**2.1. Geographic information system (GIS) and remote sensing (RS) application in land-use classification and its relationship with water quality**

Blackbird Creek is characterized by extensive salt marshes and large native populations of the salt marsh cordgrass (*S. alterniflora*). Blue crabs, as well as several other ecologically and economically important aquatic species, prosper in salt marsh environments where the leaves, roots, and stems of native plants provide sources of food and shelter from predators for many juvenile animals. Much of Delaware's other coastal wetlands including Blackbird Creek has been subject to a loss of biodiversity over the past several decades due largely to the common reed (*P. australis*) invasion. As reported by Roeske [101], this may be considered a highly disturbed ecosystem due to the invasion of *P. australis* and the intensive management (i.e., herbicide spraying) that have occurred since the early 1990s. Roeske [101] also reported that glyphosate is one of the most widely used herbicides in the world, especially with the advent of genetically modified organisms (GMOs), giving outcome to glyphosate-resistant crops. This chemical is referred to as a systemic, broad-spectrum herbicide, killing plants only a few days after application by blocking an enzyme that is vital in the production of amino acids and proteins [102]. Sadly, glyphosate is not species specific and generally kills all marsh vegetation. Roeske [101] stated that the long term and most effective management strategies for the control of *P. australis* have utilized a combination of the removal techniques including mowing, burning, herbicide treatment, increased salinity, and so on. These conditions encourage native plant colonization while pushing out non-native *Phragmites* by providing constant stress through these removal techniques. Roeske [101] also discussed that burning alone may actually reinvigorate the populations in subsequent years and thus these methods must be utilized in a strategic manner over the course of several growing seasons if any permanent removal is to be accomplished [103].

Stone [2] reported that ecologically and economically important resident fish and crustacean species in Blackbird Creek can be impacted indirectly by phosphate loading. Increased nutrient concentrations can increase phytoplankton productivity in aquatic ecosystems, which in turn can result in a drawdown of oxygen via increased respiration and decomposition of algae. Lowery and Tate [104] reported that dissolved oxygen reductions to less than 2 mg/L resulted in an increase of blue crab hemolymph lactate activity.

Stone [2] reported "as climate changes, it is simply not possible to discuss aquatic ecosystems without a detailed description of current and future trends in temperature, salinity, dissolved oxygen, pH, and the interactive effects of each of these variables on each other. This is also true with respect to change in land use. When land is modified to accommodate human needs, surface runoff, evapotranspiration, stream discharge, and sediment transport are all influenced, which in turn could have implications on the aforementioned parameters. If there is greater runoff from a crop field that was converted from a forested wetland, for example, there may be increased turbidity, reduced water temperature, and lower salinity. The potential ramifications in such a scenario would be monumental." The integrated use of remotely sensed data, geographic information system (GIS), and global positioning system (GPS) has enabled natural resource managers, state and federal governments, and consultants to develop management plans and policies for a variety of natural resource management applications for long-term sustainable options [100].

Following case studies provide research-based discussion on the application of hyperspectral remote sensing and its specific modification for the research needs. In this chapter section, our primary focus is on Mid-Atlantic Estuaries, specifically Blackbird Creek Watershed, St. Jones

**2.1. Geographic information system (GIS) and remote sensing (RS) application in land-use** 

Blackbird Creek is characterized by extensive salt marshes and large native populations of the salt marsh cordgrass (*S. alterniflora*). Blue crabs, as well as several other ecologically and economically important aquatic species, prosper in salt marsh environments where the leaves, roots, and stems of native plants provide sources of food and shelter from predators for many juvenile animals. Much of Delaware's other coastal wetlands including Blackbird Creek has been subject to a loss of biodiversity over the past several decades due largely to the common reed (*P. australis*) invasion. As reported by Roeske [101], this may be considered a highly disturbed ecosystem due to the invasion of *P. australis* and the intensive management (i.e., herbicide spraying) that have occurred since the early 1990s. Roeske [101] also reported that glyphosate is one of the most widely used herbicides in the world, especially with the advent of genetically modified organisms (GMOs), giving outcome to glyphosate-resistant crops. This chemical is referred to as a systemic, broad-spectrum herbicide, killing plants only a few days after application by blocking an enzyme that is vital in the production of amino acids and proteins [102]. Sadly, glyphosate is not species specific and generally kills all marsh vegetation. Roeske [101] stated that the long term and most effective management strategies for the control of *P. australis* have utilized a combination of the removal techniques including mowing, burning, herbicide treatment, increased salinity, and so on. These conditions encourage native plant colonization while pushing out non-native *Phragmites* by providing constant stress through these removal techniques. Roeske [101] also discussed that burning alone may actually reinvigorate the populations in subsequent years and thus these methods must be utilized in a strategic manner over the course of several growing seasons if any per-

**2. Case studies**

184 Water Quality

River Watershed, and Chesapeake Bay Watershed.

**classification and its relationship with water quality**

manent removal is to be accomplished [103].

High sediment concentrations in tidal creeks and estuaries are possibly due to transport through the catchment basin via tidal activity and wind-induced sediment resuspension [105], especially when land is modified to accommodate human uses. Wischmeier and Smith [106] reported a strong chance of the erosion of sediment from recently tilled land on any kind of slope due to rain and wind, and subsequent runoff into waterways, which could lead to a variability in light availability on the scale of hours (tides) and days (calm vs. windy days). In systems such as Blackbird Creek, primary production may be compromised due to diminished light penetration which could result in reduced zooplankton production and lead to lower recruitment of estuarine species [107], thus altering the flow of energy through food webs since the autotrophic base would not be efficient enough to sustain higher trophic levels [108].

Agricultural land use in the Blackbird Creek Watershed was identified using ArcGIS layers available in the Delaware Geospatial Exchange. To select agriculture treatments, the watershed was gridded out into 500 × 500-m cells and the percentage of agricultural land cover within each cell was calculated. Water quality analysis was conducted based on these treatments in the Blackbird Creek. Any correlation between orthophosphate concentration and turbidity in relation to the land use based on the percent agricultural use was the main focus of the project (**Figure 7**).

**Figure 7.** Sampling strategy shows the grid navigable tidal watershed into 500 m × 500 m cells and identifies the percentage of agriculture land cover in each cell. Samples were taken within each cell and differences of dependent variables across the percentages of agriculture were tested [2].

A considerable body of literature is available that explains the role of non-point source phosphorous runoff plays in the eutrophication of receiving water [109]. For agricultural watersheds such as Blackbird Creek, fertilization of soils greater than the total crop phosphorus uptake requirements leads to a buildup of phosphorous in the soil over time [110]. Studies have shown the positive relationship between soil phosphorus and surface water phosphorus, especially during high discharge events [111], indicating an association between soil management practices and surface water phosphorus concentration.

Holliday et al. [112] found 1:1 correlation between turbidity and total suspended solids for silt and clay fractions, but a smaller ratio for bulk-soil samples. Nevertheless, their research suggested that turbidity can be used to estimate sediment concentrations for fine soil fractions, which is, what the mouth of Blackbird Creek is primarily composed of. Other research has shown that reactive phosphorus increases with increased total suspended sediment concentrations [87–89, 113]. Based on these studies, our first goal was to identify trends in orthophosphate in accordance with measurements of turbidity.

Collection of water samples was undertaken over the course of 4 years (2012–2015) at various times of day between March and November. At each site (**Figure 7**), water was collected into dark polyethylene bottles (rinsed three times with sample water), then placed on ice for transport to the laboratory for nutrient analysis where samples were measured for orthophosphate and turbidity concentrations. Because orthophosphate is reactive in nature, a subsample of water was filtered on site into a separate bottle in order to minimize the potential for such reactions. Orthophosphate concentrations were determined using HACH Method No. 8048, the ascorbic acid method. This reaction provided information on soluble reactive phosphate concentrations between 0.02 mg L−1 PO4 3− and 2.50 mg L−1 PO4 3−. The measurements were performed using a HACH DR 3900 spectrophotometer (HACH, Loveland, CO). Turbidity was measured with an automatic wavelength selection on a YSI 9500 Photometer (YSI Incorporated, Yellow Springs, OH) and presented in Formazin Turbidity Units (FTUs), which is generally equivalent to Nephelometric Turbidity Units and Jackson Turbidity Units (JTUs). This method allowed for measurements between 5 and 400 FTU.

Based on the data by Stone [2], there were no correlations between turbidity and orthophosphate concentration. Linear regression analysis showed that there was a negative trend between the two parameters (*y* = −5.0822*x* + 52.734), but with an *r*<sup>2</sup> -value of 0.0016, this cannot be considered statistically significant (**Figure 8**). It did appear that the tidal regime was important with respect to the turbidity concentration. On incoming tides, there was a negative trend (*y* = −20.486*x* + 52.776; *r*<sup>2</sup> = 0.0628) (**Figure 9**), whereas on outgoing tides, there was a positive trend (*y* = 20.619*x* + 46.201; *r*<sup>2</sup> = 0.0175) (**Figure 10**). Neither of these trends was significant, but it is interesting to note the opposite slopes as they relate to the tidal movement upon collection. When the tide was receding, there was a positive slope in the data. Thus, orthophosphate concentration was greater with greater turbidity, while the opposite was true on incoming tides. It is hypothesized here that this could be due to the source of these parameters as they regard the tide. On incoming tides, there was greater phosphorus concentration (0.62 mg PO4 3−/L ± 0.020 SE) than on outgoing tides (0.55 mg PO<sup>4</sup> 3−/L ± 0.024 SE). On incoming tides, there was less turbidity (38.72 FTU ± 1.776 SE) than on outgoing tides (47.72 FTU ± 2.965 SE). Other unpublished research [101] has shown that orthophosphate concentrations upstream were considerably lower than those that were found at the sites from this study. Thus, it may be suggested that the positive trend line in the data on outgoing tides could be a function of reduced orthophosphate at the source and/or the size fractions of the sediment particles in the upstream portion of the tidal creek. Further research is necessary to verify this postulate.

A considerable body of literature is available that explains the role of non-point source phosphorous runoff plays in the eutrophication of receiving water [109]. For agricultural watersheds such as Blackbird Creek, fertilization of soils greater than the total crop phosphorus uptake requirements leads to a buildup of phosphorous in the soil over time [110]. Studies have shown the positive relationship between soil phosphorus and surface water phosphorus, especially during high discharge events [111], indicating an association between soil

Holliday et al. [112] found 1:1 correlation between turbidity and total suspended solids for silt and clay fractions, but a smaller ratio for bulk-soil samples. Nevertheless, their research suggested that turbidity can be used to estimate sediment concentrations for fine soil fractions, which is, what the mouth of Blackbird Creek is primarily composed of. Other research has shown that reactive phosphorus increases with increased total suspended sediment concentrations [87–89, 113]. Based on these studies, our first goal was to identify trends in orthophos-

Collection of water samples was undertaken over the course of 4 years (2012–2015) at various times of day between March and November. At each site (**Figure 7**), water was collected into dark polyethylene bottles (rinsed three times with sample water), then placed on ice for transport to the laboratory for nutrient analysis where samples were measured for orthophosphate and turbidity concentrations. Because orthophosphate is reactive in nature, a subsample of water was filtered on site into a separate bottle in order to minimize the potential for such reactions. Orthophosphate concentrations were determined using HACH Method No. 8048, the ascorbic acid method. This reaction provided information on

The measurements were performed using a HACH DR 3900 spectrophotometer (HACH, Loveland, CO). Turbidity was measured with an automatic wavelength selection on a YSI 9500 Photometer (YSI Incorporated, Yellow Springs, OH) and presented in Formazin Turbidity Units (FTUs), which is generally equivalent to Nephelometric Turbidity Units and Jackson Turbidity Units (JTUs). This method allowed for measurements between 5 and

Based on the data by Stone [2], there were no correlations between turbidity and orthophosphate concentration. Linear regression analysis showed that there was a negative trend

not be considered statistically significant (**Figure 8**). It did appear that the tidal regime was important with respect to the turbidity concentration. On incoming tides, there was a nega-

significant, but it is interesting to note the opposite slopes as they relate to the tidal movement upon collection. When the tide was receding, there was a positive slope in the data. Thus, orthophosphate concentration was greater with greater turbidity, while the opposite was true on incoming tides. It is hypothesized here that this could be due to the source of these parameters as they regard the tide. On incoming tides, there was greater phosphorus concentration

3− and 2.50 mg L−1 PO4


3−/L ± 0.024 SE). On incoming

= 0.0628) (**Figure 9**), whereas on outgoing tides, there was

= 0.0175) (**Figure 10**). Neither of these trends was

3−.

management practices and surface water phosphorus concentration.

soluble reactive phosphate concentrations between 0.02 mg L−1 PO4

between the two parameters (*y* = −5.0822*x* + 52.734), but with an *r*<sup>2</sup>

3−/L ± 0.020 SE) than on outgoing tides (0.55 mg PO<sup>4</sup>

phate in accordance with measurements of turbidity.

400 FTU.

186 Water Quality

(0.62 mg PO4

tive trend (*y* = −20.486*x* + 52.776; *r*<sup>2</sup>

a positive trend (*y* = 20.619*x* + 46.201; *r*<sup>2</sup>

**Figure 8.** Correlations between turbidity and orthophosphate (PO4 3−) concentration for both incoming and outgoing tides water [2].

**Figure 9.** Correlations between turbidity and orthophosphate (PO4 3−) concentration for incoming tide water [2].

**Figure 10.** Correlations between turbidity and orthophosphate (PO4 3−) concentration for outgoing tide water [2].
