Climate Change and Navigation

Navigation on the river Rhine is of great economic importance for The Netherlands. It is important to establish whether river dunes significantly restrict water depth and whether climate change influences their development. The authors present research findings that suggest minimal influence of river dunes upon navigation. River dunes are bedforms, with a height in the order of a metre. Therefore, they may restrict the water depth for navigation. Dune development shows a delayed repsonse to changing discharge. Climate change is expected to lead to a larger variation in river discharge (higher floods and more extreme droughts). Therefore, climate change may lead to higher river dunes that are still present at low discharge and thus to hindrance of navigation. In this paper it is investigated whether this is true.

Climate Change Scenarios
Many events such as more frequent flooding and extreme drought are attributed to rising average global temperature. Several scenarios have been developed to forecast the possible consequences of this type of climate change for the Rhine basin. One is the UKHI scenario used to determine expected discharges in the Rhine [1] and based on prediction of a global temperature rise of 2°C towards the year 2100. Low and high estimates of 1°C and 4°C, respectively, were specified to explore accuracy range. The scenario results in factors that can be multiplied with recorded discharges. This implies that the discharge pattern remains equal. Recorded discharges for three different years (‘base years’) with differing discharge patterns and peak discharges were selected. Multiplication with the factors derived from the UKHI scenario led to estimates of expected future discharges. Recorded and expected discharges were used as input for an 1D-hydraulic model for unsteady flows, to compute corresponding water depths. These were then used to calculate dune dimensions.

River Dune Development
River dunes are asymmetrical bed-forms that develop in sandy-bed rivers at high discharge. Earlier research on river dunes in the Rhine showed the presence of dunes of about 0.8m at a discharge of 7,000 m3/s. Dunes reached a height of about 1.6m at an extreme discharge of 12,000 m3/s.
Dunes form or degrade due to erosion and sedimentation processes. In addition, they migrate as a result of erosion of the upstream side and sedimentation on the downstream side of the dune, as shown in Figure 1. These erosion and sedimentation processes require time. Therefore, dune development shows a delayed response to changing discharges, here called the ‘time-lag effect’. Consequently, the maximum dune height is reached a few days after the occurrence of the peak discharge, when the water depth is already on the decrease.

Calculating Dimensions
Several methods have been developed for calculating dune dimensions, most based on the equilibrium situation in steady flow, e.g. Van Rijn (1982) [2]. In these methods, the time-lag effect is not taken into account. However, high dunes are mainly formed during flood waves, when the flow is unsteady. To calculate dune height in a situation like this the method of Wilbers (2004) is used [3], which is based on the ideas of Allen (1976) [4]. This method takes into account the time-lag effect in the development of river dunes in unsteady flows.

Dunes and Climate
Recorded and expected discharges, with the corresponding water depth, are used to calculate dune heights. The results show a significant influence of increasing river discharge on the development of river dunes in the Rhine. Figure 2 presents the development of dune height in January for base year 2003, and for the expected dune heights according to the UKHI scenarios for 2050 and 2100. A dune height of 1.2m is calculated for a peak discharge of 9,000m3/s in the base year. The UKHI 2100 scenario leads to a peak discharge of 14,000m3/s. The corresponding maximum dune height is 1.7m.
The time-lag is clearly perceptible in the figure: maximum dune height is reached later, while the peak discharge occurs on the same day. The maximum water depth at peak discharge is 14m for the UKHI 2100 scenario. When dune height is at its maximum, water depth has fallen to 12m, which is still large compared to the dune height of 1.7m. After this, dune height again decreases due to decreasing discharges. Despite the delayed response of river dunes to changing discharge, discharge does not decrease so quickly that dune height becomes substantial as compared to water depth. After approximately one week dune height is about 0.9m, while the water depth is still about 9m.

Height versus Depth
Figure 3 presents the calculated dune heights and water depths. This graph presents half the dune height because the hydraulic model computes the average water depth. The crest of a dune is assumed to be half a dune height above the average bed level. Local water depth is thus reduced by at most half the dune height. Figure 3 shows that when water depth decreases further (due to decreasing discharges), dune height falls very low, in the order of a few centimetres according to the calculation method of Wilbers. The height of river dunes is still not significant as compared to water depth.

Navigation Hindrance
Most vessels on the river Rhine require a minimum water depth of 4m. From these results it can be concluded that river dunes do not restrict the water depth for navigation during winter because low discharge does not occur very fast after a peak discharge. In other words, dunes get enough time to decay.
Figure 4 presents loading capacity of vessels in time, to provide a better insight into the hindrance of navigation. It is clear that navigation is restricted by low flows in summer, when discharge is about 900m3/s. The effect of climate change on loading capacity is clearly perceptible; in September loading capacity decreases from about 75% to about 55%. The influence of river dunes, however, would appear insignificant: river dunes are calculated to be about 0.05m high during summer. Restriction of loading capacity by river dunes is only in the order of 1% (see Figure 4, ‘with dunes’ and ‘without dunes’) for the base year (1991) as well as for the UKHI 2100 scenario. Although climate change is expected to lead to higher river dunes, these appear not to further decrease the loading capacity of vessels.

Concluding Remarks
In the present study expected discharges are calculated based on the UKHI scenarios. Therefore the discharge pattern of peaks and low flows in the Rhine remains the same. However, in reality climate change may cause different discharge patterns, which might lead to different results. On the other hand, the results in this research are based on the most extreme expectation of climate change, an increase in global average temperature of 4°C. For scenarios based on average or low estimates of temperature rise the discharges vary less and the effects will be smaller.
It can be concluded that river dune development in the Rhine is strongly influenced by changing discharges as a result of climate change. Higher discharges cause higher dunes and a longer time lag between peak discharge and maximum dune height. Climate change does influence the hindrance of navigation during summer due to a decrease in low discharges. However, dune height compared to water depth will always remain such that river dunes have no significant influence on the hindrance of navigation, and this effect is not enhanced by climate change.

References

1. Middelkoop, H., Van Asselt, M., Können, G., Van’t Klooster, S., Haasnoot, M., Van Deursen, W., Van Gemert, N., Kwadijk, J., Buiteveld, H., Valkering, P., Rotmans, J. (2001). Integrated water management strategies for the Rhine and Meuse basins in a changing environment. National Research Programme on Global Air Pollution and Climate Change (NRP).
   
2. Van Rijn, L. (1982). The prediction of bed forms, alluvial roughness and sediment transport. Delft Hydraulics report S 487 part III.
   
3. Wilbers, A. (2004). The development and hydraulic roughness of subaqueous dunes. PhD thesis, Netherlands Geographical Studies, Faculty of Geosciences, Utrecht University.
   
4. Allen, J. (1976a). Bed forms and unsteady processes: some concepts of classification and response illustrated by common one-way types. Earth Surface Processes, 1, 361-374.