In ‘The Other Extreme – Part 1: the Measurement of Absolutely Nothing’ I presented an argument that the difference between ‘exactly zero flow’, and ‘nowhere near enough water to generate flow’ can be important from a water resource management perspective.
We rely on discharge data to inform us about water availability.
This reliance is based on an assumption that discharge data represents information both about the hydrology upstream of the gauge as well as hydraulic conditions downstream of the gauge. Zero flow marks the point of departure between this assumption and the truth. Now I would like to further expand on this theme that our ‘measurements of nothing’ really matter.
Several years ago I walked extensive sections of the Little Campbell River at low water as part of a study investigating links between land-use and water quality. I discovered long sections of dry river bed, even though flowing water was observed both further upstream as well as downstream.
It is not unusual for land-use change to result in soil loss with resultant sediment deposition in the stream. These slugs of sediment are pushed downstream through successive storm events. The stream bed is raised in the deposition zone, a process called aggradation.
In this case, reach scale aggradation was likely due to active landscape transition from rural to suburban development. During low water events the river volume can be insufficient to breach the aggraded river reach, but the upstream water pressure can be sufficient to force the water into the hyporheic zone, from which it eventually emerges downstream of the aggraded reach. The aggraded reach can, in this way, become dewatered during the low flow season.
Consider the analysis of a time series of data for a stream with a change in land use.
A pulse of sediment may progress through the gauging section over the span of several years. There may be zero flow at the gauge during low flow periods while the flow might be perfectly ‘normal’ both upstream and downstream. A naïve analyst may be alarmed at the unprecedented occurrence of zero flow in the data and incorrectly attribute the effect to an unrelated cause (e.g. to climate change).
Another interesting project was an investigation of a regional statistical model for the estimation of the annual low flow by validation against local scale data. For this experiment we conducted a longitudinal study of discharge on the Koidern River in South West Yukon during the late winter. The results were really interesting because the data plotted almost orthogonally to the predictive relation.
The reason for this was a massive buildup of ice at each confluence of the river resulting in series of ice dams. Each dam would force more water into the hyporheic zone resulting in a net dewatering of the river in a downstream direction.
The most fascinating observations of this study were of the linkage between the hydrology of zero flow and ecology.
At each ice dam water would be forced into the hyporheic zone to emerge in a downstream pool formed behind the next ice dam. These pools were ‘nearly’ ideal winter refugia for fish. I say ‘nearly’ ideal because these refugia were not as perfect for the fish as for a family of fat river otters I enjoyed watching as they played in the snow. I presume these otters appreciated having an abundant supply of captive fish for their dining pleasure.
Even more interesting is the Fishing Branch River in the Northern Yukon. Unfortunately, I was never able to secure funding for an investigation of this river. It has been noticed in recent years that a section of river has been going dry in late August and September. This is important because the dry channel is a barrier to the run of Chum Salmon. Salmon returns are now in the range of about 30,000 fish, whereas in the early 1970’s runs in excess of 300,000 were observed.
The Fishing Branch is ideal for spawning because of a very unusual hydrological phenomenon.
It always has ample flow through the winter season. This flow must be from a deep groundwater source because there is no apparent source of runoff during the long Yukon winter and the source is warm enough that the river stays open even during extended periods of extreme cold (< -40 oC!) temperatures. The question that remains unanswered is: why would the river (increasingly) go dry in late summer but then have plenty of water during the long Yukon winter?
Even measurements of small flow require special attention. For example, late winter measurements at Big Creek in the Yukon were challenging. There would only be a very small amount of flow under 1.5 to 2.0 m of ice covering the channel at the gauge. It was necessary to drill to bed to locate the flowing channel under a very large volume of ice, taking a lot of time and ruining many ice drill bits. Conveniently, there was a section of river a short distance downstream that stayed open year round providing a location for an easy winter wading measurement saving us considerable time and effort. It took several years for us to notice that late winter flow was always substantially higher at this wading section than at the gauge in spite of no apparent source of inflow.
Understanding and accurately characterizing water availability has arguably never been more important than it is now. This is likely to become even more critical in coming years as a result of changing climate and land use. The study of ephemeral streams is needed to develop our ability to manage water wisely.
However, in order for measurements of zero flow to benefit our understanding of water availability, we need to provide sufficient context for this most important number.
Is there flowing water anywhere upstream of the gauge? Is there water flowing downstream of the gauge? Is there standing water anywhere in the channel with zero velocity? Are there fish taking refuge in these pools or disconnected river reaches? Has there been a change in the offset of the rating equation? This last question is important for attribution of cause. In any case, notes, maps, sketches, and photographs of zero flow may be as important as the data itself. Every aspect of small and zero flow gauging can be very important for accurate interpretation of the data.
How important are intermittent streams?
Acuna et al. (2014) report that flow intermittency is so common that it makes up the majority of river networks in many regions. One third of fifth-order rivers and almost 70% of first-order streams south of 60o latitude only flow intermittently. Temporary waterways perform complex and vitally important ecosystem services. Temporary waterways are under increased stress because they are not always managed with the same care as permanent waterways. In the EU a temporary stream or river may not be considered a water body, whereas Australian state and federal legislation explicitly include temporary streams and rivers. Policies will continue to evolve and be informed by new science on the role that intermittency plays in water quality and quantity. This science must, in turn, be informed by accurate measurements and monitoring of absolutely nothing.
Acuna, V., T. Datry, J. Marshall, D. Barcelo, C.N. Dahm, A. Ginebreda, G. McGregor, S. Sabater, K. Tockner, and M.A. Palmer. 2014. Why should we care about temporary waterways? Science. Vol. 343 no. 6175 pp. 1080-1081. DOI:10.1126/science.1246666. http://www.sciencemag.org/content/343/6175/1080.summary
Todd River near Alice Springs – Frank Weber photo