I recently read the document that the Suwannee River Water Management District will use to justify taking an additional 8% to 9% out of an already overdrawn and slowly dying river, which we have been trying to protect for years.
The 117-page document was a highly technical and difficult read, but I felt it was important. Entitled “Minimum Flows and Minimum Water Levels Re-Evaluation for the Lower Santa Fe and Ichetucknee Rivers and Priority Springs – Draft,” it was prepared for the district by HSW Engineering in December.
The impetus for this report is the development by the Suwannee River and St. John’s River water management district of a new groundwater flow model. I am not a hydrologist, but I have done some financial modeling, and I recognize that no matter how strong the model, results are dependent on the validity of assumptions used in the model. I also know that desired results can be obtained from any model if the assumptions are “properly” tweaked.
As I understood it, the model works like this – they establish a “reference timeframe flow” by estimating groundwater use over time and adding it back to recorded flows over the same period (I like to think of this total as “the flow that nature intended”). They then identify 10 water resource values, which include such elements as recreation on the water, fish habitat and migration, aesthetics, filtration and absorption of nutrients and other pollutants, sediment loads, etc.
Next, they try to quantify the allowable reduction in reference timeframe flow before the water resource values are significantly harmed. That reduction represents the amount available for consumptive distribution.
What are the inputs (assumptions) for this model? First of all, they need to estimate groundwater use over time, for which there are records based on consumptive use permit monitoring. However, consumption from private wells is significant, and had to be “estimated … based on population”.
They then need to determine what the flows have actually been over the same period. There are records of this from the many gauging stations on the river, although for various reasons there are gaps in the data set. They adjusted these gaps with the “infill” of data, using linear regression and linear interpolation, and other often used and well-respected mathematical manipulation. However, not having been observed is a notable weakness.
Finally, they need to determine “response functions,”, which indicate how water resource values respond to flow variations. For this they call on yet another model, the HEC-RAS developed by the U.S. Army Corp of Engineers. The model was calibrated and checked for validity by running it against historical observed flows and assigning an efficiency value.
Of all the locations tested, six were rated very good, one was good, four were satisfactory and two were unsatisfactory. Based on that record, they determined that the model “provides an acceptable simulation of flows and water depths.”
Based on the application of the model, “flow duration curves” were determined – that is, the lowest flow that would keep each of the water resource values from significant harm. That determination led to the conclusion that additional flow reductions of between 8% and 9% were available on the lower Santa Fe River, and additional flow reduction of 2.8% is available on the Ichetucknee.
It is important to note that this is a gross oversimplification of what was a very detailed process, and I don’t question that the best available science went into this analysis. What I do question is whether or not, even in the hands of knowledgeable and capable scientists, best available science is strong enough to justify additional reduction in flow in light of 1) the vast amount of data “infill,” data estimates, and questionable validity results coupled with 2) the precautionary principle that we would hope our resource protectors are following.
For all the applied science, simple observation of the river and the springs tells us that they are in poor health, and that health has declined over the last decade even as protections for impairment have been ostensibly applied. When observation contradicts modeled results, shouldn’t the precautionary principle dictate no change to action until that contradiction can be reconciled?
My skepticism is bolstered by the fact that this was not an unexpected result – when they went to a new model, additional water availability was sure to result.
We are left with the hope that those individuals selected to protect our resources will do so fairly and diligently.
Michael Roth is president of Our Santa Fe River Inc.