TURBIDITY MEASUREMENTS TO OPTIMIZE MEASUREMENTS OF SUSPENDED PARTICULATE MATTER CONCENTRATION 

Get Complete Project Material File(s) Now! »

COMBINED MEASUREMENTS OF ABSORPTION AND ATTENUATION

The development of multi-wavelength and hyperspectral combination a and c meters in the mid 1990s (Moore et al. 1992; Pegau et al. 1995) led to significant advancements in the measurement of IOPs. The most widely used instruments are the WET Labs ac-9 (9 wavelengths) and ac-s (hyperspectral), where a() is determined based on the reflective tube principle (Zaneveld et al. 1990; Kirk 1992). Instrument design is similar to transmissometer design, except that the configuration is optimized to collect as much scattered light as possible, allowing measurement of0()-a(). The optical path is surrounded by a quartz reflective tube, reflecting forward scattered light back into the sampling path. For natural particle suspensions the majority of light is scattered forward, but a small amount of light will be lost by backscattering or imperfect reflection at the walls, and needs to be accounted for (Sullivan et al. 2006) in order not to overestimate a(). Additional corrections for the dependency of a() on temperature and salinity are needed (Pegau et al. 1997). Measurement uncertainties of optical properties using the ac-9 instrument have been reported by Leymarie et al. 2010.
The simultaneous c() and a() measurements are blanked to ultrapure Milli-Q water so that the combined attenuation and absorption due to dissolved and particulate matter are measured directly. The particulate scattering coefficient, bp(), can be derived from these by difference.

VOLUME SCATTERING FUNCTION AND BACKSCATTERING COEFFICIENTS

Measurement of the VSF over the full angular range (between 0° and 180°) involves making a large number of measurements at discrete angles, each covering a large dynamic range especially in the near-forward angles. Instruments satisfying these complex requirements have been developed (Petzold 1972; Lee and Lewis 2003; Sullivan and Twardowski 2009; Twardowski et al. 2012), but none are commercially available at present. Since the late 1990s, bb() has been routinely determined from one or a couple of measurements of the VSF in the backward direction. Measurement of() at a single angle m>90° has been found to be significantly correlated to bb() (Oishi 1990; Boss and Pegau 2001Oishi1990; Maffione and Dana 1997) via a proportionality factor: bb () 2 (m ,) (1.36).
Errors are minimized to a few percent atm between 117° en 120° (Oishi 1990; Boss and Pegau 2001) whereis independent of the contribution ofw() to() (Boss and Pegau 2001). Removal of the variable contribution ofw() to() prior to conversion leads to more accurate results, especially in clear waters (Boss and Pegau 2001), hence: bb () 2 p (m ,)w (m ,) bbw () (1.37).
wherep is the proportionality factor betweenp(m) and bbp() defined analogously toin Eq. (1.36). Values ofw(m) and bbw() for pure seawater from Zhang et al. (2009) are thought to be accurate within 2%. Empirically determinedp factors as a function ofm are reported by Sullivan and Twardowski 2009 and have lowest uncertainty of maximally 2% in the range 110°-120°. The WET Labs BB-9 and the HOBI Labs Hydroscat instruments are widely used multi-wavelength backscattering instruments based on a single angle measurement of(). In practice, the measured(m) is the result of a weighted portion of() in the vicinity ofm, described by an instrument geometry dependent angular weighting function W():  (m ,) ( ,)W ( )d (1.38).

Turbidity measurements

Turbidity, T, defined by ISO 1999 as ‘the reduction of transparency of a liquid caused by the presence of undissolved matter’, can be quantified in various ways (e.g., Secchi disk, light attenuation, side scatter). The Hach 2100P portable turbidity instrument measures the ratio of LED light scattered at an angle of 90°±2.5° at a wavelength of 860 nm ±60 nm to forward transmitted light, as compared to the same ratio for a standard suspension of Formazine. This optical measurement technique of T from the side scattering coefficient is in accordance with ISO 1999. T is expressed in Formazine Nephelometric Units (FNU) and instruments are calibrated using a set of Formazin Turbidity Standards. At the start of each sea campaign, instrument stability is ensured by recording turbidity of Hach STABLCAL Formanzin standards of 0.1, 20, 100, and 800 FNU and instrument recalibration is done if necessary. Side scattering signals are averaged over 10 measurements at 1.2 second intervals. Glass sample cells of 10 mL are used to record seawater T. The glass cell is rinsed with sampled seawater before filling. The exterior of the sample cell is rinsed with milliQ water, dried with paper tissue, swiped with a soft microfiber lint-free cloth treated with silicon oil, and swiped again with a dry cloth. Prior to turbidity measurement, the sample cell is visually inspected for dust particles or air bubbles. T was recorded in triplicate, gently tumbling the sample cell between each measurement.
T is recorded before and after [SPM] filtrations to check adequate mixing of the water sample during subsampling for filtration. Each T measurement typically takes about four minutes to complete and portable turbidity meters can be purchased for less than 1000 US dollars.

T AS PROXY FOR [SPM]

T and [SPM] measurements were carried out in surface waters in coastal and offshore waters around Europe and French Guyana between 2007 and 2010. Sampling sites are described in Neukermans et al. 2012. A ‘least squares cubic’ type II regression (York 1966) is applied to the log transformed T and [SPM] data. The least squares cubic regression, which takes into account measurement uncertainties, is applied after removal of outliers identified by the MATLAB robustfit.m routine. Correlation coefficients are given with their 95% confidence intervals, obtained from bootstrapping. Details of these statistical procedures are described in section 3.5.1.3 on page 82. Based on the [SPM]-T regression, an estimate of [SPM] can be derived from measurements of T prior to filtration. From this estimate of [SPM], the volume of seawater to be filtered so that an optimal mass is retained by the filter can then be estimated as described in section 2.2.3.

DETERMINING OPTIMAL FILTRATION VOLUME

The filtration volume, V, should be high enough so that the dry mass of the particles retained by the filter, , is sufficient to be precisely measured, but not so much that the filter clogs. Its estimation requires a quantification of minimum measurement uncertainty on , assessed from procedural control filters, and a maximum value for the relative uncertainty on [SPM]. The approach is described below.
For measurements of weight, the detection limit (DL) of the balance gives the minimum measurement uncertainty. The minimum uncertainty on the difference between filter weights before and after filtration, , is then given by (ISO 1995):wbal (2.1).

Effect of filtration volume on precision of [SPM] measurements

To investigate the effect of filtration volume on the precision of [SPM] measurements, six experiments were carried out in the southern North Sea in September 2009 and 2011 for clear (T < 5 FNU), moderately turbid (5 FNU<T<20 FNU), and turbid waters (T>20 FNU). For each water sample listed in Table 2.2, [SPM] measurements were performed with filtration volumes of 0.2, 0.5, 1, and 2 times Vopt and every filtration was done on five replicate filters. T was continuously monitored during the course of each filtration experiment, to ensure good mixing of the sampled seawater. Table 2.2 lists the sampling time and location, salinity, temperature, depth, [Chl a], and the mean and standard deviation of T for each sample.

Between group statistical analysis

Between group differences are investigated by one-way analysis of variance (ANOVA), comparing the means of several groups to test the hypothesis that they are all the same, against the alternative that they are not all the same. To test which pairs of means are significantly different, paired-sample t-tests were done at the 5% significance level. Analyses were carried out using the statistics toolbox of MATLAB, version R2011b.
Graphical illustrations of the distribution of observations are done with boxplots. The edges of the box indicate the 25th and 75th percentiles, while the red line represents the sample median. The length of the box is called the interquartile range (IQR). Observations further than 1.5 IQR from the 25th and 75th percentiles are marked as outliers and indicated by red crosses. This range corresponds to ±2.7 standard deviations and 99.3% data coverage if normally distributed. The whiskers extend to minimum and maximum observations that are not marked as outliers.

READ  Evaluating the Resilience of Browser Fingerprinting to Block Adversarial Crawlers

SALT RETENTION TESTS

Results from lab experiments with SSW show significantly higher residual weight (p=0.003, F=11.76, d.f.=17, ANOVA) when the rim is not rinsed (see Figure 2.2), compared to when the filter rim is rinsed. This is in accordance with previous works that stressed the importance of rinsing of the filter rim to flush out diffused salts (Strickland and Parsons 1968; van der Linde 1998). The volume of SSW filtered was not found to affect SSW blank residual weights (p=0.72, F=0.57, ANOVA, see Figure 2.3), which were not significantly different from zero. This suggests that salts are washed out using a wash volume of 400-450 mL of milliQ, independent of the volume of SSW filtered. This is partially in accordance with Stavn et al. 2009 who found salt retention of 1.1 mg, independent of volume of seawater filtered for a salinity of 34 PSU and a wash volume of 300 mL (see their Table 1).

Table of contents :

CHAPTER 1 OPTICAL PROPERTIES AND SEAWATER CONSTITUENTS: DEFINITIONS AND MEASUREMENT 
1.1 SEAWATER CONSTITUENTS
1.2 OPTICAL PROPERTIES
1.2.1 INHERENT OPTICAL PRPERTIES (IOPS)
1.2.1.1 Definition of IOPs
1.2.1.2 Partitioning of IOPs
1.2.1.3 Concentration-specific IOPs
1.2.2 APPARENT OPTICAL PROPERTIES (AOPS)
1.2.2.1 Reflectances
1.2.2.2 Diffuse attenuation coefficients
1.2.3 LINK BETWEEN IOPS AND AOPS
1.3 MEASURING OPTICAL PROPERTIES
1.3.1 IOPS
1.3.1.1 Beam attenuation
1.3.1.2 Combined measurements of absorption and attenuation
1.3.1.3 Volume scattering function and backscattering coefficients
1.3.1.4 Turbidity
1.3.2 AOPS
1.3.2.1 Above-water radiometry
CHAPTER 2 TURBIDITY MEASUREMENTS TO OPTIMIZE MEASUREMENTS OF SUSPENDED PARTICULATE MATTER CONCENTRATION 
ABSTRACT
2.1 INTRODUCTION
2.2 MATERIALS AND METHODS
2.2.1 MEASUREMENT OF [SPM]
2.2.1.1 Measurement protocol
2.2.1.2 Filter blanks
2.2.1.3 Salt retention tests
2.2.2 TURBIDITY MEASUREMENTS
2.2.3 OPTIMAL FILTRATION VOLUME
2.2.3.1 T as proxy for [SPM]
2.2.3.2 Determining optimal filtration volume
2.2.4 EFFECT OF FILTRATION VOLUME ON PRECISION OF [SPM] MEASUREMENTS
2.2.5 BETWEEN GROUP STATISTICAL ANALYSIS
2.3 RESULTS AND DISCUSSION
2.3.1 UNCERTAINTIES IN [SPM] MEASUREMENT
2.3.1.1 Salt retention tests
2.3.1.2 Filter blanks
2.3.1.3 Sample mixing
2.3.2 UNCERTAINTIES IN [SPM] MEASUREMENT FROM FILTRATION VOLUME
2.3.2.1 Determining optimal filtration volume
2.3.2.2 Effect of filtration volume on precision of [SPM] measurement
2.4 CONCLUSION
CHAPTER 3 IN SITU VARIABILITY OF MASS-SPECIFIC BEAM ATTENUATION AND BACKSCATTERING OF MARINE PARTICLES WITH RESPECT TO PARTICLE SIZE, DENSITY, AND COMPOSITION 
ABSTRACT
3.1 INTRODUCTION
3.2 METHODS
3.2.1 DESCRIPTION OF THE STUDY AREAS
3.2.2 OPTICAL MEASUREMENTS
3.2.3 WATER SAMPLING
3.2.4 PARTICLE SIZE AND MEAN APPARENT DENSITY
3.2.5 MASS-SPECIFIC ATTENUATION AND BACKSCATTERING COEFFICIENTS
3.3 RESULTS
3.3.1 INSTRUMENT INTERCOMPARISON FOR PARTICULATE BEAM ATTENUATION
3.3.2 RELATIONSHIPS BETWEEN SCATTERING PROPERTIES AND PARTICLE CONCENTRATION BY AREA OR DRY WEIGHT
3.3.3 INVESTIGATION OF THE VARIABILITY OF MASS-SPECIFIC SCATTERING PROPERTIES
3.3.4 INVESTIGATION OF THE VARIABILITY OF ATTENUATION AND BACKSCATTERING EFFICIENCY
3.3.5 SCATTERING MEASUREMENTS AS PROXIES FOR [SPM]
3.4 DISCUSSION AND CONCLUSION
3.5 APPENDICES
3.5.1 DEALING WITH MEASUREMENT UNCERTAINTIES: PROPAGATION OF UNCERTAINTIES AND LINEAR REGRESSION METHODS
3.5.1.1 Repeat measurements of IOPs and particle concentration
3.5.1.2 Propagation of uncertainty
3.5.1.3 Least squares regression for measurements with uncertainties
3.5.1.4 Correlation coefficient uncertainty estimation
3.5.2 IOPS AND PARTICLE SIZE DISTRIBUTIONS
3.5.2.1 Scattering properties vs. area concentration by size class
3.5.2.2 Particle size distributions
3.5.2.3 Relationship between Junge PSD slope and cp spectral slope from in situ measurements
CHAPTER 4 ATMOSPHERIC CORRECTION OF SEVIRI IMAGERY 
ABSTRACT
4.1 INTRODUCTION
4.2 MATERIALS AND METHODS
4.2.1 THE SEVIRI RADIOMETER
4.2.2 STUDY AREA: SEVIRI SUBSCENE AND VIEWING ANGLES
4.2.3 DEFINITIONS AND GENERAL APPROACH
4.2.4 SEVIRI SOLAR CHANNELS CALIBRATION
4.2.5 ATMOSPHERIC CORRECTION OF SEVIRI VIS06
4.2.5.1 Rayleigh and gas corrections
4.2.5.2 Aerosol correction
4.2.5.3 Atmospheric correction assumptions
4.2.5.4 Solving for marine and aerosol reflectances
4.2.5.5 Atmospheric correction processing steps
4.2.6 ATMOSPHERIC CORRECTION USING THE HRV BAND
4.2.6.1 Relating marine reflectances in the HRV and VIS06 bands
4.2.6.2 Estimating spatial variability of 0(0.6) w  on the HRV grid from HRV spatial anomaly
4.2.7 ESTIMATE OF UNCERTAINTY ON MARINE REFLECTANCE
4.2.7.1 Uncertainties associated with the atmospheric correction assumptions
4.2.7.2 Digitization uncertainties
4.2.7.3 Combined atmospheric-digitization uncertainty
4.2.7.4 Uncertainty estimate of marine reflectance in the VIS06 band on the HRV grid
4.2.8 CROSS-VALIDATION WITH MODIS REFLECTANCE DATA
4.3 RESULTS
4.3.1 MAPPING MARINE REFLECTANCE AND ITS UNCERTAINTIES
4.3.2 MARINE REFLECTANCE ON THE HRV GRID
4.3.3 SEVIRI – MODIS CROSS VALIDATION
4.3.3.1 VIS06 reflectance
4.3.3.2 VIS06 reflectances on the HRV grid
4.3.4 DIURNAL VARIABILITY OF MARINE REFLECTANCES AND AEROSOLS
4.4 DISCUSSION AND CONCLUSION
4.4.1 OPTIMAL REFLECTANCE RANGE FOR ATMOSPHERIC CORRECTION
4.4.2 SPATIAL RESOLUTION
4.4.3 EXPLOITING HIGH TEMPORAL RESOLUTION IMAGERY
4.4.4 EXPANDING TO SEVIRI FULL DISK IMAGERY
4.5 APPENDICES
4.5.1.1 Estimation of uncertainty on 0(0.6) w  using band combination (VIS06, NIR16)
4.5.1.2 Band pair selection based on minimization of  0(0.6) w 
4.5.2 INTER-CALIBRATION OF VIS06 AND VIS08 BANDS USING RAYLEIGH AND GAS-CORRECTED REFLECTANCES
CHAPTER 5 DIURNAL VARIABILITY OF TURBIDITY AND LIGHT ATTENUATION IN TURBID SOUTHERN NORTH SEA WATERS FROM THE SEVIRI GEOSTATIONARY SENSOR 
5.1 INTRODUCTION
5.2 MATERIALS AND METHODS
5.2.1 RETRIEVAL OF TURBIDITY AND SUSPENDED MATTER FROM SEVIRI
5.2.1.1 Retrieval Algorithms
5.2.1.2 Uncertainty on SEVIRI turbidity and suspended matter products
5.2.2 RETRIEVAL OF PAR ATTENUATION
5.2.2.1 PAR attenuation algorithm
5.2.2.2 Uncertainty on PAR attenuation retrieval
5.2.3 VALIDATION OF SEVIRI TURBIDITY AND PAR ATTENUATION PODUCTS
5.2.3.1 SmartBuoy measurements of turbidity and PAR attenuation
5.2.3.2 Comparison of instantaneous SEVIRI-SmartBuoy observations
5.2.3.3 Validation of SEVIRI product time series
5.3 RESULTS
5.3.1 COMPARISON OF INSTANTANEOUS SEVIRI-SMARTBUOY OBSERVATIONS
5.3.1.1 Turbidity observations
5.3.1.2 PAR attenuation observations
5.3.2 VALIDATION OF SEVIRI PRODUCT TIME SERIES
5.3.2.1 Time series of turbidity
5.3.2.2 Times series of PAR attenuation
5.4 DISCUSSION
5.4.1 SEVIRI ATMOSPHERIC CORRECTION AND DIGITIZATION UNCERTAINTIES
5.4.2 LIMITATIONS OF SEVIRI SPATIAL RESOLUTION
5.4.2.1 sub-pixel scale variability
5.4.2.2 Use of the HRV band
5.4.3 LIMITATIONS OF THE T AND KD RETRIEVAL ALGORITHMS
5.4.3.1 Turbidity retrieval algorithm
5.4.3.2 KPAR retrieval algorithm
5.4.4 INCOMMENSURABILITY BETWEEN IN-SITU AND REMOTELY SENSED PRODUCTS
5.4.4.1 remotely sensed vs. in-situ T
5.4.4.2 Remotely sensed vs. in-situ KPAR
5.4.5 PERSPECTIVES FOR THE DESIGN OF FUTURE GEOSTATIONARY SENSORS AND SYNERGY WITH POLAR-ORBITING SENSORS
5.5 CONCLUSION
5.6 APPENDICES
5.6.1 CALIBRATION DATASET FOR TURBIDITY AND [SPM] RETRIEVAL ALGORITHMS
5.6.1.1 Data collection, treatment, and selection
5.6.1.2 Impact of [SPM] measurement improvement on the [SPM] retrieval algorithm
5.6.2 VARIABILITY OF TURBIDITY-SPECIFIC BACKSCATTERING
5.6.3 UNCERTAINTIES OF THE T AND [SPM] RETRIEVAL ALGORITHMS
5.6.3.1 Variability of the mass- and turbidity-specific backscattering coefficient
5.6.3.2 regional calibration
5.6.4 COUPLING BETWEEN TURBIDITY AND HYDRODYNAMICS
CHAPTER 6 GENERAL CONCLUSIONS AND PERSPECTIVES
6.1 IN-SITU VARIABILITY OF SCATTERING PROPERTIES AND SUSPENDED MATTER CONCENTRATION
6.2 REMOTE SENSING OF DIURNAL VARIABILITY OF SUSPENDED PARTICLES FROM THE
GEOSTATIONARY SEVIRI METEOROLOGICAL SENSOR
REFERENCES

GET THE COMPLETE PROJECT

Related Posts