The major theme of this paper is that management of seagrass must deal with issues of geographic scale. Approaches at several scales are needed. Examples are drawn primarily from management programmes for the 250 km long Indian River Lagoon system on the south-east coast of Florida. The Lagoon has several attributes of spatial variation that require approaches at various scales (e.g., from 1:1 000 000 to 1:1). Risks and errors of scaling up and scaling down are described. For large-scale approaches, remote-sensing mapping methods are generally appropriate. In the Indian River Lagoon, true-colour aerial photographs. are typically taken every 2?3 years at 1 :24 000 scale. Such Lagoon-wide maps have fuzzy boundaries and cannot be scaled down to fine scale, but they can be scaled up. At large scale, seagrass restoration/protection targets (to a depth of 1.7 m) are reasonable, but are unreasonable at fine scale. For monitoring change within a bed or meadow at metre to 500 m scale, monitoring of fixed transects is a powerful tool. However, the technique has limited power for comparisons among beds, which requires multiple transects. To build a predictive model, a site-specific study examined the relationships among light, water quality, and seagrasses. The link between seagrass and water quality is made through a light attenuation model incorporating both water column and epiphytes. Extensive sampling is required to test the robustness of the model at all scales. No single scale is appropriate for all approaches, and no approach applies over all scales. If such considerations of scale are not incorporated, errors of measurement, inappropriate techniques for assessment, implementation of wrong solutions, and a lack of understanding of the system under study can result.
Spatio-Temporal Analysis of Water Quality Parameters in Machángara River with Nonuniform Interpolation Methods
