THE EVALUATION OF THE MINIMUM FLOWS FOR THE BIOLOGICAL LIFE IN SAVIO RIVER NEAR CESENA (FO), NORTHERN ITALY
Rambaldi, A. (*), Rizzoli, M. (**), Patrizi, G. (*)
(*) Singea srl, Ravenna
(**) Bologna Province
1. PREMISE
One of the principal consequences of river water extractions is the variation in and stabilization of the instream flows.
In this context one assumes as a great importance the quantification of the "vital minimum instream flow" which one intends to mean the flow level necessary to guarantee the continuance of biological life in the river.
The most established and reliable methods for the determination of such parameters were discussed in a recent conference in Quebec, Canada (June, 1996) on the hydraulics of the river habitats.
These methods were developed predominantly in North America where regulations provide that each and every water extraction project be negotiated in order to obtain the best compromise between the project’s economic benefits and the safeguards of the river habitat. The concept of "vital minimum instream flow" for Italian rivers was introduced with Law 183/89, which specifies among the objectives of programming, planning and realisation of the Basin Plains, the "rational utilisation of the surface and deep water resources with the guarantee, however, that the entire extraction not damage the minimum constant vital instream flow in the riverbeds". This concept was restated in Law 36/94:
"Dispositions Relating to Water Resources", which provides that in basins characterised by withdrawals, the extractions are regulated in a way so as to not damage the equilibrium of the affected ecosystem.
The intention of the texts sited above is evident: one intends to maintain under control all extractions so as to avoid a condition in which they might approach the limit beyond which a crisis may result in the equilibrium of the acquatic biological community.
Some perplexing and uncertain interpretations are derived from the use of the adjectives "minimum" and "constant". Consider, for example, a river without a regulation system on the natural flow: after identifying the vital minimum instream flow, one can determine the amount of flow that may be safely extracted as a difference between the natural flow and said vital minimum instream flow. (Manciola et al., 1994).
In some rivers the natural flow could always be greater than that which is vitally necessary and for which the adjectives "minimum" and "constant" need be applied; on the other hand, cases, as for example, torrential rivers, in which the natural flow can be less than vital, the generalization would not be appropriate. In such cases, one part of a contained water volume could contribute to increasing the amount of flow, assisting it to approach the "vital minimum". In this way, the minimum vital instream flow is an extraction restraint and a necessary objective when considering the flow of a river.
To this objective one must point out that in some countries (USA, Britain and France) methodologies based on biological indexes have been developed for directly identifying the flows that give favourable assurances to the development of each fish species. In Italy, however, the problem has been confronted without expert scientific support and at different times with different criteria. For example, in the "General Plan Project for the Utilization of Public Water in the Province of Bolzano" requires that the minimum flows, which are determined case by case, should not be less than 2 I/sec per km2 of the river basin.
Another example, in the Piemonte Region, water storage containers receive authorization under the condition that, immediately downstream from the dams, sudden flows should be released which would be equal to 10% of the upstream natural flows. Finally, a River Po Basin Authority proposal (The Adjustment of Hydroelectric Concessions, Law 102, approved on 02.05.90) stipulates that the minimum vital discharge is to be 1.6 I/sec per km2 of the river basin, which is then to be multiplied by 4 coefficients, respective functions of the precipitations, altitudes, environmental quality and natural characteristics.
2. METHODOLOGICAL REFERENCES
The experience herein described refers to the application of "PHABSIM" which is one of the numerable widely used and reliable methods (Bovee, 1996) for formulating quantitative forecasts regarding this material; the method, developed by the United States Fish and Wildlife Service (Milhous et al., 1989) permits physical habitat simulation based on minimum vital flows criteria for the protection of aquatic life in rivers subjected to the imposition of stable state conditions on their flows. This approach (Trihey, 1979; Stalnaker, 1980; Bovee, 1982) is known also as the "Instream Flow Incremental Methodology," applicable also to the microhabitats.
The method is based on an evaluation of the physical habitat as a function of depth, velocity of the instream flow and river substrate-type, each being a function of the flows’ geometry and regime; one analyses, that is, the relationship between the flow, real or simulated, the physical usable habitat for the fish species and the vital phase determined as significant.
The system provides for the use of geometry based on a subdivision of the river into homogeneic tracts featuring certain morphological characteristics (shape, type of substrate, etc), described by way of opportune section-types, in correspondence of which are required the riverbed altimetrical (altitude measurement) profile and the relationship of the instream flow between the measurement of the flow and the hydrometric height (the water surface level altitude measurement).
In each case the sections are divided into vertical cells, which constitute the effective calculation reference, both for hydraulic simulation and biological simulation. The hydraulic simulation essentially consists in the calculation of the hydrometric heights and of relative velocities of simulated flows.
Given the objective desired, the biological component is of fundamental importance. The species and the vital phases considered as biological indicators of the river are chosen from among those highly representative of a designated area and as such are understood to be particularly sensitive to the stresses derived from water environmental modifications.
The relationships between the system’s variables (velocity, depth and substrate) and the exigencies of the targeted species permit the acquisition of suitability indicators that constitute the basis for usable habitat calculations. This is expressed as the weighted usable area (WUA), a function of the flow.
2.1. The hydraulic submodel
The hydraulic simulation consists in the calculation of two relationships between the flow (Q) and the hydrometric level (water surface level: WSL) and between the flow (Q) and the velocity (v). The first does not require a simulation at the single cell level because for every (Q) simulated in the section WSL is constant. WSL is determined by the average "best-fit" between the experimental relationship WSL (Q) and the equation’s theoretical curve
Q = a (WSL-WSL0)^(b )
in which:
WSL0 = hydrometric height for Q = 0; a , b = empirical coefficient
The second relationship is resolved by the Manning formula, separately applied to each section for calculating the average velocity of each cell:
vk = Rk^(2/3) SI^(1/2) nk^(-1)
in which:
vk = average velocity of Kth cell;
nk = Manning roughness coefficient of the Kth cell;
Rk = hydraulic radius of the Kth cell
SI = slope
2.2 The suitability curve
For evaluating usable habitat in a river, or, better, the impact that change in the water conditions can have on the aquatic relationships, it is necessary to develop suitable biological criteria for each important species in the water tract under examination; this information is provided by a system of calculations in the form of biological suitability curves.
The criteria taken in consideration are primarily limited to those variables that condition the distribution of the aquatic life and that are directly correlated to the hydraulic characteristics of the river, that is depth, velocity and substrate-type. The simulation of the habitat is based on the hypothesis that each species selects the most favorable conditions in a river, occupying first the space with the most suitable, and then, gradually, the other usable spaces.
The functions of suitability must be developed on the basis of experimental observation of the river or by analogy to similar environments, on the basis of direct experience or of bibliographical resources. Effectively, the functions are constructed by relating a suitability index, comprised of 0 and 1, which respectively correspond to the condition less suitable and that more suitable, with the significant variables mentioned above (velocity, depth and substrate).
2.3. The habitat submodel
The habitat submodel represents the process by which, the morphological system, the hydraulic stable state and the suitability functions become synthesized into one function that expresses the weighted usable area (WUA), expressed both in absolute value and in terms of percentage, related to the flows (Q).
The expressions by which WUA can be calculated are the following:
WUA(Q) = S Ak(f(vk)*g(dk)*l(dk)) (1)
WUA(Q) = SAk(f(vk)*g(dk)*l(dk))^(1/3) (2)
WUA(Q) = SAkmin (f(vk)*g(dk)*l(dk)) (3)in which:
Ak is the area of the Kth cell;
f(vk) is the suitability index relative to the velocity in the Kth cell;
g(dk) is the suitability index relative to the depth in the Kth cell;
l(sk) is the suitability index relative to the substrate in the Kth cell.
The WUA can be expressed both in absolute value, and in terms of percentage. The simple product of the suitability index (1) is used when one assumes that each factor has significance in the determination of the biological distribution, independently of its value. The geometric average (2) is used on the other hand when one assumes that all three functions are important in conditioning the quality of the habitat; but is, in any case, a compensation corrrelation between the three. Finally, the equation (3) is used if one considers that only the minimum function influences the distribution of the biological species.
3. THE CASE STUDIED
3.1. The geometry and hydraulics system
The sources of the Savio River are located near Verghereto in the Province of Forli, and empty into the Adriatic Sea at Lido di Savio after having flowed its last kilometers in the Province of Ravenna. The studied tract (Fig.1) is relatively homogeneous, especially from a hydraulic point of view: the principle tributary, the Borello Stream, converges upstream to the first hydraulic section herein considered, while all the successive affluents, mainly from the left, are due to small ditches and charactensed by a strictly hydrologic stable state. The fundamental hypothesis is, therefore, that in this zone there exist no significant flow variations between the initial sections and the final sections studied. The system is, however, heterogeneous from a morphological point of view: the first part, in fact, is characterised by a rectilinear flow, with low water; and, at times, the substrate at surface level.
A Savio River tract with deep and slow water meanders for a short distance, then again it follows a rectilinear tract, until finally, it enters the city of Cesena. The tract pertaining to this study extends from Molino di Cento to a bridge on Emilia Street in Cesena, a section of about 12 Km. The fundamental geometric parameters for the morphological description of the river are taken from the most recent topographic survey and constitute a longitudinal profile and 12 cross-sections (Tab. 1).
Reach Length (m) Slope( 10-3) Manning’s coefficient
1 521 2,0 0,033
2 401 2,5 0,033
3 375 0,5 0,033
4 488 4,5 0,040
5 658 3,4 0,040
6 225 5,0 0,040
7 428 5,0 0,040
8 515 1,2 0,025
9 648 1,5 0,033
10 540 1,5 0,025
11 640 3,3 0,033Tab.1-. Characteristics of the studied sections.
The nature of the floor of the river bed (Tab.2) is of particular importance for the definition of habitat suitability in relation to the aquatic biological associations because the organisms are very selective in their requirements, above all, in the more delicate biological phase: that of reproductivity. The description reflects specifications presented by Bovee (1978).
Section Description
1 Silty banks and gravel in the central bar
2 Rock without sediment
3 Rock without sediment
4 Silty banks and gravel in the middle of the stream
5 Silt and vegetation
6 Silt and vegetation
7 Silt and vegetation
8 Silty banks and sandy in the central bar
9 Silt and sandy silt
10 Sift and sandy silt
11 Silt and sandy silt
12 Silt and sandy silt
Tab.2- River bottom substrate.
For the hydrologic characterisation reference is made to the downstream data relative to the Savio River at S. Vittore, published by the Italian Hydrographic Service, from which have been calculated the average monthly relationships for the years 1922-43, 1946, 1948-72 (Tab.3):
month flow
G 15,56
F 20,09
M 18,76
A 12,69
M 8,63
J 5,15
J 1,87
A 1,24
S 2,09
0 5,48
N 12,26
D 16,26Tab.3- Average monthly flows (m3/s).
In every case a deficiency exists in direct surveys of other significant variables and in particular, the instream flow velocity.
The utilised procedure permits one to estimate the necessary coefficients, a and b , to define the relationship between the flows and the water surface level using the Manning equation. The Manning coefficients were assigned on the basis of the morphometric characteristics of the river. The last step of the calculation consists of the definition of the hydrometric height (water surface level) and the relative velocity in each sections’ cell. One example of the results obtained is presented in Fig. 2 in which are indicated the hydrometric velocities and the heights relative to one of the sections studied.
3.2. Suitability curves
The system of biological characterisation represents a delicate and critical point in the development of simulation models for the quality of river habitat in relation to flows. The suitability curves graphically represent the level of acceptance that each fish species shows in relation to the various environmental conditions examined. Presently, for the fish fauna of Italy, there are no known significant examples of the curves’ construction based on experiments. In anticipation of such, however, to be able to arrange accurate information by way of a focussed research program, theoretical suitability curves have been constructed utilizing the information on environmental tendencies furnished from multi-year analyses of data on the rheophile ichthyic fauna in the Bologna Appennine. Specifically, the data relative to samples collected between 1987-1995 of Reno River Basin fish were analysed.
The Leciscus cephalus and the Barbus barbus plebejus were the targeted species chosen to characterise the Savio River near Cesena. More diffuse and less demanding, the L. cephalus has proven to be the most abundant species in this zone; the B. barbus p., itself highly represented there, is generally understood to be more selective of its environment.
For each species, three critical vital phases have been considered: that of reproductivity, applicable to the moment of egg-deposition; that of offspring growth and development during the first months after birth; and finally, that of the entire conditions which characterise the juvenile and adult life. The various curves relative to each phase of the vital cycle should be utilized as a reference for stipulating the regime of the river’s instream flow during a determinant critical period. The curves relative to the reproductive phase will represent the referential elements for the months of April and May; while those relative to the conditions of offspring growth and development will be applicable to the period June-October. For the remander, it will be sufficient to refer to the curves relative to the juvenile or adult stage.
For all the species and for each vital phase, there were constructed three suitability curves relating to the environmental fluctuations: water depth, current velocity and substrate-type. The curves relative to the L. cephalus (Fig. 3) evidence a limited selectivity during the juvenile and adult stage in confronting the conditions of the environment. The reproductive and offspring stages, on the other hand, indicate an extreme sensitivity to the substrate-type and to the current velocity; while to all the other factors they seem to adapt very well.
In Fig. 4, the curves relative to the B. barbus p. confirm the greater environmental selectivity of these species which appears to be very sensitive at each vital stage to the substrate type. An important role that was fully scrutinized was the depth factor that appears to have a greater influence on the reproductive and offspring stages.
4. RESULTS
Table 4 indicates the evaluation of the usable habitat.
Month Flow L. Ad. L. Fry L. Spa B. Ad. B. Fry B. Spa.
1 15.56 22.0 1.1 1.6 6.0 0.3 0.8
2 20.09 22.2 0.1 1.5 6.3 0.2 1.4
3 18.79 22.1 0.3 1.5 6.3 0.2 1.3
4 12.69 21.0 2.0 1.6 5.7 1.3 0.4
5 8.63 21.0 2.1 1.0 5.9 3.2 0.0
6 5.15 22.9 1.9 0.2 6.2 1.5 0.0
7 1.87 14.2 15.9 0.0 2.8 7.0 0.0
8 1.24 8.8 20.0 0.0 1.5 15.5 0.0
9 2.09 15.8 14.0 0.0 3.3 5.4 0.0
10 5.48 22.9 1.6 0.3 6.3 1.7 0.0
11 12.54 20.9 2.1 1.6 5.8 1.4 0.4
12 16.26 22.1 0.9 1.6 6.1 0.3 0.9Tab.4 - WUA usable water area expressed as % of the total area.
The first important consideration ecologically speaking is that of the two species’ acceptance of the location: in fact, one notes that the usable habitat for the L. cephalus is systematically superior to that of the B. barbus p. and this difference is especially high in the adult stage, with the average percentage of usable area equal to 19.7% for the L. cephalus; and 5.2% for the B. barbus p. This result is confirmed by experimental observations based on the presence of the two species in the tract under consideration, valued highly by one and scarcely by the other (AA.W., 1992). In the light of the above considerations, inclusive of the hydrologic character and the biological character, this difference is attributable to the importance that the river bed substrate assumes in the determination of the usable area by the B. barbus p., because of its low tolerance to modifications of this parameter. In fact, the river bed conditions most suitable to this species (rocky, gravelly, or at least, sandy) is provided only at a few sites (Tab.2).
Also, as regards to the offspring, in the months of July, August and September, the usable area for the L. cephalus is superior to that for the B. barbus p., even though the difference in sensitivity is less, given that the ecological valence is enough similar for the two species’ offspring.
An interesting result is determined by the identification of the flow conditions that guarantee the maximum usable area for each species.
In Fig.5, relative to the adults of both species, it is evident that for flows less than 5 m2/s the usable habitat quickly decreases, while the progressive increases of usable area remain almost constant. This flow value, in fact, seems to guarantee area of maximum use for the adult L. cephalus and B. barbus p.. This behaviour can be connected to the river bed morphology; that is, in the presence of chanalized rivers the increase of flow does not increase the usable habitat for the fish fauna.
In reference to the average hydrologic regime of the Savio River, the conditions most critical for the adult stage are verified in the months of July, August and September when the summers low water level reduces the river’s flow to the point of 1-2 m3/s (Tab.4).
The trend of the "usable area-flows" curves, relative to the offspring stage for both species permits one to specify (Fig. 6) a maximum of usable habitat corresponding to the approximate flow 1 m3/s, while for other conditions of instream flow the area seems to suffer a drastic reduction almost to the point of non existence. Keeping in mind that summer months are those most important for the offspring, one must point out that the natural hydrologic conditions guarantee these offspring the maximum usable habitat. As regards to the reproductive phase it is not possible to specify a critical flow value.
The indicated flow curves in Fig.7 evidence the increase of usable area as the flow increases, with values that, however, only slightly surpass 1.5% of the overall area. This result, apparently, not consistent with those preceding, in reality, contributes significantly to the microenvironmental characterisation of this specific tract of the Savio River, indicating that this studied zone one may connote as one of marginal reproductivity for the targeted species, while it presents particularly ideal characteristics for the growth and development of their offspring and for the settlement of their combined populations.
Reiterated, one may connote this zone as one for establishing and growing offspring, juvenile and adult populations of B. barbus p., and L. cephalus that will use as a zone of reproduction that part of the river more upstream. It becomes, therefore, particularly important to give attention to the protection and revitalisation of the fish fauna’s free circulation in the zones immediately upstream. Successively, the minimum flow criteria was compared with the hydrologic regime of the Savio; this aspect of the analysis is particularly important for rivers that, like the Savio, have relatively small cachment areas and an hydrologic regime strictly provided by rainfall.
For example, indicated in Table 5, is the frequency of low-water level flows less than a 0.2 m3/s in the periods 1937-43, 1946,1948-72 (Singea, 1988, b), expressed by the days in which they were measured: the frequencies appear substantially casual and the water resource demand for channels cannot be separated from the effective usable river water.
Year Days of occourrence
1962 67
1943 62
1956 44
1946 43
1952 41
1958 40
1971 36
1957 31
1970 21Tab.5- Frequency of low flows (< 0.2 m3/s).
To provide, in part, to the complete hydrologic analysis it is acceptable to use the data of the daily average instream flow of a significant year among those published by the Italian Hydrographic Service for the Savio River at S.Vittore; the year utilised is 1968 and Tab. 6 indicates the characteristic elements.
1968 1937-43
1946
1948- 71
Qmax (m3 s-1) 198,0 469,0
Qave(m3 s-1) 10,2 10,0
Qmin (m3 s-1) 0,31 0,02
Qave (l s-1 km2) 17,1 16,8
Outflow (mm) 536,5 528,8
Inflow (mm) 1124,7 1100,7
Outflow ratio 0,48 0,48Tab.6 Hydraulic parameters of Savio River (S.Vittore station).
The average daily flows are represented by the time-flow diagram in Fig. 8 with monthly average flow for the period 1922-72. On the same graphic are indicated also the recommended flows for adults and offspring in significant months for each vital stage. It is indicated from the analysis that at times natural flows are less than the minimum vital instream flow and therefore they indicate critical conditions.
5. CONCLUSIONS
The experiences described above permits one to affirm with a high degree of reliability that the proposed methodology indicates the minimum vital instream flow for a river like the Savio with its indicative hydraulic, hydrologic and biological characteristics. Without a doubt the degree of approximation of the model increases with the increase in precision by which the system-imput experimental data are defined: the data of which pertains to the geometry and hydraulics of the system, as well as, to the ecological needs of the species specified as biological indicators of the tract studied. The description of the system’s geometry consists of the morphological survey of the sections’ characteristics and of the longitudinal profile of the tract studied. The hydraulic system is defined, on the other hand, by way of determining, in correspondence to each section, the instream flow curves and by a series of measurements of the current velocity.
Of particular importance, in the final analysis, is the definition of the suitability curves, relative to the level of acceptance by the fish species under examination, which must be collected by way of accurate and focussed research programs. One mantains, therefore, that the application of the method has to be developed, not only for the evaluations of the impact of large hydraulic plants on the river, but also for the determination of the impact of water withdrawal, above all, for that which is for agricultural use, which, individually, have scarce impact, but as an entire network can cause an acute water crisis to the point of endangering the continued existence of the river.
BIBLIOGRAPHY
AA.W., Basic Elements for the Predisposition of the Regional lchthyic Map, Vol. 1 and 2, Emilia-Romagna Region, Food and Agriculture Assessor, Bologna, 1992.
AA.W., Second International Symposium on Habitat Hydraulics. Proceedings. Quebec, Canada, June 1996.
Bovee, K.D., Probability-of-Use Criteria for the Family Salmonidae: Instream Flow Information Paper, n.4, 88 pp, Fort Collins, GO, 1978.
Bovee, K.D., A User Guide the Instream Flow Incremental Methodology, U.S. Fish and Wildlife Service, Biological Services Program, FWIOBS-80-52, 1982
Bovee, K.D., Perspectives on Two-Dimensional River Habitat Models: thePHABSIM Experience. In the Second International Symposium on Habitat Hydraulics. Proceedings. Quebec, Canada, June, 1996.
Cotta, Ramusino, M., Crosa, G., Forni, A. & Marchetti, R. The Po River As An Example of Environmental Quality Evaluation of Rivers Subjected to Extraction.Acts of the Convention OThe Quality of th Water of th Po River in the I 9900s0. The Water Research Institute Journal, 1992.
Crosa, G., Cotta, Ramusino, M. & Marchetti, R., Determination of the Necessary Minimum Flows for the Protection of Aquatic Life in Rivers Subject to Extraction and Retention. Water-Air, Volume. 7, pp.839-850, 1988.
Idroser SpA, The Regional Hydrologic Journal, in; Plan Project for the Best Safeguard and Use of Hydric Resources in Emilia-Romagna, vol.VI, 6 sections, Bologna, 1978.
Manciola, P., Martini, P. and Ubertini, L., The Meaning of Minimum Vital Instream Flow and its Determination for the Rivers of Central Italy. Hydrotecnics, Vol 2, pp.79-91, 1994.
Milhous, R.T., Instream Flow Values as a Facror in Water Management and Energy: Tecnical and Policy Issues, American Society of Civil Engineers,pp.489-494, N.Y., 1983.
Milhous, R.T., Updike, M.A. & Schneider, D.M., Physical Habitat Simulation System Reference Manual, Version II. National Ecology Research Center, Fish and Wildlife Service, Instream Flow Information, Paper n. 26, Fort Collins, CO, 1989.
Minister of Public Works, Hydrographic Service, Hydrographics Annals, The State Institute of Poligraphics, Rome, 1976.
Singea, Feasibility Study For An Urban Hippodrome Park Basin. Ravenna, 1988.
Stalnaker, C.B., The Use of Habitat Structure Proterenda For Establishing Flow Regimes Necessary for Maintenance of Fish Habitat. The Ecology of Regulated
Streams, Plenum Publishing Corporation, pp. 321-337, 1980.
Trihey, E.W., The IFG Incremental Methodology, in: G.L.Smith (ed.), Workshop In Instream Flow Habitat Criteria and Modeling. Colorado Water Research Institute. Colorado State University Information Series 40, Fort Collins, CO: 24-44, 1979.