Analysis of classification accuracy for pre-filtered multichannel remote sensing data

Noise is one of the main factors degrading the quality of original multichannel remote sensing data and its presence influences classification efficiency, object detection, etc. Thus, pre-filtering is often used to remove noise and improve the solving of final tasks of multichannel remote sensing. Recent studies indicate that a classical model of additive noise is not adequate enough for images formed by modern multichannel sensors operating in visible and infrared bands. However, this fact is often ignored by researchers designing noise removal methods and algorithms. Because of this, we focus on the classification of multichannel remote sensing images in the case of signal-dependent noise present in component images. Three approaches to filtering of multichannel images for the considered noise model are analysed, all based on discrete cosine transform in blocks. The study is carried out not only in terms of conventional efficiency metrics used in filtering (MSE) but also in terms of multichannel data classification accuracy (probability of correct classification, confusion matrix). The proposed classification system combines the pre-processing stage where a DCT-based filter processes the blocks of the multichannel remote sensing image and the classification stage. Two modern classifiers are employed, radial basis function neural network and support vector machines. Simulations are carried out for three-channel image of Landsat TM sensor. Different cases of learning are considered: using noise-free samples of the test multichannel image, the noisy multichannel image and the pre-filtered one. It is shown that the use of the pre-filtered image for training produces better classification in comparison to the case of learning for the noisy image. It is demonstrated that the best results for both groups of quantitative criteria are provided if a proposed 3D discrete cosine transform filter equipped by variance stabilizing transform is applied. The classification results obtained for data pre-filtered in different ways are in agreement for both considered classifiers. Comparison of classifier performance is carried out as well. The radial basis neural network classifier is less sensitive to noise in original images, but after pre-filtering the performance of both classifiers is approximately the same.

An intercomparison of bio-optical techniques for detecting dominant phytoplankton size class from satellite remote sensing

Satellite remote sensing of ocean colour is the only method currently available for synoptically measuring wide-area properties of ocean ecosystems, such as phytoplankton chlorophyll biomass. Recently, a variety of bio-optical and ecological methods have been established that use satellite data to identify and differentiate between either phytoplankton functional types (PFTs) or phytoplankton size classes (PSCs). In this study, several of these techniques were evaluated against in situ observations to determine their ability to detect dominant phytoplankton size classes (micro-, nano- and picoplankton). The techniques are applied to a 10-year ocean-colour data series from the SeaWiFS satellite sensor and compared with in situ data (6504 samples) from a variety of locations in the global ocean. Results show that spectral-response, ecological and abundance-based approaches can all perform with similar accuracy. Detection of microplankton and picoplankton were generally better than detection of nanoplankton. Abundance-based approaches were shown to provide better spatial retrieval of PSCs. Individual model performance varied according to PSC, input satellite data sources and in situ validation data types. Uncertainty in the comparison procedure and data sources was considered. Improved availability of in situ observations would aid ongoing research in this field.

Macroscale factors affecting diatom abundance: a synergistic use of Continuous Plankton Recorder and satellite remote sensing data

Diatoms exist in almost every aquatic regime; they are responsible for 20% of global carbon fixation and 25% of global primary production, and are regarded as a key food for copepods, which are subsequently consumed by larger predators such as fish and marine mammals. A decreasing abundance and a vulnerability to climatic change in the North Atlantic Ocean have been reported in the literature. In the present work, a data matrix composed of concurrent satellite remote sensing and Continuous Plankton Recorder (CPR) in situ measurements was collated for the same spatial and temporal coverage in the Northeast Atlantic. Artificial neural networks (ANNs) were applied to recognize and learn the complex non-monotonic and non-linear relationships between diatom abundance and spatiotemporal environmental factors. Because of their ability to mimic non-linear systems, ANNs proved far more effective in modelling the diatom distribution in the marine ecosystem. The results of this study reveal that diatoms have a regular seasonal cycle, with their abundance most strongly influenced by sea surface temperature (SST) and light intensity. The models indicate that extreme positive SSTs decrease diatom abundances regardless of other climatic conditions. These results provide information on the ecology of diatoms that may advance our understanding of the potential response of diatoms to climatic change.

Remote sensing of assimilation number for marine phytoplankton

Estimating primary production at large spatial scales is key to our understanding of the global carbon cycle. Algorithms to estimate primary production are well established and have been used in many studies with success. One of the key parameters in these algorithms is the chlorophyll-normalised production rate under light saturation (referred to as the light saturation parameter or the assimilation number). It is known to depend on temperature, light history and nutrient conditions, but assigning a magnitude to it at particular space-time points is difficult. In this paper, we explore two models to estimate the assimilation number at the global scale from remotely-sensed data that combine methods to estimate the carbon-to-chlorophyll ratio and the maximum growth rate of phytoplankton. The inputs to the algorithms are the surface concentration of chlorophyll, seasurface temperature, photosynthetically-active radiation af the surface of the sea, sea surface nutrient concentration and mixed-layer depth. A large database of in situ estimates of the assimilation number is used to develop the models and provide elements of validation. The comparisons with in situ observations are promising and global maps of assimilation number are produced.

Volcanic ash supply to the surface ocean – remote sensing of biological responses and their wider biogeochemical significance

Transient micronutrient enrichment of the surface ocean can enhance phytoplankton growth rates and alter microbial community structure with an ensuing spectrum of biogeochemical feedbacks. Strong phytoplankton responses to micronutrients supplied by volcanic ash have been reported recently. Here we: (i) synthesize findings from these recent studies; (ii) report the results of a new remote sensing study of ash fertilization; and (iii) calculate theoretical bounds of ash-fertilized carbon export. Our synthesis highlights that phytoplankton responses to ash do not always simply mimic that of iron amendment; the exact mechanisms for this are likely biogeochemically important but are not yet well understood. Inherent optical properties of ash-loaded seawater suggest rhyolitic ash biases routine satellite chlorophyll-a estimation upwards by more than an order of magnitude for waters with <0.1 mg chlorophyll-a m-3, and less than a factor of 2 for systems with >0.5 mg chlorophyll-a m-3. For this reason post-ash-deposition chlorophyll-a changes in oligotrophic waters detected via standard Case 1 (open ocean) algorithms should be interpreted with caution. Remote sensing analysis of historic events with a bias less than a factor of 2 provided limited stand-alone evidence for ash-fertilization. Confounding factors were poor coverage, incoherent ash dispersal, and ambiguity ascribing biomass changes to ash supply over other potential drivers. Using current estimates of iron release and carbon export efficiencies, uncertainty bounds of ash-fertilized carbon export for 3 events are presented. Patagonian iron supply to the Southern Ocean from volcanic eruptions is less than that of windblown dust on thousand year timescales but can dominate supply at shorter timescales. Reducing uncertainties in remote sensing of phytoplankton response and nutrient release from ash are avenues for enabling assessment of the oceanic response to large-scale transient nutrient enrichment.

Uncertainty estimates of remote sensing reflectance derived from comparison of ocean color satellite data sets

Assigning uncertainty to ocean-color satellite products is a requirement to allow informed use of these data. Here, uncertainty estimates are derived using the comparison on a 12th-degree grid of coincident daily records of the remote-sensing reflectance RRS obtained with the same processing chain from three satellite missions, MERIS, MODIS and SeaWiFS. The approach is spatially resolved and produces σ, the part of the RRS uncertainty budget associated with random effects. The global average of σ decreases with wavelength from approximately 0.7– 0.9 10−3 sr−1 at 412 nm to 0.05–0.1 10−3 sr−1 at the red band, with uncertainties on σ evaluated as 20–30% between 412 and 555 nm, and 30–40% at 670 nm. The distribution of σ shows a restricted spatial variability and small variations with season, which makes the multi-annual global distribution of σ an estimate applicable to all retrievals of the considered missions. The comparison of σ with other uncertainty estimates derived from field data or with the support of algorithms provides a consistent picture. When translated in relative terms, and assuming a relatively low bias, the distribution of σ suggests that the objective of a 5% uncertainty is fulfilled between 412 and 490 nm for oligotrophic waters (chlorophyll-a concentration below 0.1 mg m−3). This study also provides comparison statistics. Spectrally, the mean absolute relative difference between RRS from different missions shows a characteristic U-shape with both ends at blue and red wavelengths inversely related to the amplitude of RRS. On average and for the considered data sets, SeaWiFS RRS tend to be slightly higher than MODIS RRS, which in turn appear higher than MERIS RRS. Biases between mission-specific RRS may exhibit a seasonal dependence, particularly in the subtropical belt.

Space-based lidar measurements of global ocean carbon stocks

Global ocean phytoplankton biomass (C-phyto) and total particulate organic carbon (POC) stocks have largely been characterized from space using passive ocean color measurements. A space-based light detection and ranging (lidar) system can provide valuable complementary observations for C-phyto and POC assessments, with benefits including day-night sampling, observations through absorbing aerosols and thin cloud layers, and capabilities for vertical profiling through the water column. Here we use measurements from the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) to quantify global C-phyto and POC from retrievals of subsurface particulate backscatter coefficients (b(bp)). CALIOP b(bp) data compare favorably with airborne, ship-based, and passive ocean data and yield global average mixed-layer standing stocks of 0.44 Pg C for C-phyto and 1.9 Pg for POC. CALIOP-based C-phyto and POC data exhibit global distributions and seasonal variations consistent with ocean plankton ecology. Our findings support the use of spaceborne lidar measurements for advancing understanding of global plankton systems.

Epipelagic mesozooplankton distribution and abundance over the Mascarene Plateau and Basin, south-western Indian Ocean

The crescent shaped Mascarene Plateau (southwestern Indian Ocean), some 2200 km in length, forms a partial barrier to the (predominantly westward) flow of the South Equatorial Current. Shallow areas of the Mascarene Plateau effectively form a large shelf sea without an associated coastline. Zooplankton sampling transects were made across the plateau and also the basin to the west, to investigate the role the partial interruption of flow has on zooplankton biomass and community structure over the region. Biomass data from Optical Plankton Counter (OPC) analysis, and variability in community structure from taxonomic analysis, appear to indicate that the obstruction by the plateau causes upwelling, nutrient enrichment and enhanced chlorophyll and secondary production levels downstream. The Mascarene Basin is clearly distinguishable from the ridge itself, and from the waters to the south and north, both in terms of size-distributed zooplankton biomass and community structure. Satellite remote sensing data, particularly remotely-sensed ocean colour imagery and the sea surface height anomaly (SSHA), indicate support for this hypothesis. A correlation was found between OPC biovolume and SSHA and sea surface temperature (SST), which may indicate the physical processes driving mesozooplankton variability in this area. Biomass values away from the influence of the ridge averaged 24 mg m-3, but downstream if the ridge biomass averaged 263 mg m-3. Copepods comprised 60% of the mean total organisms. Calanoid copepods varied considerably between regions, being lowest away from the influence of the plateau, where higher numbers of the cyclopoid copepods Oithona spp., Corycaeus spp. and Oncaea spp., and the harpacticoid Microsetella spp. were found.

Use of Continuous Plankton Recorder information in support of marine management: applications in fisheries, environmental protection, and in the study of ecosystem response to environmental change

The Continuous Plankton Recorder (CPR) survey was conceived from the outset as a programme of applied research designed to assist the fishing industry. Its survival and continuing vigour after 70 years is a testament to its utility, which has been achieved in spite of great changes in our understanding of the marine environment and in our concerns over how to manage it. The CPR has been superseded in several respects by other technologies, such as acoustics and remote sensing, but it continues to provide unrivalled seasonal and geographic information about a wide range of zooplankton and phytoplankton taxa. The value of this coverage increases with time and provides the basis for placing recent observations into the context of long-term, large-scale variability and thus suggesting what the causes are likely to be. Information from the CPR is used extensively in judging environmental impacts and producing quality status reports (QSR); it has shown the distributions of fish stocks, which had not previously been exploited; it has pointed to the extent of ungrazed phytoplankton production in the North Atlantic, which was a vital element in establishing the importance of carbon sequestration by phytoplankton. The CPR continues to be the principal source of large-scale, long-term information about the plankton ecosystem of the North Atlantic. It has recently provided extensive information about the biodiversity of the plankton and about the distribution of introduced species. It serves as a valuable example for the design of future monitoring of the marine environment and it has been essential to the design and implementation of most North Atlantic plankton research.