes. Because of the high spatial resolution of the Formosat-2 image, more roads, open spaces and small vegetation areas could be distinguished from buildings in urban areas; however, misclassifications of land cover in such areas using ASTER data would overestimate the sensible heat flux. On the other hand, the small vegetated areas detected from the Formosat-2 image slightly increased the estimation of latent heat flux. As a result, the storage heat flux derived from Formosat-2 is higher than that derived from ASTER data in most areas. From these results, we can conclude that the higher resolution land coverage map increases accuracy of the heat balance analysis. Storage heat flux occupies about 60 to 80% of the net radiation in most of the artificial surface areas in spite of their usages. Because of the homogeneity of the building roof materials, there is no contrast between the storage heat flux in business and residential areas. In sparsely vegetated urban areas, more heat is stored and latent heat is smaller than that in the forested suburbs. This result implies that density of vegetation has a significant influence in decreasing temperatures.

Optical model for use in oceanic ecosystem models

Modeling the plankton ecosystem requires a code for simulating the profile of irradiance from the chlorophyll profile at each time step of the integration. We have compared two existing codes with data from the Biogeochemical Ocean Flux Study: the Hydrolight radiative transfer model is accurate but too slow to use interactively in ecological models; Morels [J. Geophys. Res. 93, 10, 749 (1988)] empirical model is much faster but produces substantial error. We have developed a streamlined version of the Hydrolight radiative transfer model that is 20 times faster than the full Hydrolight code, while limiting errors to less than 12% within the euphotic zone. This new code is both fast and accurate and is, therefore, suitable for use interactively in oceanic ecosystem models.

Prediction of ocean colour: Monte Carlo simulation applied to a virtual ecosystem based on the Lagrangian Ensemble method

The annual variation of ocean-colour signals in the North Atlantic Ocean is successfully simulated by combining the plankton ecosystem model and the optical model. The first model uses the Lagrangian Ensemble method of Woods and Barkmann to simulate the upper-ocean ecosystem, including the vertical profile of chlorophyll concentration. The second model employs the Monte Carlo technique to compute the optical environment for that virtual ecosystem with 31 wavebands in the visible spectrum (400–700 nm). Ocean-colour signals are determined from the spectrum of the relatively few photons that are scattered back up through the sea surface. This information is then substituted into various satellite ocean-colour algorithms to calculate the satellite-derived surface chlorophyll concentration. Results show that substantial differences exist among the predictions from different ocean-colour algorithms. In addition, the average chlorophyll concentration of the upper few layers is not equal to the satellite-derived surface chlorophyll concentration. Our research encourages the adoption of the Monte Carlo optical model to simulate the satellite ocean-colour signals for the purpose of evaluating the plankton ecosystem model.