Interaction of Coastal aquifer and Lake Urmia

Hey guys.

As I discussed previously, I am a true believer of  existence of interaction between coastal aquifer and Lake Urmia water level. Many authorities and politicians refuse to accept the theory and there are some research articles based on rejection of existence of such an interaction.

Recently I used to publish a conference paper (ASCE, EWRI 2015) about the interaction of water level in some random coastal aquifers in West coast of the Lake Urmia basin and water level in the Lake itself. I used a soft computational method named "Decision Tree" to manipulate my model. It is based on Entropy and probability. Evidence and results of this model are in agreement with a theory of existence of such interaction in Coastal aquifer.

Fig. 1 shows the schematic relation between lake and coastal aquifer which I believe that exist in the hydrological process. In general in closed basin lakes, such interaction is one of the main hydrological variables that should be considered and studied carefully.

Fig. 1. Schematic of interaction of coastal aquifer and Lake Urmia in balance

So I used to select some random wells just near to the west coast of the lake. You can find the position of this wells in Fig. 2. Data in east coast is not ready for use for now and I will try to manipulate them a.s.a.p. Followingly, a Pearson correlation coefficient test between Lake water level and water level in wells is done and interesting results are shown in Fig. 3 with a radar chart including the direction of such relations.

Fig. 3. Correlaogram radar chart

Fig. 2. Position of wells in west coast of the Lake

It is obvious that, there is strong linear relationship specially in North and South of the basin all with negative values. Same analysis on probability distribution function of lake water and water level in wells showed strong similarities in shape and moments of distribution. I have done some investigations on the structure of cross-correlations in time and space between lake and coastal aquifer. Two samples of such investigation are shown in Fig. 4. You can see seasonality and strong interaction between lake and coastal aquifer. As shown in Fig. 3 and 4,  these two stations (Station 1 and 6) have the most impact on the interaction.

Fig. 4. Cross-correlation between lake water level and water level in wells of station 1 and 6

I though a model may reveal more detailed structure of the relation, so I used to select a probabilistic one. As entropy concept is very popular now a days I used DT for manipulation of data and calibrated my tree. Here is the scatter plot of my model in Fig. 5. As you can see these are strong estimation result and I personally satisfied with the results.

Fig. 5. Scatter lot of DT model

That is all I was eager to share for now!

So I think I proved my theory at least to some extent. You may find out my paper's abstract in Related page in my weblog and/or download the whole article from ASCE library.

Please share your points of view with me.

Thank you

ANN, Decision Tree and Regression Methods for Forecasting Monthly Lagged Lake Water Level

Hi guys,
Here I am sharing the abstract of my presentation (Paper) in EWRA 2015 which be held June 10-13, 2015 in Istanbul. I will share more details as soon as it is published in the abstract book.

ANN, Decision Tree and Regression Methods for Forecasting Monthly Lagged Lake Water Level 

Babak Vaheddoost(1), Hafzullah Aksoy(2), Hirad Abghari(3), Saieed Zare Naghadeh(4)

(1) Department of Civil Engineering, Istanbul Technical University, Turkey, e-mail: vaheddoostb@itu.edu.tr
(2) Department of Civil Engineering, Istanbul Technical University, Turkey, e-mail: haksoy@itu.edu.tr
(3) Faculty of Natural Resources, Urmia University, Iran, e-mail: h.abghari@urmia.ac.ir
(4) Department of Civil Engineering, Dokuz Eylul University, Turkey, e-mail: saieed_zare@yahoo.com

Abstract:

The forecasting of lake fluctuations with previously observed data is a necessary task in closed basin lake hydrologic analysis. Known as a hyper-saline lake, Lake Urmia is dealing with atrophy and mismanagement through recent years. In this investigation, nine methods in the category of parametric and nonparametric approaches were applied for modelling Lake water level. Eleven months in the form of monthly lagged data were used as independent variables and used arbitrarily for each tested model. The present situation of the lake was considered as the dependent variable. The parametric approaches, used in modelling were multi linear regression (MLR), nonlinear regression (NLR) and decision tree (DT). Nonparametric approaches applied in the modelling were artificial neural networks (ANN) with different functions. Generalized regression neural network (GRNN), radial base function (RBF) and feed forward back propagation (FFBP) were used as ANN models. Three criteria of coefficient of determination, Lin’s concordance correlation coefficient and root mean square percentage error were used in comparison of the results. Each model was given a grade through each criteria of the measurement in the magnitude of zero to ten for comparison. The summation of the grades for each method was accepted as the prosperity of each method. Results show that the NLR is the superior method of all, while GRNN shows the worst results.

Key words: Decision tree, ANN, lake water elevation, hyper-saline lake, Lake Urmia

Bbak VDST