Lian Feng

Dynamic range and sensitivity requirements of satellite ocean color sensors: learning from the past

Sensor design and mission planning for satellite ocean color measurements requires careful consideration of the signal dynamic range and sensitivity (specifically here signal-to-noise ratio or SNR) so that small changes of ocean properties (e.g., surface chlorophyll-a concentrations or Chl) can be quantified while most measurements are not saturated. Past and current sensors used different signal levels, formats, and conventions to specify these critical parameters, making it difficult to make cross-sensor comparisons or to establish standards for future sensor design. The goal of this study is to quantify these parameters under uniform conditions for widely used past and current sensors in order to provide a reference for the design of future ocean color radiometers. Using measurements from the Moderate Resolution Imaging Spectroradiometer onboard the Aqua satellite (MODISA) under various solar zenith angles (SZAs), typical (L(typical)) and maximum (L(max)) at-sensor radiances from the visible to the shortwave IR were determined. The L(typical) values at an SZA of 45° were used as constraints to calculate SNRs of 10 multiband sensors at the same L(typical) radiance input and 2 hyperspectral sensors at a similar radiance input. The calculations were based on clear-water scenes with an objective method of selecting pixels with minimal cross-pixel variations to assure target homogeneity. Among the widely used ocean color sensors that have routine global coverage, MODISA ocean bands (1 km) showed 2-4 times higher SNRs than the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) (1 km) and comparable SNRs to the Medium Resolution Imaging Spectrometer (MERIS)-RR (reduced resolution, 1.2 km), leading to different levels of precision in the retrieved Chl data product. MERIS-FR (full resolution, 300 m) showed SNRs lower than MODISA and MERIS-RR with the gain in spatial resolution. SNRs of all MODISA ocean bands and SeaWiFS bands (except the SeaWiFS near-IR bands) exceeded those from prelaunch sensor specifications after adjusting the input radiance to L(typical). The tabulated L(typical), L(max), and SNRs of the various multiband and hyperspectral sensors under the same or similar radiance input provide references to compare sensor performance in product precision and to help design future missions such as the Geostationary Coastal and Air Pollution Events (GEO-CAPE) mission and the Pre-Aerosol-Clouds-Ecosystems (PACE) mission currently being planned by the U.S. National Aeronautics and Space Administration (NASA).

Natural and unnatural oil slicks in the Gulf of Mexico

Abstract When wind speeds are 2–10 m s−1, reflective contrasts in the ocean surface make oil slicks visible to synthetic aperture radar (SAR) under all sky conditions. Neural network analysis of satellite SAR images quantified the magnitude and distribution of surface oil in the Gulf of Mexico from persistent, natural seeps and from the Deepwater Horizon (DWH) discharge. This analysis identified 914 natural oil seep zones across the entire Gulf of Mexico in pre‐2010 data. Their ∼0.1 µm slicks covered an aggregated average of 775 km2. Assuming an average volume of 77.5 m3 over an 8–24 h lifespan per oil slick, the floating oil indicates a surface flux of 2.5–9.4 × 104 m3 yr−1. Oil from natural slicks was regionally concentrated: 68%, 25%, 7%, and <1% of the total was observed in the NW, SW, NE, and SE Gulf, respectively. This reflects differences in basin history and hydrocarbon generation. SAR images from 2010 showed that the 87 day DWH discharge produced a surface‐oil footprint fundamentally different from background seepage, with an average ocean area of 11,200 km2 (SD 5028) and a volume of 22,600 m3 (SD 5411). Peak magnitudes of oil were detected during equivalent, ∼14 day intervals around 23 May and 18 June, when wind speeds remained <5 m s−1. Over this interval, aggregated volume of floating oil decreased by 21%; area covered increased by 49% (p < 0.1), potentially altering its ecological impact. The most likely causes were increased applications of dispersant and surface burning operations.

Satellites Capture the Drought Severity Around China's Largest Freshwater Lake

A severe drought occurred in the region of Poyang Lake, the largest freshwater lake in China in early 2011, causing various environmental problems and posing a significant threat to local animals and humans. The event has caught extensive attention from many environmental groups as well as from national and international news media. Yet there has been no quantitative report on the severity of the drought except some qualitative description of water levels and local dry/wet conditions. Here we demonstrate how satellites can provide a quantitative measure of the severity of the large-scale drought, and how the current drought event may compare with historical conditions. The inclusion of the multi-source satellite data over a 12-year period also led to the conclusion that the drought was primarily caused by the significantly low precipitation. While there was a widespread rumor in China that the impoundment of the Three Gorges Dam (TGD) could lead to the severe drought, available data did not support such a hypothesis. Through this example, we hope to provide a perspective on the unique value of multi-source satellite data in quantifying environmental conditions in a changing climate.

Oil slick morphology derived from AVIRIS measurements of the Deepwater Horizon oil spill: Implications for spatial resolution requirements of remote sensors.

Using fine spatial resolution (~7.6m) hyperspectral AVIRIS data collected over the Deepwater Horizon oil spill in the Gulf of Mexico, we statistically estimated slick lengths, widths and length/width ratios to characterize oil slick morphology for different thickness classes. For all AVIRIS-detected oil slicks (N=52,100 continuous features) binned into four thickness classes (≤50 μm but thicker than sheen, 50-200 μm, 200-1000 μm, and >1000 μm), the median lengths, widths, and length/width ratios of these classes ranged between 22 and 38 m, 7-11 m, and 2.5-3.3, respectively. The AVIRIS data were further aggregated to 30-m (Landsat resolution) and 300-m (MERIS resolution) spatial bins to determine the fractional oil coverage in each bin. Overall, if 50% fractional pixel coverage were to be required to detect oil with thickness greater than sheen for most oil containing pixels, a 30-m resolution sensor would be needed.

Assessment of total suspended sediment concentrations in Poyang Lake using HJ-1A/1B CCD imagery

We explored the potential of the environment and disaster monitoring and forecasting small satellite constellations (HJ-1A/1B satellites) charge-coupled device (CCD) imagery (spatial resolution of 30 m, revisit time of 2 days) in the monitoring of total suspended sediment (TSS) concentrations in dynamic water bodies using Poyang Lake, the largest freshwater lake in China, as an example. Field surveys conducted during October 17–26, 2009 showed a wide range of TSS concentration (3–524 mg/L). Atmospheric correction was implemented using the Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes (FLAASH) module in ENVI with the aid of aerosol information retrieved from concurrent Terra/Moderate Resolution Imaging Spectroradiometer (MODIS) surveys, which worked well at the CCD bands with relatively high reflectance. A practical exponential retrieval algorithm was created between satellite remote sensing reflectance and in-situ measured TSS concentration. The retrieved results for the whole water area matched the in-situ data well at most stations. The retrieval errors may be related to the problem of scale matching and mixed pixel. In three selected subregions of Poyang Lake, the distribution trend of retrieved TSS was consistent with that of the field investigation. It was shown that HJ-1A/1B CCD imagery can be used to estimate TSS concentrations in Poyang Lake over synoptic scales after applying an appropriate atmospheric correction method and retrieval algorithm.

Satellite observations make it possible to estimate Poyang Lake?s water budget

Using moderate resolution imaging spectroradiometer (MODIS) satellite imagery with hydrologic and meteorological data, we developed a box model to estimate the water exchange between Poyang Lake (the largest freshwater lake of China) and the Changjiang (Yangtze) River from 2000 to 2009. Significant intra- and inter-annual variability of the water budget was found, with an annual mean outflow of Poyang Lake of 120:2 31:2 billion m 3 during 2000‐2009 and a declining trend of 5.7 billion m 3 yr 1 (pD 0:09). The impoundment of the Three Gorges Dam (TGD) on the Changjiang River in June 2003 led to a rapid lake‐river outflow of 760.6 million m 3 day 1 , resulting in a loss of 7864.5 million m 3 of water from the lake in a short period. Shortly thereafter, a statistically significant decrease in the drainage basin’s runoff coefficient was discovered. These findings provide large-scale evidence on how local precipitation and the TGD control the lake’s water budget, where continuous monitoring using the established approach and satellite data may provide critical information to help make water management decisions.

Variability of particulate organic carbon in inland waters observed from MODIS Aqua imagery

Surface concentrations of particulate organic carbon (POC) in shallow inland lakes were estimated using MODIS Aqua data. A power regression model of the direct empirical relationship between POC and the atmospherically Rayleigh-corrected MODIS product (Rrc,645-Rrc,1240)/(Rrc,859-Rrc,1240) was developed (R2?=?0.72, RMSE?=?35.86 ?gL?1, p?<?0.0001, N?=?47) and validated (RMSE?=?44.46 ?gL?1, N?=?16) with field data from 56 lakes in the Middle and Lower reaches of the Yangtze River, China. This algorithm was applied to an 11 year series of MODIS data to determine the spatial and temporal distribution of POC in a wide range of lakes with different trophic and optical properties. The results indicate that there is a general increase in minimum POC concentrations in lakes from middle to lower reaches of the Yangtze River. The temporal dynamics of springtime POC in smaller lakes were found to be influenced by local meteorological conditions, in particular precipitation and wind speed, while larger lakes were found to be more sensitive to air temperature.

Comparison of Valid Ocean Observations Between MODIS Terra and Aqua Over the Global Oceans

Ocean color satellite missions to measure the biophysical and geochemical properties of the surface ocean need to consider not only the spectral and spatial requirements of the sensors but also the satellite overpass time to maximize valid observations. The valid observations are impacted not only by cloud cover but also by other perturbations such as sun glint and stray light. Using Level-3 global composites of three ocean products (chlorophyll a or Chl-a, normalized florescence line height or nFLH, and sea surface temperature or SST), the daily percentage valid observations (DPVOs) over the global oceans were calculated, from which the differences between MODIS Aqua (afternoon pass) and MODIS Terra (morning pass) have been analyzed. For all three products, Aqua shows more valid observations than Terra over the Southern Ocean, the ocean near Peru and Chile, and the ocean around Angola and Namibia, with relatively >30% more valid observations in boreal winter months due to lower cloud coverage in the afternoon. In contrast, more than 20% of valid Chl-a and nFLH observations are obtained by Terra in the North Indian Ocean, and 10%-30% more valid observations by Terra are also found for the Equatorial Pacific and Atlantic oceans. These can be possibly linked to the lower presence of sun glint for Terra. Compared with Chl-a and nFLH, SST retrievals are more tolerant to sun glint and other perturbation factors, leading to much higher DPVOs. The implications of these findings to future satellite mission design and field campaigns are also discussed.

Evaluation of GOCI sensitivity for At-Sensor radiance and GDPS-Retrieved chlorophyll-a products

The signal-to-noise ratio (SNR, or sensitivity) of an ocean color instrument is a critical parameter to determine the accuracy and precision of the data products. Yet published literature showed various formats in SNR specifications under different conditions, making a direct cross-sensor comparison difficult. Here, we compared the SNRs of GOCI spectral bands with those of SeaWiFS and MODIS/Aqua under the same radiance inputs. We also compared their ability to resolve small changes in the retrieved chlorophyll-a data products (Chl). While GOCI visible bands showed similar at-sensor SNRs to SeaWiFS, the near-infrared (NIR) bands showed significantly higher SNRs. Because the NIR bands were used for atmospheric correction, the increases in SNRs led to reduced noise in the retrieved Chl, as shown in the GOCI and SeaWiFS Chl products for Chl < 0.1 mg m−3. The noise in the retrieved products also depends on the retrieval algorithms in addition to the sensor SNR. When a new band-subtraction algorithm (the Ocean Color Index or OCI algorithm) was applied to the same GOCI remotesensing reflectance data derived from the GDPS software package, significant noise reduction was found in the Chl product for low concentrations (< 0.25 mg m−3), leading to product precision (∼3% in Chl) comparable to those from MODIS/Aqua measurements. This is certainly a significant achievement, as GOCI spatial resolution is much higher than MODIS (500 m versus 1 km). In addition, artifacts across image mosaic edges over low-concentration waters have been removed nearly completely by the OCI algorithm. Data analyses also indicated that GOCI radiometric calibration requires further improvement.

Long-Term Distribution Patterns of Chlorophyll-a Concentration in China’s Largest Freshwater Lake: MERIS Full-Resolution Observations with a Practical Approach

Abstract: A new empirical Chl-a algorithm has been developed and validated for the largest freshwater lake of China (Poyang Lake) using a normalized green-red difference index (NGRDI), where the uncertainty was estimated to be ~25 mg·L −1 total suspended sediments or TSS) masked. The long-term Chl-a distribution showed significant spatial gradient and temporal variability, wi th Chl-a ranging between 2.4 ± 0.2 mg·m −3 in April and 4.4 ± 1.0 mg·m −3 in July and no significant increasing or decreasing trend during the 10-year period. In waters where Chl-a was retrievable (