**3. Final remarks**

The impacts of land use on aquatic health and marsh habitat have been a constant battle, and the rate of our wetlands lost is far more than they can be replaced. Tidal marsh ecosystems serve as great examples of dynamic ecosystems that can provide numerous lessons in restoration and management strategies. The ubiquitous incidence of aquatic species as both temporal and spatial perspectives indicates the importance of these types of systems to sustain the health of aquatic species and other important species in the face of anthropogenic and environmental stresses. Thus, it would be pertinent for managers to maintain the health of these areas by preventing any further alterations and damage, especially from nutrient runoff and abusive use of our coastal lands, and the invasion of *P. australis*.

The Blackbird Creek Watershed is composed of only 4% urban development. This provided a great opportunity to study the area that has very little anthropogenic impact, relative to other watersheds. How trophic dynamics can be affected as a result of various land uses is one of the concerns. There are several plots of land designated as cropland in Blackbird Creek. An understanding of crop rotation in these plots both within and between seasons can provide insight on whether changes in crops across years can affect trophic interactions and food web dynamics. Additionally, effectiveness of riparian buffers as blockades for fertilizer runoff is a significant impact on water quality. With sites associated with buffers and sites without buffers, different food web characteristics can be identified, especially if there are inconsistencies in the nutrient concentrations. While this is intriguing, it would be more beneficial if a second watershed with more urban areas and anthropogenic effects was used. The St. Jones Watershed is composed of over 21% urban development. If similar work can be done in each of these watersheds, we may be able to discern the effects of different land uses on Delaware's coastal waterways.

The Chesapeake Bay is under increasing pressures from anthropogenic disturbances at various temporal and spatial scales. Water quality monitoring is vital for assessing such impacts, and further provides important information for sustainable water resource managements. The research in this chapter demonstrates the applications of hyperspectral remote sensing in retrieval of the water quality parameters in such an optically complex system. Further development of retrieval algorithms is still needed in order for the remote sensing to be routinely used in the water quality monitoring.

Based on the studies discussed in this chapter, in almost every watershed studied, major land use and cover change are due to the conversion of agriculture land to developed land in the Mid-Atlantic region. Urban runoff has gradually become a dominate pollution sources to the natural environment, including wetlands and open water bodies. More attention needs to be paid to locate the position and size of land conversion in order to effectively manage and control urban runoff. Also, it is necessary to monitor the spread of invasive plants, especially *P. australis* and its impact on vegetation communities in coastal watersheds within the region. Hyperspectral images have advantages in accurate land classification and can keep track of detailed changes in land use and cover. This will help build more effective land management practices in Mid-Atlantic region.

All studies above reveal that it is feasible to measure water quality indicators including chlorophylla, TSS, and CDOM using hyperspectral images and simple spectral algorithms. These algorithms involve spectral bands in the red and red-edge ranges and they can capture unique spectral signatures of these water quality indicators such as chlorophyll-*a* in coastal waters. All studies suggest the advantage of using hyperspectral images in estimating chlorophyll a-concentrations, TSS, and CDO in various coastal water bodies. Future research focusing on the direct measurements of in situ inherent optical properties (IOPs) such as absorption coefficients and backscattering for different estuarine systems is needed to develop better algorithms for water quality retrievals which include TSS and CDOM by remote sensing in coastal waters. With application of remote-sensing technology, real-time data collection of water quality parameters and plant community changes for the entire watershed will help resource managers to develop better management strategies.
