6 Comments

By Daniel Collins

As Ross mentioned some time ago, one of the frontiers of hydrological research at present is the interface between surface water and groundwater. On the one hand, we need to understand how aquifers are recharged from the surface; on the other, how aquifers in turn discharge water back to the surface. This is important to water resource managers so that they can determine how water use at one location may affect water availability and aquatic ecosystems elsewhere.

One question, particularly relevant to both Canterbury and Hawke’s Bay, is how fluctuations in groundwater levels affect spring flow near the coast. This is actually part of a larger research programme we have looking at the environmental effects of water use.

As part of this research I have been examining streamflow and groundwater data around Lake Ellesmere/Te Waihora, Canterbury. Environment Canterbury has an extensive network of groundwater monitoring wells and streamflow monitoring sites in the area. The wells measure what is known as the piezometric head of an aquifer, derived from the Greek piezein meaning ‘to press or squeeze’, and which refers to the height that water would rise to if subject to atmospheric pressure. For the top-most aquifer, this height basically means the water table (see this earlier post on groundwater anatomy for more).

IrwellSWGWBy comparing piezometric and streamflow data, we can develop a picture of how fluctuations in groundwater translate into fluctuations in streamflow. In the figure below I’ve chosen just one well (L36/0141) and one stream (Irwell River). The first thing you see is a roughly linear increase in flow with piezometric height, at least on the right. Linear relationships like this are typical of water flowing through a saturated porous medium (see Darcy’s law for more). But where the flow drops to around zero, at a piezometric height of about 68 m above mean sea level, there is a hint that this line bends — a kink in the groundwater-streamflow relationship. In fact, it has to bend because you can’t go lower than zero flow.

This kink indicates a threshold groundwater level that divides streamflow behaviour in two. Above the threshold, streamflow increases roughly in step with groundwater level; below the threshold, there is basically no flow at all, no matter how low the groundwater drops. So if the level in well L36/0141 drops below about 68 m AMSL, Irwell River is likely to go dry.

While in some instances this drying is entirely natural, changes in groundwater recharge (say, due to climate change) and abstraction for irrigation could shift the hydrological regime of the stream to become more ephemeral. It could also mean less water flows into Lake Ellesmere. This is all very important when allocating water while avoiding undue water resource or environmental impacts.

Threshold behaviour like this is actually quite common in hydrology, and can be traced back to the underlying physics of water movement. Having a physical explanation of the interaction between aquifers and streams allows us to make more robust predictions of spring flow, and of the ecological and water resource implications that follow.