Brian Finlayson

A global classification of river regimes

Abstract Global analysis of streamflow regimes is a neglected aspect of comparative hydrology. The regionalization of river regimes provides a useful tool for extrapolation of hydrological variables and for the identification of natural flow regimes where intensive river regulation has occurred. Previous studies at the global scale have used climatic characteristics to define regime types and the boundaries of climatic zones for extrapolation. These procedures have been made necessary by the lack of a suitable data set of stream flows with a global coverage. Such a data set has now been assembled for the first time and is used here to produce a classification of seasonal flow regimes based on a cluster analysis of the monthly mean flows expressed as percentages of the mean annual flow. Clustering was carried out using the within-group average method with cosine similarity measure after removing streams with no significant seasonal fluctuations. A set of algorithms in the form of a decision tree are presented. They can be used to place any stream into this classification and the process can be halted at any desired level of subdivision up to the 15-group level selected for global mapping in this paper. A preliminary world map of regime types is presented and is the first to be drawn on the basis of streamflow characteristics alone.

The relation of fluorescence to dissolved organic carbon in surface waters

Abstract A rapid method for the estimation of organic carbon in natural waters which could be adapted to continuous measurement is needed to improve water quality monitoring. The fluorescent properties of natural organic molecules offer such a method. The fluorescence of natural waters from a variety of surface sources, using an excitation wavelength of 365 nm and emission wavelengths in the range 400–600 nm have been correlated with total organic carbon levels. The results are highly significant statistically and show that fluorescence can be used as a predictor of TOC over a wide range of concentrations.

Distribution and hydraulic significance of large woody debris in a lowland Australian river

The line-intersect technique was used to measure the loading of large woody debris in a 1.8 km reach of the Thomson River, Victoria (catchment area of 3540 km2). A debris census (measuring every item present) was done over 0.775 km of this reach. The transect technique over-estimated the actual loading revealed by the census. The loading of debris ≥0.01 m in diameter for the total 1.8 km reach was 0.0172 m3 m−2, which is higher than that measured in many headwater streams in other parts of the world. The volume loading of debris measured from low level aerial photographs was only 4.8% of the value estimated by the line-intersect technique. The line-intersect estimates were biased due to non-random orientation of debris in the stream (causing estimated errors of +8% for volume loading and +16% for surface area loading). It is recommended that to avoid this problem, when using the line-intersect transect technique in lowland rivers, each line should comprise at least two obliquely-angled transects across the channel. The mean item of debris (≥0.1 m in diameter) had a trunk basal diameter of 0.45 m, a length of 7.4 m, and volume of 0.7 m3. The riparian trees and the in-channel debris were of similar dimensions. The debris tended to be close to the bed and banks and was oriented downstream by the flow at a median angle of 27°. Because of this orientation, most debris had a small projected cross-sectional area, with the median value being only 1 m2. Thus, the blockage ratio (proportion of projected area of debris to channel cross-sectional area) was also low, ranging from 0.0002 to 0.1, with a median value of 0.004. The average item of debris, which occupied only 0.4% of the cross-section, would have minimal influence on banktop flow hydraulics, but the largest items, which occupied around 10%, could be significant. Judicious re-introduction of debris into previously cleared rivers is unlikely to result in a large loss of conveyance, or a detectable increase in flooding frequency.

Water quality in mountain ash forests — separating the impacts of roads from those of logging operations

The purpose of the two catchment studies reported here was to allow the effects on water quality of road use and maintenance to be separated from the effects of a logging operation. In the first project, known as the Myrtle experiment, two small catchments in an old-growth mountain ash (Eucalyptus regnans) forest were chosen for a paired catchment study of the effects on physical and chemical water quality (baseflow and stormflow) of logging under a strict code of practice and with no roads crossing runoff producing areas. In the second project, known as the Road 11 experiment, the effect on sediment production from unsealed forest roads of vehicle use and level of road maintenance was assessed. The Myrtle experiment showed that the harvesting and regeneration operation did not have a major impact on the stream physical or chemical water quality. Increases were detected in turbidity, iron and suspended solids at baseflows, but these were small in absolute terms and of similar magnitude to the measurement error. The stormflow data revealed no significant influence of the logging operation. The suspension of logging during wet weather, the protection of the runoff producing areas with buffer strips and the management of runoff from roads, snig tracks and log landing areas eliminated intrusion of contaminated runoff into the streams, thereby avoiding the adverse effects of logging. The Road 11 study determined that annual sediment production from forest roads was in the range of 50–90 t of sediment per hectare of road surface per year, with approximately two-thirds being suspended sediment and one-third coarse material. The use of gravel reduced sediment production, provided a sufficient depth of material was used. Increasing the level of road maintenance with increasing traffic load controlled sediment production rates, but when maintenance was not increased, sediment production increased by approximately 40%. The results indicate that by identifying the areas that produce runoff it is possible to prevent contaminated runoff reaching the streams. Roads, on the other hand, produce large quantities of sediment, even when well maintained, so careful consideration of their placement and management is paramount.

Catchment-wide impacts on water quality: the use of ‘snapshot’ sampling during stable flow

Abstract Water quality is usually monitored on a regular basis at only a small number of locations in a catchment, generally focused at the catchment outlet. This integrates the effect of all the point and non-point source processes occurring throughout the catchment. However, effective catchment management requires data which identify major sources and processes. As part of a wider study aimed at providing technical information for the development of integrated catchment management plans for a 5000 km 2 catchment in south eastern Australia, a ‘snapshot’ of water quality was undertaken during stable summer flow conditions. These low flow conditions exist for long periods so water quality at these flow levels is an important constraint on the health of in-stream biological communities. Over a 4 day period, a study of the low flow water quality characteristics throughout the Latrobe River catchment was undertaken. Sixty-four sites were chosen to enable a longitudinal profile of water quality to be established. All tributary junctions and sites along major tributaries, as well as all major industrial inputs were included. Samples were analysed for a range of parameters including total suspended solids concentration, pH, dissolved oxygen, electrical conductivity, turbidity, flow rate and water temperature. Filtered and unfiltered samples were taken from 27 sites along the main stream and tributary confluences for analysis of total N, NH 4 , oxidised N, total P and dissolved reactive P concentrations. The data are used to illustrate the utility of this sampling methodology for establishing specific sources and estimating non-point source loads of phosphorous, total suspended solids and total dissolved solids. The methodology enabled several new insights into system behaviour including quantification of unknown point discharges, identification of key in-stream sources of suspended material and the extent to which biological activity (phytoplankton growth) affects water quality. The costs and benefits of the sampling exercise are reviewed.

Identification and explanation of continental differences in the variability of annual runoff

Continental differences in the variability of annual runoff were investigated using an expanded and improved database to that used in previous work. A statistical analysis of the data, divided by continent and Koppen climate type, revealed that continental differences exist in the variability of annual runoff. The variability of annual runoff for temperate Australia, arid southern Africa and possibly temperate southern Africa were noted to be generally higher than that of other continents with data in the same climate type. A statistical analysis of annual precipitation by continent and Koppen climate type revealed that differences in the variability of annual precipitation could account for some but not all the observed differences in the variability of annual runoff. A literature review of potential causes of continental differences in evapotranspiration resulted in the hypothesis that the significantly higher variability of annual runoff in temperate Australia and possibly temperate southern Africa may be due to the distribution of evergreen and deciduous vegetation. The process model Macaque was used to test this hypothesis. The model results indicate that the variability of annual runoff may be between 1 and 99% higher for catchments covered in evergreen vegetation as opposed to deciduous vegetation, depending on mean annual precipitation and the seasonality of precipitation. It is suggested that the observed continental differences in the variability of annual runoff are largely caused by continental differences in the variability of annual precipitation and in temperate regions the distribution of evergreen and deciduous vegetation in conjunction with the distribution of mean annual precipitation and precipitation seasonality.

Implications of flow control by the Three Gorges Dam on sediment and channel dynamics of the middle Yangtze (Changjiang) River, China

The impacts of a dam on the river downstream in terms of hydrology and morphology are determined by a complex mix of variables that includes the patterns of release of water through the dam and the characteristics of the downstream channel. Scour of the downstream channel is a common response because large dams cause a significant interruption to sediment continuity. Here we show that in the case of China9s Three Gorges Dam on the Yangtze River the outcome is complicated, as is commonly the case in large rivers. The downstream channel and floodplain system compose an area of long-term sediment accumulation and unstable channels with seasonally contrasting erosion and deposition patterns related to the migrating seasonal monsoon rainfall zones. In achieving one of the main purposes of this dam, that of flood control in the middle and lower basins, the pattern of flows released from the dam will closely resemble those seasonal flows that are responsible for channel instability in the middle catchment, thus effectively making erosive conditions the most common during a year. There is obviously concern about the ultimate impact of sediment storage in the dam on the dynamics of the delta and adjacent coast, and we show that this depends on the trajectory and duration of the erosive responses in the middle Yangtze basin. In this particular case, the outcome is of great significance to the well being of the densely populated riparian areas of the river.

Flood management on the lower Yellow River: hydrological and geomorphological perspectives

Abstract The Yellow River, known also as “Chinas Sorrow”, has a long history of channel changes and disastrous floods in its lower reaches. Past channel positions can be identified from historical documentary records and geomorphological and sedimentological evidence. Since 1947, government policy has been aimed at containing the floods within artificial levees and preventing the river from changing its course. Flood control is based on flood-retarding dams and off-stream retention basins as well as artificial levees lining the channel. The design flood for the system has a recurrence interval of only around 60 years and floods of this and larger magnitudes can be generated downstream of the main flood control dams at Sanmenxia and Xiaolangdi. Rapid sedimentation along the river causes problems for storage and has raised the bed of the river some 10 m above the surrounding floodplain. The present management strategy is probably not viable in the long term and to avoid a major disaster a new management approach is required. The most viable option would appear to be to breach the levees at predetermined points coupled with advanced warning and evacuation of the population thus put at risk.

Analysis of periodicity in streamflow and rainfall data by Colwell's indices

Indices of predictability, constancy and contingency which have been used to describe aspects of periodicity in biological phenomena are applied to monthly streamflow and rainfall data. The indices are first defined and the method of derivation described. The degree of correlation of the indices with each other, with record length and with the common statistical measures are investigated. Certain shortcomings are identified and some applications of these indices to hydrologic time series are suggested within the framework of their limitations.

Variability of Annual Precipitation and Its Relationship to the El Niño–Southern Oscillation

Abstract The importance of El Nino–Southern Oscillation (ENSO) to annual precipitation variability is assessed by dividing stations into regions of ENSO and non-ENSO influence as defined by Ropelewski and Halpert and then comparing these regions by Koppen climate type for significant differences in the zone average of the variability of annual precipitation. Of the 13 Koppen climate types that had 10 or more precipitation stations in both ENSO and non-ENSO influenced regions, 8 had significant differences in the zone average of the variability [defined as the coefficient of variation (Cυ)] of annual precipitation with the variability of annual precipitation being higher in the ENSO than the non-ENSO zone in every case. The range of increase in the Cυ of annual precipitation was generally 5%–25%. These results confirm earlier findings in relation to the variability of annual runoff and also are consistent with earlier findings in relation to the variability of annual precipitation. However, the impact of E...