Assessment of spatial structure of groundwater quality variables based on the entropy theory

Abstract. Fundamental to the spatial sampling design of a groundwater quality monitoring network is the spatial structure of groundwater quality variables. The entropy theory presents an alternative approach to describe this variability. A case study is presented which used groundwater quality observations (13 years; 1987-2000) from groundwater domestic wells in the Gaza Strip, Palestine. The analyses of the spatial structure used the following variables: Electrical Conductivity (EC), Total Dissolved Solids (TDS), Calcium (Ca), Magnesium (Mg), Sodium (Na), Potassium (K), Chloride (Cl), Nitrate (NO₃), Sulphate (SO₄), alkalinity and hardness. For all these variables the spatial structure is described by means of Transinformation as a function of distance between wells (Transinformation Model) and correlation also as a function of distance (Correlation Model). The parameters of the Transinformation Model analysed were: (1) the initial value of the Transinformation; (2) the rate of information decay; (3) the minimum constant value; and (4) the distance at which the Transinformation Model reaches its minimum value. Exponential decay curves were fitted to both models. The Transinformation Model was found to be superior to the Correlation Model in representing the spatial variability (structure). The parameters of the Transinformation Model were different for some variables and similar for others. That leads to a reduction of the variables to be monitored and consequently reduces the cost of monitoring.

Keywords: transinformation, correlation, spatial structure, municipal wells, groundwater monitoring, entropy