I have run a hydrological & hydrodynamic model of a river system for 72 hours time lag using MIKE 11 software. When I evaluate the performance of the model for 24, 48 & 72 hours, the R square value is coming 0.78, 0.38 & 0.39 and Nash Sutcliffe Coefficient (NSC) value 0.73, (-) 0.49 & (-) 0.79. How we can explain the negative value of NSC?
Dear Mani Kumar ,
NSC can range between -inf and +1. Special attention can be attributed to negative and positive values as well as 0 and +1. NSC = 1 means a perfect agreement between modelled and observed data; NSC = 0 indicates that the mean of observed data is as accurate as model predictions; NSC < 0 means the mean of observed data predicts better than the model.
All the best,
Alonso
An efficiency less than zero (NSE < 0) occurs when the observed mean is a better predictor than the model.
A negative NSE means with R2 fairly above zero means that your model has some relevance with reality in terms of the variation, but fails to reproduce the mean. You may try to change the model parameters focusing on preservation of the mean.
Article Evaluation of Variable-Infiltration Capacity Model and MODIS...
Dear Mani,
Following articles maybe helpful in improving NSE values:
https://www.diva-portal.org/smash/get/diva2:1060593/FULLTEXT01.pdf
Article Hydrological Response to Agricultural Land Use Heterogeneity...
Cheers
Simulated data variance is more than observed data variance.
E takes values in the interval [–∞, 1]. A value of E equal to unity shows a perfect match and an E value of zero means that the average of the observed values is as good as predicted values. Negative E values mean that the model is worse for prediction than taking only the mean of the observed values. E values higher than 0.9 show that the slope between observed and predicted values is nearly unity.
NSE indicates how well the plot of observed versus simulated data fits the 1:1 line.
Nash-Sutcliffe efficiencies range from -Inf to 1. Essentially, the closer to 1, the more accurate the model is.
-) NSE = 1, corresponds to a perfect match of modelled to the observed data.
-) NSE = 0, indicates that the model predictions are as accurate as the mean of the observed data,
-) -Inf < NSE < 0, indicates that the observed mean is better predictor than the model.
So, you have to rearrange the model's parameters depend on your objective. such as by showing pattern, maximum, and minimum value of the observed as well as the simulated output.
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators