Table 1 Descriptive statistics of EC values in the stations of interest (2012–2021).

From: Multi-step ahead forecasting of electrical conductivity in rivers by using a hybrid Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model enhanced by Boruta-XGBoost feature selection algorithm

Metric

Albert River

Barratta Creek

Number of datapoints

3653

3653

Minimum

151

69

Maximum

924

726

Mean

459

380

Median

439

367

Standard deviation

160

118

Coefficient of variation (%)

34.8

31.1

Q1

345

299

Q2

439

367

Q3

563

453

Skewness

0.527

0.314

Kurtosis

− 0.141

0.087

Back to article page