>Flow Patterns and Alteration

Flow Patterns and Alteration

Timing and volume of water is important for maintaining geomorphic processes, literally keeping the channel wet for fish, and keeping water temperatures reasonable for aquatic life.
Biotic Condition

QT: Indicators

What is it?

Hydrologic alteration is a measure of how the current flow pattern of a river compares to natural flows. The amount of water in a river changes every day, sometimes a little bit, sometimes a lot. A hydrograph is a daily flow record over an entire year at one location on a river. The hydrograph changes throughout the year as seasons change and as water delivery and power generation schedules change. The natural hydrograph can be drastically different from the current hydrograph. Calculating the degree of hydrologic alteration is a way of quantifying how different the current conditions are from natural conditions.

Why is it important?

The ecological integrity of river ecosystems depends on their dynamic character (Poff et al.1997). The ecosystem in a river or stream lives in balance with the hourly, daily, monthly, and annual changes in water level. The transportation and germination of some plant species depend on flow pulse events, such as a spring storm runoff, followed by gradual receding flows. The flow pulse transports the seeds up onto the river bank and the receding flow provides the young seedling with sufficient moisture while its roots grow deeper where it can access more permanent water sources. If pulse flows from storm events do not occur because they are held back by upstream dams, the downstream reaches do not receive any ecological benefits from that storm. Similarly, macroinvertebrates and fish depend on storm flows for both habitat (such as the creation of cool pools that are safe for rearing young), as well as for the transportation of sediment and valuable nutrients. Maintaining the hydrologic variability of the natural flow regime is essential for the conservation of native riverine biota and maintaining river ecosystem integrity (Richter et al. 1998).

Target or Desired Condition

The desired condition for the hydrologic regime is to be as close to the hydrologic regime that existed in the watershed naturally prior to construction of any dams, weirs, or other structures that can attenuate flow, as well as any significant landscape alterations such as roads or change in type and amount of vegetation cover in a watershed. Each watershed has a unique natural hydrologic regime which depends on the type of vegetation in a watershed, the form of precipitation (snow or water), soil and sediment composition, groundwater connectivity, etc. Consequently, it is not possible to have a single hydrologic regime that all watersheds should follow.

What can influence or stress condition?
As previously mentioned, hydrologic alteration is caused by the presence of dams, weirs, paved surfaces (delivering runoff), road-crossings, and/or other structures that can alter flow. In addition, changes to vegetation cover, channel/floodplain connectivity, and soil composition (e.g., loss of some types of soils from surface erosion) can all contribute to hydrologic alteration because different types of vegetation and soil have different abilities for soaking up and releasing water


In order to calculate the degree of hydrologic alteration the historic flow record was divided into two groups, pre-project (before the dam was constructed) and post-project (after the dam was constructed). The construction of the most resent and/or downstream dam was used as the project by which to divide the flow record (Table 2). The pre-and post-project data sets were compared using a program called Indicators of Hydrologic Alteration (IHA version 7.1). Six different hydrograph components were selected to represent the hydrograph from which the degree of hydrologic alteration was calculated (Goa et al. 2009). These hydrograph components were chosen because, selected together, they capture the signature of the hydrograph and reduce information redundancy (measuring the same thing more than once) that can occur when too many hydrograph components are used. The selected six hydrograph components that were used for this analysis are:

  • 30-day minimum, represents base flow or summer monthly flow
  • 7-day maximum, represents high flow
  • February flow, represents winter monthly flow
  • November flow, represents fall to winter monthly flow rate of change
  • June Flow, represents spring to summer monthly flow
  • March flow, represents the timing of extreme events

Selection of hydrograph components: The six hydrograph components that were measured using IHA (version 7.1) are called range of variability (RVA) values and they represent six principal component axes (PCA axes) described in an analysis by Goa et al. (2009). Two of the eight IHA parameters selected by Goa at al. (2009) were excluded from this analysis; these were high pulse duration and rise rate. The reason these two parameters were excluded was that they fell into group 4 and 5 respectively of the IHA analysis and only groups 1 and 2 are considered reliable when portions of data are missing in the period of record. In addition, these two parameters explained the least amount of variation in the overall principal components analysis.

Determining the natural range of variability: Each hydrologic characteristic is calculated for a pre-project period which creates a data set that represents a more natural flow regime. The acceptable RVA is between the 25th and 75th percentile determined by the pre-project data set. The pre-project data set is compared to the post-project data set to determine the frequency with which the post-project data set falls within the acceptable range of variability. A positive deviation indicates that annual parameter values for the hydrologic characteristic fell inside the RVA target window more often than expected, a negative value indicates that the annual values fell within the RVA target window less often than expected.

Calculating the degree of hydrologic alteration: The absolute value (distance from zero) of the RVA values for each of the six hydrologic characteristics were averaged to get an overall percent of hydrologic alteration for each subwatershed (Richter et al. 1998).

The degree of hydrologic alteration is considered high, if the average is greater than 60%, low, if it is less than 33%, and medium, if it is between 34 and 59%. The average degree of hydrologic alteration is converted into the report card format by subtracting the absolute value of the average hydrologic alteration from 100. In this way a high degree of hydrologic alteration for an average score of -74, becomes a report card score of 26, a very low grade. Table 3 gives the report card score for each subwatershed.

This indicator's data resource has not been uploaded yet.

What did we find out/How are we doing?

The streams and rivers in the Feather River Watershed are moderately to highly altered. The most highly altered subwatersheds are the Lower Yuba and Lower Bear (Figure 1 and Table 1). The factor most strongly driving the hydrologic change in the Yuba and Bear Rivers is the timing of extreme flow events. Spring pulse flows are typical of regulated, i.e., dammed, rivers because reservoirs that are empty from high summer water demand fill up all winter long and in the spring spill more often than early in the rainy season. Spring extreme events may have the same geomorphic (physical) benefits as extreme events in the winter, but the biological benefits of the extreme events can be drastically different for fish and macroinvertebrate populations, depending on the time of year when they occur. In general, the construction of dams has shifted the timing of extreme events from winter to spring.

One of the least altered of the analyzed subwatersheds is Deer Creek. The most highly altered aspect of the Deer Creek hydrology is the 30-day minimum, i.e., summer base flow. Post-project construction (after the installation of Scotts Flat Reservoir) summer base flows on Deer Creek are abnormally high. Unnaturally high summer flows in Deer Creek are likely a result of agricultural water demand in the lower watershed during the summer months. The ecological impacts of unnaturally high flows is that it may make summer rearing for sensitive species, such as young fish and frogs, more difficult. In addition, water release from both Lake Wildwood and Scotts Flat can result in massive scour can occur if the water is released too high, too fast.

Temporal and spatial resolution

The data used in this analysis only represent a small portion of the whole river. A measure of hydrologic alteration at a particular stream gauge site is limited in geographical extent and cannot be representative of the entire watershed. Dams tend to re-set the system by introducing a new flow release schedule and tributaries tend to mask the signal of the main stem hydrology by contributing either a more natural contribution or a more highly altered contribution depending on if the tributary has a dam on it or not. Therefore, the measurement of hydrologic alteration was considered to extend downstream of the gauge site to the first confluence with a major tributary and extend upstream to the location of the first upstream dam (Richter et al. 1998). The result is that there are large areas of stream reaches that are not covered by this analysis, especially in the upper watersheds.

How sure are we about our findings?

There is no perfect record for the natural flow regime. This analysis is limited by the available historical flow data which was divided up into two time periods; a pre-project period, which was prior to a dam being constructed, and a post-project period, which was after the construction of a dam. Invariably, other landscape changes are also taking place that also influence hydrology such as stream incision, road density and vegetation changes, but this analysis is not sensitive enough to capture these changes. In addition, each river has multiple dams on it making it difficult to select the appropriate pre- and post-project period, which in turn makes it more complex to measure the degree of hydrological alteration. In some cases, especially for the Upper Feather River the gauges were installed at the same time that the dam was constructed, therefore, there is no pre-project flow data, or natural flow record, to compare to current flows. Ideally, naturalized flows for each subwatershed would be generated using hydrologic models and would constitute the baseline that the current flow regime would be compared to. This is a critical data gap and source of uncertainty.

Comparing two different time periods can be like comparing apples to oranges. When working with historical hydrological data that covers a long time frame, say twenty years before a dam was built to twenty years after a dam was built, it is important to remember that other climate-related changes have occurred which may influence the hydrograph. For example, if the twenty years of pre-project data occurred during a drought then the hydrograph does not capture the extent of variability in flow/hydrologic regimes that the watershed can experience. In an attempt to overcome this potential source of error, hydrological analyses take into account different water year types (critical, dry, below normal, above normal, wet) when comparing pre- and post-project hydrology. This analysis uses the complete flow record that was available and as a result does not explicitly look at different water year types.