東華大學圖書館 |

Remote Sensing of Suspended Sediment in San Francisco Bay Using Satellite and Drone Imagery.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Remote Sensing of Suspended Sediment in San Francisco Bay Using Satellite and Drone Imagery./
作者:	Adelson, Joseph Henry.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
面頁冊數:	218 p.
附註:	Source: Dissertations Abstracts International, Volume: 82-02, Section: B.
Contained By:	Dissertations Abstracts International82-02B.
標題:	Remote sensing. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28104005
ISBN:	9798662511071

Remote Sensing of Suspended Sediment in San Francisco Bay Using Satellite and Drone Imagery.
Adelson, Joseph Henry.

Remote Sensing of Suspended Sediment in San Francisco Bay Using Satellite and Drone Imagery. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 218 p.

Source: Dissertations Abstracts International, Volume: 82-02, Section: B.

Thesis (Ph.D.)--Stanford University, 2020.

This item must not be sold to any third party vendors.

Suspended sediment in San Francisco Bay affects the economic and ecological health of the estuary and its surrounding region by limiting light availability for photosynthesis, transporting contaminants, nourishing marsh restoration projects, infilling shipping channels, and providing protection to the shoreline from sea level rise via accretion on mudflats. Traditional efforts to study sediment transport phenomena have relied upon in situ measurements and numerical modeling, but these approaches have limitations. In situ measurement techniques rely on point measurements with high temporal resolution, yet they are difficult to deploy over large spatial areas. Models provide useful insight into the spatial heterogeneity of sediment processes. However, they rely on initial and boundary conditions and parameterizations that are based on observations, therefore the accuracy of models is also constrained in part by the limitations of in situ measurements. This dissertation presents remote sensing measurements from satellites and unmanned aerial vehicles (UAVs) to understand suspended sediment transport processes in estuaries like San Francisco Bay. Twelve methods for inferring suspended sediment concentration (SSC) from Landsat 7 imagery were compared using k-folds validation and assessed based on their abilities to recreate in situ SSC measurements from one meter below the surface. The best performer was the model of Nechad et al. (2010) using the red wavelength band with coefficients determined via Huber regression, with mean absolute error of 5.94 mg L-1 and bias of 0.15 mg L-1. Satellite-derived SSC observations compare well with USGS transects indicating that the method is well-suited to supplement cruise data that is costly to acquire and therefore limited in its frequency. Remote sensing measurements were aggregated by location, season, or tidal phase to understand the variability of SSC and to compare probability densities with in situ measurements. These results show that surface SSC is heightened in the shoals during summer months and has trended downward in Suisun and Grizzly Bays since 1999. Using satellite imagery from 2014-2017, remotely sensed surface SSC derived from the Nechad method was paired with bottom stress estimates based on two-dimensional hydrodynamic and fetch-limited wave models to investigate the relationship between surface SSC and flow. Observations of SSC closely fit a lognormal distribution though the shape, characterized by the modal value, depend on binning criteria including embayment, depth, and wave height. When binned by model-derived bottom shear stress, the modal value of the SSC distribution exhibited an inflection point at the critical shear stress for erosion. This suggests that remote sensing can be used to derive critical stresses that are otherwise difficult to measure. To account for the limitations of satellite imagery such as low spatial resolution and low temporal resolution (Landsat 7 overpasses occurred roughly once every 16 days), a method was developed to infer surface SSC from UAV-based imagery. While traditional remote sensing platforms take imagery at approximately a nadir viewing angle and provide multispectral images that are aligned with one another, an off-the-shelf camera aboard a UAV may not adhere to those qualities. Low cost multi-spectral cameras often include individual sensors for each band. The slight misalignment between images violates assumptions in two-band glint correction algorithms. Additionally, UAVs must tilt to fly and compensate for wind requiring images to occasionally be taken at angles more oblique than most satellite imagery. The method developed in this dissertation adapts previous techniques for sun glint correction for misaligned multispectral images and offers a novel approach to reduce the effects of camera orientation for oblique angles. During a field campaign, the UAV-based method to capture remote sensing reflectance was validated via comparison with in situ measurements made with a hyperspectral radiometer, and its ability to accurately infer SSC was verified based on in situ water samples. It was found that a polarizing filter is necessary to mitigate much of the glare on the water surface. A series of test flights were conducted to measure the surface SSC along a transect parallel to the Dumbarton Bridge during different phases of the tidal cycle. To reduce the impact of variability of incoming light, the flights were conducted over a period of 12 days at the same solar zenith angle during each day. Because the tide arrives later by roughly 50 minutes each day, consecutive daily transects over 12 days provided the variability over a tidal cycle. Cross-sectional sediment flux was computed from the remotely sensed surface SSC measurements and compared well to flux values estimated from in situ USGS observations.

ISBN: 9798662511071Subjects--Topical Terms:

535394
Remote sensing.
Subjects--Index Terms:

Estuaries

Remote Sensing of Suspended Sediment in San Francisco Bay Using Satellite and Drone Imagery.
LDR:06091nmm a2200361 4500 001 2281870
005 20210927083416.5
008 220723s2020 ||||||||||||||||| ||eng d
020 $a 9798662511071
035 $a (MiAaPQ)AAI28104005
035 $a (MiAaPQ)STANFORDzd976km4704
035 $a AAI28104005
040 $a MiAaPQ $c MiAaPQ
100 1 $a Adelson, Joseph Henry. $3 3560580
245 1 0 $a Remote Sensing of Suspended Sediment in San Francisco Bay Using Satellite and Drone Imagery.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 218 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-02, Section: B.
500 $a Advisor: Fringer, Oliver;Arrigo, Kevin;Kitanidis, Peter;Monismith, Stephen.
502 $a Thesis (Ph.D.)--Stanford University, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a Suspended sediment in San Francisco Bay affects the economic and ecological health of the estuary and its surrounding region by limiting light availability for photosynthesis, transporting contaminants, nourishing marsh restoration projects, infilling shipping channels, and providing protection to the shoreline from sea level rise via accretion on mudflats. Traditional efforts to study sediment transport phenomena have relied upon in situ measurements and numerical modeling, but these approaches have limitations. In situ measurement techniques rely on point measurements with high temporal resolution, yet they are difficult to deploy over large spatial areas. Models provide useful insight into the spatial heterogeneity of sediment processes. However, they rely on initial and boundary conditions and parameterizations that are based on observations, therefore the accuracy of models is also constrained in part by the limitations of in situ measurements. This dissertation presents remote sensing measurements from satellites and unmanned aerial vehicles (UAVs) to understand suspended sediment transport processes in estuaries like San Francisco Bay. Twelve methods for inferring suspended sediment concentration (SSC) from Landsat 7 imagery were compared using k-folds validation and assessed based on their abilities to recreate in situ SSC measurements from one meter below the surface. The best performer was the model of Nechad et al. (2010) using the red wavelength band with coefficients determined via Huber regression, with mean absolute error of 5.94 mg L-1 and bias of 0.15 mg L-1. Satellite-derived SSC observations compare well with USGS transects indicating that the method is well-suited to supplement cruise data that is costly to acquire and therefore limited in its frequency. Remote sensing measurements were aggregated by location, season, or tidal phase to understand the variability of SSC and to compare probability densities with in situ measurements. These results show that surface SSC is heightened in the shoals during summer months and has trended downward in Suisun and Grizzly Bays since 1999. Using satellite imagery from 2014-2017, remotely sensed surface SSC derived from the Nechad method was paired with bottom stress estimates based on two-dimensional hydrodynamic and fetch-limited wave models to investigate the relationship between surface SSC and flow. Observations of SSC closely fit a lognormal distribution though the shape, characterized by the modal value, depend on binning criteria including embayment, depth, and wave height. When binned by model-derived bottom shear stress, the modal value of the SSC distribution exhibited an inflection point at the critical shear stress for erosion. This suggests that remote sensing can be used to derive critical stresses that are otherwise difficult to measure. To account for the limitations of satellite imagery such as low spatial resolution and low temporal resolution (Landsat 7 overpasses occurred roughly once every 16 days), a method was developed to infer surface SSC from UAV-based imagery. While traditional remote sensing platforms take imagery at approximately a nadir viewing angle and provide multispectral images that are aligned with one another, an off-the-shelf camera aboard a UAV may not adhere to those qualities. Low cost multi-spectral cameras often include individual sensors for each band. The slight misalignment between images violates assumptions in two-band glint correction algorithms. Additionally, UAVs must tilt to fly and compensate for wind requiring images to occasionally be taken at angles more oblique than most satellite imagery. The method developed in this dissertation adapts previous techniques for sun glint correction for misaligned multispectral images and offers a novel approach to reduce the effects of camera orientation for oblique angles. During a field campaign, the UAV-based method to capture remote sensing reflectance was validated via comparison with in situ measurements made with a hyperspectral radiometer, and its ability to accurately infer SSC was verified based on in situ water samples. It was found that a polarizing filter is necessary to mitigate much of the glare on the water surface. A series of test flights were conducted to measure the surface SSC along a transect parallel to the Dumbarton Bridge during different phases of the tidal cycle. To reduce the impact of variability of incoming light, the flights were conducted over a period of 12 days at the same solar zenith angle during each day. Because the tide arrives later by roughly 50 minutes each day, consecutive daily transects over 12 days provided the variability over a tidal cycle. Cross-sectional sediment flux was computed from the remotely sensed surface SSC measurements and compared well to flux values estimated from in situ USGS observations.
590 $a School code: 0212.
650 4 $a Remote sensing. $3 535394
650 4 $a Physical oceanography. $3 3168433
650 4 $a Environmental engineering. $3 548583
653 $a Estuaries
653 $a San Francisco Bay
653 $a Sediment transport
690 $a 0775
690 $a 0415
690 $a 0799
710 2 $a Stanford University. $3 754827
773 0 $t Dissertations Abstracts International $g 82-02B.
790 $a 0212
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28104005