Analysis Of Watersheds
2. Methodology 2.1 Study Area The water quality data used in this study comes from the six
watersheds that drain to Lake Erie: Maumee, Cuyahoga, Muskingum, Raisin, Grand, and Sandusky
(Figure 1). These six basins are close in spatial proximity (e.g., similar climate, ecoregion, and
cropping systems), and have publicly available long–term, daily time–series datasets (14–35 years).
The soil in Sandusky, the northern reaches of Maumee, and Raisin watersheds were clayey soil
which formed from beach sediments and glacial till associated with the glacial lakes. The central
part of Maumee and the southern part of Sandusky watersheds are characterized by coarse–textured
soils formed from glacial deposits. Muskingum watershed is characterized by ... Show more content
on Helpwriting.net ...
With 52% agricultural and 43% forest land, Muskingum was considered the mixed watershed, while
Grand is a forested (52%) watershed. Urban, agricultural, and forest categories are herein
characterized as having high, medium, and low anthropogenic impact, respectively. (Table 1)
(Figure 1) 2.2 Data Mining Techniques The Weka software package (Hall et al., 2009) was used to
implement the four classical data mining methods: multilayer perceptrons, K–nearest neighbor,
REPTree, and random forest. The linear regression was used as the reference method to predict the
nitrate concentrations in a multiyear time–series. 2.2.1 Multilayer Perceptrons (MLP) Multilayer
Perceptron (MLP) is an artificial neural network that learns nonlinear function mappings. An MLP
can be viewed as a logistic regression classifier where the input is first transformed using a learned
non–linear transformation. This transformation projects the input data into space where it becomes
linearly separable. This intermediate layer is referred to as a hidden layer. A single hidden layer is
sufficient to make MLPs a universal approximator. Formally, a one–hidden–layer MLP is a function
f: RD→ RL, where D is the size of input vector x and L is the size of the output vector f(x), such
that, in matrix notation: (Equation 1) with bias vectors , weight matrices and activation functions G
and s. The vector constitutes the
... Get more on HelpWriting.net ...