Real-Time Forecasting of Deep Creek Hydro Operations, Part 3

Preface

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.

When you click the Knit button in RStudio a document will be generated that includes both content as well as the output of any embedded R code chunks within the document within a HTML document.

This part 3 of the report is about attempting to produce some kind of rain forecast that can be used in WAM (Water Allocation Methodology).

Introduction

The WAM attempts to allocate the water of Deep Creek Lake that exists above the lower rule band equitably among its shareholders. It might be made part of issuing a new operations permit to Brookfield, which is coming up for renewal in the spring of 2019.

It’s easy to determine how much water can be released through the generators on a given day. This has been analyzed in a previous report. This basic issue is: “Can we predict how much water to release without ever drawing down the lake level below the lower rule-band?”

At any given time, the water level of the lake is based on past practices and on whether it rained since the last forecast or not. Here are the approaches that one could use to make a release forecast.

1. A conservative approach to allocating the water is to assume that no water is going to be added to the lake during the forecasting period. That ensures pretty well what is possible.
2. A second approach is to use a weather forecast and to account for the the rain that has fallen on the watershed since the last forecast. To make this approach as reliable as possible one would have to correlate the forecasted rainfall with what actually happened and we would like to know how much of the forecasted rain actually contributes to rise of the lake level.
3. A third approach is to use historical weather and use Markow chains to forecast the following days rain.
4. The fourth approach is to use the average of weather forecasts done by multiple weather services.
5. The fifth approach is to use a mixture of the four previous approaches.

In the following sections we shall look how each of these approaches can be implemented. But first we need to define our stakeholders and what they require.

The Stakeholders

There are basically two groups of stakeholders that are affected by the number, duration and timing of the water releases from Deep Creek Hydro, those that depend on the water level of the lake, and those that depend on the waters of the Youghiogheny River, the river that receives the waters from a release through the Deep Creek Hydro facility.

The stakeholders that depend on the water level of the lake are:

1. Brookfield,Inc., because the more water there is in the lake the more power they can produce and sell.
2. The property owners around the lake, because the higher the water level is the longer they can keep the boats in the lake during the important summer months and moored at their respective boat-slips.

The stakeholders that depend on the water flow in the Youghiogheny are:

1. The white water rafters and kayakers because they need water depth in the river to navigate the river safely
2. The trout fishermen because without the cooling effects of water releases from the Deep Creek Hydro Electric facility the trout population can decrease significantly.

The ability to use the waters of Deep Creek Lake to generate power depends on being able to satisfy the other requirements first. Brookfield Inc., who owns and operated the Deep Creek Hydroelectric project and the dam, have as their first duty to keep the facilities safe. Mostly this means keeping the water from flowing over the spillway and minimizing shoreline erosion

Historically, the first has been relatively easy to accomplish since in its 60 yr history it’s been reported that water flowed over the spillway only three times.

The second part, minimizing shoreline erosion is less defined and requires more research. The goal is to keep the water from going above 2461 ft elevation. At the higher levels, shoreline erosion is highest from waves induced by 24/7 weather. The 2461 ft elevation allows for a one feet margin before going over the spillway. The water is at it highest levels in the springtime, dropping eventually to its lowest levels in December and January. These lowest levels are defined by the lower rule-band curve

The property owners around the lake have a significant interest in keeping the water level of the lake as high as possible since that will allow their boats, or the visitor boats, to remain a float at all times. The Southern part of the lake is most susceptible to lower lake levels since it is overall shallower. Boat-slips are important to rental properties around the lake.

White-water rafting has been a relatively stable business over the years. The issue arises whether the economic benefits derived from it adequately compensate for the loss of business for boat-slip owners at the lake. There is no such data one way or the other.

Trout fishing in the Yough is strongly protected by national organizations although there is no data about how much they contribute. Access to the Yough along the stretch of the river from the tailrace of the hydroelectric facility to the Sang Run River bridge is highly limited because of protected land and private property. There is also virtually no knowledge of how the fish population survives in that area. Currently, Temperature Enhancement Releases are scheduled when the Yough waters get too high, but there is no evidence that it benefits the trout fishing experience.

Given all of this, the hydroelectric facility benefits from ALL releases. Power is generated and income is produced. So the only real requirement is to keep the boats in the lake afloat and to maximize the number of white-water releases. The boat-afloat issue is basically taken care of by the lower rule-band, which may require a slight tuning as suggested in the graph in Part 1 of this report. Hence the focus should be on accommodating and curtailing white-water releases.

The operating permit for the hydroelectric facility dictates when white water releases should be held. Rather than dictating them, the schedule should be a suggested schedule, accommodated by the availability of water as predicted by the WAM. In all likelihood it is entirely possible that more than the suggested water releases can be made.

The rest of the report will focus on generating a schedule for white water releases and develop lake water accumulation rate from historical data, and use that date to test a WAM scheme.

White Water Release Schedules

Reverse Engineering of Water Entering the Lake

Accomodating White Water Releases by Reversed Water Inflow Schedules

Conclusion

Author: PLV
First Published: March 3, 2018
FILE: 2018-03-03-Real-Time Automated Analysis, Part 3
2018 PLV Some Rights Reserved
Contact: pete@senstech.com