Looking at the study A compromise study of the effect of GIS data on hydrology and non-point source pollution modeling found in Agriculture Water Management, the authors of the study were looking to find and identify the problems of current sample mapping using the GIS method.
Using the Daning watershed in China, they studied different GIS studies done on the area. The GIS is a model that can take data from other studies and put all the data in one prediction looking at different variables it predicts certain events in the watershed. They found the biggest problem was that the method did better when it was a coarse study. Doing a fine study they could run the same data taken from different studies and get very different outcomes.
They looked at how the different variables relate to the GIS method. The GIS method was turning out to be bias against the SWAT data in certain instances. The SWAT data was used to predict the H/NPS for the GIS. This was one of the things the study set out to find. They wanted to find out if this was the case. The GIF program is growing in popularity and availability. Many are using it in their research
They would use twelve different combinations to produce the data. GIS was to help determine and predict the level of uncertainty at the watershed. Different GIS methods led to different input and different answers. GIS was chosen because of the pollutants in the water. Agriculture is the biggest cause of the pollution of the watershed. The authors of the study concluded that a coarse GIF method was better than a fine GIF model because of all the data they were using. They broke the watershed they were looking at into twenty-two separate lakes to use in the study. The sample size they were using was part of the problem of getting biased outcomes on the part of GIF. They had too much data leading to more uncertainty. A smaller sample size may have allowed them to use the fine model but considering the size of the watershed they were using was impossible.
This study was done mainly to look at the GIF’s ability to handle data from the SWAT model. The authors wanted to look at the newest and seemingly best way to graph samples. The authors all being experts in the field and knew the area. They were not so surprised to find that the fine GIF model was not suitable to their purposes and a coarse model of the GIF did much better considering all the data they were looking at.
The Daning watershed was the only source used to gather data from. It is not known if the fine model would act the same on all watersheds like it did in Daning. The uncertainty of pollution in the watersheds would also change from watershed to watershed. A lack of similar studies to compare to is another problem the researchers faced.
The study found a few things that could lead to break throughs when more study is done. More watersheds being looked at can only help in this area. More refinement of the program when looking at SWAT data from large samples might be called for. The study of more watersheds to see if this a reoccurring problem or just the size of the selected watershed was the problem I would like to see other watersheds from around the world studied. This study could lead to protection of underground water around the world when more study is done. It shows that manmade pollution of watersheds is on the rise and just how vunerable our water supply is.
Sources
Shen, Z., Chen, L., Liao, Q., Liu, R., & Huang, Q. (2013). A comprehensive study of the effect of GIS data on hydrology and non-point source pollution modeling. Agricultural Water Management, 118, 93-102. doi:10.1016/j.agwat.2012.12.005
