Paper (rus)

On the Hypotheses Regarding
the Causes of Channel Formation

A. N. Kondrat'ev

Received March 14, 2000

In modern science dealing with channel processes, the carrying capacity of flow [1] and the ratio of the elevation of the maximum channel-forming flow rate to the floodplain elevation [2, 3] are considered the principal channel-forming factors. In fact both these factors can explain the same types of channel processes (the same types can be placed along both channel-forming axes, some of them being placed better and some, worse). The question arises as to which types of the channel processes cannot be accounted for if only one of the channel-forming factors is taken into account? First, a change in the flow carrying capacity of sediment discharge does not initiate the development of floodplain branches nor does it bring about the formation of multibranch system in the floodplain, while the rise of the elevation of the channel-forming flow rate above the floodplain elevation initiates the development of floodplain branches [2]. Second, the multibranch pattern cannot be associated with the relative positions of the maximum of the channel-forming flow rate and the "alluvial islets" line but can be easily derived from the channel being overloaded with sediment.

A contradiction in this case consists in that different governing factors are chosen for the same types of channel processes. On the other hand, each such channel-forming factor has its benefits. Therefore, it is good reason to consider these factors in combination. To do this we place all the types of channel processes on both the axes. The horizontal axis will now represent the degree of floodplain inundation, while the vertical axis will represent the carrying flow capacity (Fig. 1). The relative carrying flow capacity increases downward.

Now two channel-forming factors are considered for lowland rivers with a wide floodplain. Each of these independent factors has its effect on the type of channel processes. The first factor---the relative carrying capacity of the flow---increases along the following series of the channel process types: multichannel pattern, band--ridge type, shoal process, limited meandering, free meandering. The second factor---the ratio of the elevation of the channel-forming flow rate to the floodplain elevation. When the channel-forming level is lower than the floodplain elevation, the above series of the channel process types will take place (in a single channel), and when this level is higher than the floodplain elevation, the same channel process types will take place in a multichannel case.

It is worth mentioning that the order of these channel process types in the first column is the same as that in the classification proposed by N.E. Kondrat'ev and I.V. Popov and in the classification proposed by B.F. Snishchenko [1]. The contradiction between these classifications emerges when braided rivers are introduced. The second column begins with the type, which corresponds to a combination of the floodplain and channel branching. As to its carrying capacity, this type corresponds to the channel branching (it is overloaded with sediment as well), while the conditions of branch formation correspond to the floodplain branching type. Such type of channel processes can be observed on the Amur River at Khabarovsk. This type is in contradiction with both the above classifications, in which the channel and floodplain branching patterns are placed at the opposite ends of classification.

The proposed two-factor scheme of channel process types (Fig. 1) permits prediction of changes in the type that can be caused by variations in the channel-forming factors. For example, when the flow is overloaded with sediment, according to Fig. 1, we can expect a change in the channel process and the formation of islands. Such prediction is confirmed by the processes observed in the lower reaches of the Zeya River, whose right-hand bank has approached the Belye Mountains composed of sandstone. Scouring the mountains, the river became overloaded with sediments. The type of channel processes has changed within a reach with a length of several tens kilometers; the previously meandering channel became straight, and a large number of islands formed in it. On the other hand, when the carrying capacity increases, according to Fig. 1 we can forecast the opposite process to take place (bank scouring, meandering) as was the case with the Mississippi, where a large number of meander scrolls were straightened.

According to Fig. 1, when the level of channel-forming flow rate is higher than the floodplain elevation, an improvement of the conditions for the floodplain branch formation can be predicted (for example, when the lower reaches of the Tulva River became subject to the backwater effect of the Votkinskii Waterworks Hydraulic Power System, the free-meandering type of channel process changed to the braided river type (following the type of floodplain branching)). Conversely, when the channel-forming level drops and secondary branches disappear, gradual formation of a single channel can be expected. The advantages of Fig. 1 in solving both scientific and practical problems are obvious.

Commonly meandering is divided into limited and free. Figure 1 shows that the limited and free meandering differ in the relative carrying capacity of the flow. Accentuating the different origin of the meandering types, it may have been better to call these types in some other way (for example, developed and undeveloped meandering). The different types of meandering should also be divided in accordance with the extent of influence of the limiting conditions. A river with high relative carrying capacity is able to develop following the type of free meandering. However, if this river flows in a narrow valley, the limited width of the belt of meandering allows it to develop only within limited meandering type. The opposite situation is also possible---a river with the carrying capacity lower than the amount of sediment, which enter it, will develop following the scheme of limited meandering even within a wide floodplain where there is enough space for developed scrolls in the case of free meandering.

This conclusion allows us to give positive answers to the following questions: can limited meandering exist without limiting factors and a direct channel form without limiting factors (Fig. 2)?

Therefore, the two terms ("limited" and "free") are not sufficient to characterize meandering as a process. The hydromorpholigical theory [1] considers only the options corresponding to Figs. 2c and 2d. A possible solution may be to use dual classification: by the degree of limiting (limited or free meandering) and by genetic cause (for example, developed and undeveloped).

The conclusion is that the limiting conditions are the third independent channel-forming factor. At the same time, one should not take into account the floodplain-to-channel width ratio as the sole controlling factor for all the types of channel process. Rivers with wide floodplains flowing in wide valleys may be not meandering, but straight and braided as well.

Let us consider the two-factor table of the morphodynamic types of channel processes suggested by R.S. Chalov [3]. This means that the channel types considered are controlled by two channel-forming factors. Thus, the vertical axis represents the limiting condition, that is, the floodplain-to-channel width ratio. Indeed, a wide channel can contain either of straight, meandering, and braided channels. The horizontal axis represents meandering, straight channels without branching, and braided rivers. Figure 1 shows that the controlling factor for this series of types is the relative carrying capacity of flow. In this table, the types are ranked in accordance with two channel-forming factors, i.e., the relative carrying capacity and the limiting conditions.

The next step is combining three channel-forming factors, i.e., the relative carrying capacity, the relative inundation of the floodplain, and the relative width of the floodplain. The obtained cube of channel process types characterized by the three channel-forming factors is difficult to represent on the plane. Figure 1 presents a two-factor table obtained as a projection of the three-dimensional cube of channel-forming factors on two axes; the vertical axis represents the relative carrying capacity and the horizontal axis represents the relative inundation of the floodplain. The two-factor Chalov table [3] is a projection of the same three-dimensional cube onto the plane with two other channel-formoing factors, i.e., the relative carrying capacity and the relative width of the floodplain.

The approach based on combining hypotheses is fruitful [4]. It allows the use of the best of each hypothesis and avoid the effect of their faults. Combining the hypotheses for solving the problems of channel process type formation allows us to forecast the changes in the channel process types caused by changes in several independent factors.

REFERENCES

1. Kondrat'ev, N.E., Popov, I.V., and Snishchenko, B.F., «Osnovy gidromorfologicheskoi teorii ruslovogo protsessa» (Fundamentals of the Hydromorphological Theory of Channel Process), Leningrad: Gidrometeoizdat, 1982.

2. Makkaveev, N.I., «Ruslo reki i eroziya v ee basseine» (River Channel and Erosion in Its Basin), Moscow: Akad. Nauk SSSR, 1955.

3. Makkaveev, N.I. and Chalov, R.S., «Ruslovye protsessy» (Channel Processes), Moscow: Mosk. Gos. Univ., 1988.

4. Mitrofanov, V.V., «Ot tekhnologicheskogo braka do nauchnogo otkrytiya» (From Technological Waste to Scientific Discovery), St. Petersburg: Assotsiatsiya TRIZ Sankt-Peterburga, 1998.

FIGURES

Fig. 1. The types of channel processes in wide-floodplain rivers presented as a table in accordance with the contolling channel-forming factors. The first column represents the channels without branching that form in floodplains with low inundation ((a) alluvial islet type; (b) shoal type; (e) limited and (g) free meandering); the second column represents branching channels that form in floodplains with high inundation ((b) alluvial islets in branching channel; (d) shoal type in branching channel; (f) interrupted (uncompleted) meandering; (h) meandering of the branches in a branching channel).

Fig. 2. Types of meandering as a function of controlling factors. In the case of low relative carrying capacity, (a) undeveloped meandering in a narrow valley on narrow floodplain (this type could be called limited since there is a limitation, but it is not a factor to determine the type); (b) undeveloped meandering in a wide valley on a wide floodplain (no limitation). In the case of high relative carrying capacity, (c) potentially free meandering in a narrow valley (limited meandering, that is, the limitation exists and has its effect); (d) free (developed) meandering in a wide valley (this is true free meandering).

Kondratyev A.N.

Parkovaya Str., 8, 28., Ilichovo, 188 838, Vyborgsky region, Leningrad area.

ancondratyev@peterlink.ru                      http://bedload.boom.ru