Grandi Laghi Lombardi - FLOOD
MANAGEMENT POLICES FOR VERBANO RESERVOIR TO REDUCE FLOOD ISSUES IN VERBANIA CITY
A group project research was finalized for Natural Resources Management and Integrated Water Resources Management class, in Spring semestr 2015 at Politecnico di Milano and it is presented in this wiki page. Grandi Laghi Lombardi (GLL) Basin is located in a Northern West region of Italy and involves several water bodies, one of them is Lake Maggiore(or Verbano) that is the second greatest Italian lake with regard to area and volume (Premazzi, et al., 2003) and is located in Italian Lombardy, Piedment regions and Switzerland Ticino district. The lake is fed by Ticino River and Tresa River and the inflow is stored in Verbano Reservoir in order to provide water to downstream demand sites.
The changes in level of the lake causes flood issues in some cities that have a shoreline such as Belgirate, Stresa and also in Verbania city that is taken into consideration within the scope of this report due high historical flood events in the past (Coluccino, et al., 2012). The purpose of this project is reduce the flood area by analyzing previous datas using Water Evaluation and Planning (WEAP) software (SEI, 2007). Three different scenarios were run in GLL model and were evaluated through performance indicators using Matlab (MathWorks,2015), finally other stakeholders indicator were evaluated to analyze and visualize the conflict among all stakeholder using DiscoveryDV software (Kollat, et al., 2013). The outcome of this study will be the preference scenario according the trade-offs with other water user involves in this river basin.
Lake Maggiore(or Verbano) receives 1800 mm water, that is substantial rainfall regime is in Lombardy region, in a year in the period of May and November (Premazzi, et al., 2003) An alpine watershed of Lake Maggiore that comprises 210 km2 surface area is approximately 6,600km2 (Guariso, et al., 1985). The lake is full when the snow melts and autumn precipitation occurs and it provides sufficient water supply in order to irrigate large areas in Lombardy and Piedmont, and to meet the hydroelectric power production demand (Guariso, et al., 1985). In addition to this, a population that live in the Lake Maggiore area is divided into three parts (Figure 1): The Lombardy side (55,000), Piedmont (72,000) and Switzerland (41,000) (Engelhardt, n.d.). The municipalities which has the most population is Verbania (30,000) in Italy and Locarno(15,000) in Switzerland (Engelhardt, n.d.). Moreover, when the damages of flood events are compared between the Italian and Swiss parts, Italian territory is more suffered inundation issues according to (Figure 2) since building the mobile dam on the lake (1943), agreeable management of it is required due have an international lake characteristics (Betti, et al., 2013).
Figure 1. Lake Maggiore Map
Figure 2. This indicates the flood damages arising from Lake Maggiore. (Guariso, et al., 1985)
According to the STRADA Project in Verbano Cusio Ossola’s Province (Progetto Strada Azione 6.3 Scenari di evento e di vulnerabilita) the most significant flood issues on Lake Maggiore shores occur in Verbania city (Figure 3) that has three major areas that get caught more inundation problems (Figure 4) (Coluccino, et al., 2012). Thus, the report focuses on this municipality and presents proposed management releasing policy and recommendation for this area.
When flood event occurs, the water crosses the threshold defined and an area is covered with high amount of water that arise from a water bodies like lakes, affecting the territory adversely (i.e Pallanza flooded area in 2000 (Figure 5) (Coluccino, et al., 2012). Overflow water from the lake affects people who live on the lakeshores directly in some ways: loss of human life, physical injuries, infrastructure issues and damage to material goods, roads and bridges. The goal of the project is to reduce maximum flood extension area in city of Verbania that has a shore on Lake Verbano, is the upstream of the lake with regard to the dam (Betti, et al., 2013) and it is also capital city of Provincia Verbano-Cusio-Ossola. To achieve this purpose, WEAP software is used for creating and simulating scenarios for 5 stakeholders that are flood reducing, environmental, agricultural, hydropower generation and transportation and DiscoveryDV (Kollat, et al., 2013) is also used to visualize of trade-offs and conflicts among stakeholders and to choose one alternative. In addition, Matlab (MathWorks,2015) is used to obtain for all stakeholders’ performance indicators.
Figure 3. The lines show the function with regard to flooded areas and lake levels. (Coluccino, et al.,2012)
Figure 4. The three main towns that have inundation issues in city of Verbania. (Coluccino, et al.,2012)
Figure 5. Flooded area on Verbania city’s shore on October of 2000 (Coluccino, et al.,2012)
There are lots of researches that have been done about flooding issues on Lake Verbano shores including Verbania city. One of them is Strada Project introduced before. In addition to this, other projects have studied about flooding like Verbano Project, including users that are hydropower production and irrigation, the control of floods and the conservation with regards to environmental, is done to manage the water resource sustainably and in order to come to an agreement as increase the stakeholder engagement (Betti, et al., 2013).
Another previous research has been done to minimize the impacts of flooding on Lago Maggiore basin that has suffered extreme inundation issues between 1993 and 2000, a EU founded project that is called RAPHAEL was planned to specifically investigate the current potential of flood forecasting by coupling high-resolution weather prediction models with distributed hydrological models in complex terrain (Jasper, et al., 2002). In addition to this, one study related to warn people before flooding and protect them from flooding issues on Locarno shores due to increasing Lake Verbano level has been done (Cannata, et al., 2013).
Federal Office for the Environment (FOEN) of Switzerland forecasts the risk level of all the lake and river which include the Lake Verbano (FOEN, 2015), and makes a flood alert map (Figure 6) that includes five levels for river and lakes of national interest and allow people to check the possibiliy of flood occurance everyday (FOEN, 2015).
Figure 6. Flood alert map for rivers and lakes of national interest
Area of interest
The area of interest is Verbania city, located in the northern of Italy in the shoreline of the Lake Maggiore. is the first city upstream the Verbano reservoir and have an elevation of 197 m.a.s.l. The Figure 7 shows the location in the Grandi Laghi Lombardi Model.
Figure 7. Area of İnterest in Grandi Laghi Lombardi Model – Verbania city
Basic legal framework for flood risk in Italy and Lombardy Region
The legislative framework of flood issues in Italy is content in the Directive 2007/60 and includes the management and assessment in the European territory. The aim of this Directive is establish the procedures to reduce the consequences of inundation in all the humans´ aspects (Poretti, 2010).
The main objective of this project is model and manage some of the reservoir releases polices to reduce the potential flood area on Verbania city, located in the northern of Italy in the shoreline of Lake Maggiore. This area experimented flood events in the past, and due to the location and hydraulics/hydrology behavior in the zone, could take place again in the future. Three scenarios were modeled and compared with the base scenario, which is the scenario with the current conditions of the Grandi Laghi Lombardi River basin.
We defined four indicators related with the flood performance in Verbania City (Reliability, maximum flood area, maximum flood duration and flood vulnerability), to achieve our goal (Reduce flood area in Verbania city). This four performance metrics were calculated using the results of WEAP model of daily data of Grandi Laghi Lombardi.
WEAP software does not allow to model flood events, so in order to understand flood events in Verbania city with the WEAP model, we analyzed the following considerations: Verbania is located upstream of Verbano reservoir, so we needed to focus on the inflow from Ticino and Tresa rivers, which was stored in the Verbano reservoir, and the releases from the reservoir to meet downstream stakeholders water demands (Hydropower generation, agricultural irrigation and minimum environmental flow). The releases summed with the storage indicated the available water. As this project interest area is city of Verbania and threshold depends on the the location (Betti, et al., 2013), the threshold of lake level for Verbania city is 195,5 m.a.s.l, was established according with the study done by the researchers of the center for Geoscience and the Institute of Ecosystem in Verbania Pallanza, they analyzed flood events in the period of 1965-2006 and the levels reached all those years in Verbania (Kampf, et al., 2012). Moreover, by the elevation-volume curve of Lake Maggiore we obtained the relationship between the water level and the volume (Figure 8), so the volume necessary to reach flood is 634 million m3.
So, as it seems daily river discharge is an alternative definition of flood if:
- [Outflow/release downstream million m3] + [storage millions m3] > 634 million m3 Flood
- [Outflow/release downstream million m3] + [storage millions m3] ≤ 634 million m3 No Flood
Figure 8. Elevation - Volume curve of Lake Maggiore
Once we understood the process to analyze flood, the flood indicators were calculated in Matlab (MathWorks,2015) as described below.
Maximum Flood Duration
Indicates maximum number of consecutive days of flooding. After calculated the flood events with our considerations mentioned above, we identified 25 flood events. Thus, in order to know the days with consecutive flood we defined a logical vector indicating 1 for flood events and 0 for not flood events, then we made the difference between them and the flood duration is the sum from 1 to -1.
Maximum Flooded Area
Indicates maximum flooded area in Verbania. To obtain the maximum flooded area value, we used the volume-elevation curve of the Lake Verbano Reservoir from WEAP to get the daily level of the lake, the data missed were obtained by making a linear regression of this curve (Figure 8). An interesting report was presented by the Politecnico di Milano in 2013 about the research program of Lake Maggiore, in the report they presented the relationship between the level of the Lake Maggiore and the flooded area in Verbania city (Anghileri,et al., 2013), with this information we interpolated some data, graphed the curve and founded the tendency polynomial regression that best fitted (Figure 9). Finally, for each day with flood we related the area with the level previously calculated in Matlab and obtained the maximum flooded area.
Figure 9.Elevation Lake Maggiore level vs. flooded area in Verbania city
Reliability is the ratio between number of days when there is not flooding and the total number of days. As we calculated before, during the period 1974- 2010, there were 25 days flood events. To verify our calculation, we checked the flood events data during the period of 1965-2006 from journal (Kampf, et al., 2012). The total days in the data series in the Grandi Laghi Lombardi is 13505 days.
Flood vulnerability indicates average flooded area when flooding occurs. We calculated the flood vulnerability by summing all flooded areas in the time series and dived by the total of flood days.
Method used to evaluate alternatives against the stated performance indicators
To evaluate the alternatives, we observed the behavior of previous flooding events in base scenario. We realized that almost all the inundations occurred in the months of October and November as a results of the rains. To illustrate the seasonal comportment of the Verbano reservoir, we considered the inflows of one year, it is clear that from April to June is higher due to the smelting snow in summer, but inflow is instantly increased on October and November because of high precipitation (Figure 10). The precipitation is a phenomena very difficult to predict but very important to consider, nevertheless was unconsidered in the model in the reservoir zones, and fmore specifically in the top of conservation zone (Figure 11). We decided to manage this zone in order to allow more space for the incoming rain water. The releases polices that we thought out were reduce top of conservation in scenario 1 and 2 (see Scenario 1 and Scenario 2.
Figure 10.Average daily inflow of Verbano Reservoir
Figure 11. Top of Conservation and Average Daily Storage Volume
An alternative scenario study the possibility of release more water by changing the minimum flow requirement for Ticino River downstream (See Scenario 3).
The base scenario shows the calculations of the performance indicator in the current conditions of the Grandi Laghi Lombardi Model. For the Verbania Reservoir the managements decided to operate the reservoir zones as follow in the Figure 12:
Figure 12. description Verbano Reservoir zones for the base case
Scenario one: Reduce Top of conservation (Average)
The aim of the first scenario is allowing storing more water in the reservoir by reducing the value of top of conservation zone with the average of maximum exceeded water in October and November minus the threshold, according with the exceeded water of previous flood events presented in this months. The equations used to achieve this analysis are described below (Equation 1 and Equation 2) and the values are shown in Table 1.
TC1 october= TC - (Vea October -Th)
Equation 1. Reduction in the Top of Conservation zone in October
TC1 November= TC - (Vea November -Th)
Equation 2. Reduction in the Top of Conservation zone in November
- TC : Top of Conservation (m3) of Verbano Reservoir at case base
- Vea : Average available water (m3) when flood occurred in October and November
- Th : Threshold volume (m3) of Verbania
- TC1: Calculated Top of Conservation (m3) of Verbano Reservoir for Scenario 1 in October and November.
Table 1. Values of the Top of conservation in the Scenario 1
Scenario two: Reduce Top of conservation (maximum)
The aim of this scenario is the same of the scenario one, but the differences lies in the value used to reduce the top of conservation zone, instead of using the average, we used the maximum exceeded water in October and November minus the threshold. The equations used to achieve this analysis are described below (Equation 3 and Equation 4) and the values are shown in Table 2.
TC2 october= TC - (Vem October -Th)
Equation 3. Reduction in the Top of Conservation zone in October
TC2 November= TC - (Vem November -Th)
Equation 4. Reduction in the Top of Conservation zone in November
- TC : Top of Conservation (m3) of Verbano Reservoir at case base
- Vem : Maximum available water (m3) in October and November.
- Th : Threshold volume (m3) of Verbania
- TC2 : Calculated Top of Conservation (m3) of Verbano Reservoir for Scenario 2 in October and November.
Table 2. Values of the Top of conservation in the Scenario 2
Scenario three: Increase the minimum environmental flow
In this scenario we increased the minimum environmental flow downstream of October and November in order to release more water from the reservoir by increasing the downstream demand for Verbano Reservoir, to prevent upstream flooding as a result of the rains that are generated in the late summer and early fall.
The increase was from 13 m3/s to 800 m3/s on October and November. This increased minimum flow requirement value is the sum of minimum flow requirement (m3/s) in base case and average release water (m3/s) when flood occurred on October and November.
MFRsc3 = MFR + Average (Rf)
Equation 5.İncrease of minimum environmental flow downstream
- MFR : minimum flow requirement (m3/s)
- Rf : Average release water (m3/s) when flood occurred
- MFRsc3 : Calculated minimum flow requirement (m3/s) for Ticino River for Scenario 3.
Results, discussion and recommendation
After having run the three scenarios in WEAP we obtained the data necessary to evaluate our performance indicators. As we can see in the Table 3, the indicators achieve improvement with respect base case. The maximum flooded area decreased considerably with the management of the Top of Conservation zone as well as with the increasing of the minimum environmental flow in Ticino River. The maximum flood duration changed slightly because only decreased one day, although this indicator cannot be improved more since floods occurs within hours. The vulnerability of having flood events decreased 0.11 km2 in Scenario 2, whereas in the other alternatives was reduced about 0.065 km2.
Table 3.Performance indicators in base case, fist, second and third scenario
On the other hand, the improvements also are being reflected in the storage capacity of Verbano Reservoir (Available Water). In the first scenario the stored water exceeded the threshold about 9 days (Figure 13), meanwhile in the scenario 2 and 3 the topped was 5 (Figure 14) and 6 (Figure 15) days respectively. This indicates the storage capacity can be further improved, but the flood events remaining could be due to heavy rains with short duration.
Figure 13. Daily Available Water Scenario 1
Figure 14. Daily Available Water Scenario 2
Figure 15. Daily Available Water Scenario 3
Conflict among other stakeholder in GLL model
When designing planning and management polices of water many stakeholders are involved (Caminola, 2015). In the GLL system we have 4 different water users with their own goals, they are Farmers, Environmental Protectors, Hydropower Producers and Navigation. The stakeholders are located downstream and have diverse objectives to achieve: maximize farmer production, optimize flow range for brown and marble trout, maximize hydropower production in order to meet the demand, and keep the lake level in optimal conditions to be able to navigate, respectively. As we already mention before, Floods issues’ attenuation aims not to exceed the lake level with regard to threshold and to release excess water if it is necessary to reduce flooded area in Verbania city. The combination of all stakeholders’ objectives may cause conflict each other, so we decide to visualize theirs performance indicators in our alternatives scenarios to decide the best option for all users, taking into account to allocate water resources effectively.
The Performance indicators chosen of the others stakeholders involved were:
- Farmer: Vulnerability of unmet demand in Regina Elena Irrigation site.
- Environmental protectors: Reliability of the flow of Ticino River for the brown and marble trout`s preservation.
- Hydropower Producers: Vulnerability of unmet demand to generate enough energy.
- Navigation: Vulnerability of reduce the lake level for proper navigation conditions.
The objectives were visualized in DiscoveryDV (Kollat, et al., 2013) and are shown in the Figure 16. The desirable scenario for us is the second one, because have the highest improvements for our purpose (reliability 99.95%, maximum flooded area 0.81 km2, maximum flood duration 2 days and vulnerability 0.13 km2), nevertheless the result indicated that flood protectors have conflict with navigation user and farmers, for them, the indicator chosen by us do not have the optimal operation with our second scenario. The preferable scenario for them would be the third one in which they have less vulnerability of unmet demand in Regina Elena irrigation site and less vulnerability of reduce the lake level for proper navigation conditions, but since we are caring about upstream flood protection, we decide on the second scenario.
Figure 16. Visualization of Stakeholders’ objectives in the GLL model
The results of the other stakaholders' indicators are shown in the Appendix A
As the inundation events on Lake Verbano shores, especially Verbania that is our area of interest within the scope of the project, are significant issues that should be reduced for public weal. To do this, in GLL basin that has 5 different stakeholders including flood protectors, three scenarios were constructed and simulated to see which actions achieve our goal that is to reduce maximum flooded area as comparing with base case which represent the state of the GLL basin system without taking any action using WEAP software. Calculations of all stakeholders’ performance metrics were made in Matlab and visualizations of trade-offs among the all stakeholders in order to decide one alternative in DiscoveryDV (Kollat, et al., 2013) were evaluated. The second scenario that reduces the top of conservation with the maximum value that can be possible is the preferred and recommended for management release policies. This scenario improved the values of performance indicators of the base case scenario from 99.81%, 1.38 km2, 3 days, 0.24 km2, to 99.95%, 0.81 km2, 2 days, 0.13 km2 of reliability, maximum, flooded area, maximum flood duration, vulnerability, respectively. As a result of this, our preferred alternative was not proper for farmers and navigation because they do not meet their objectives.
- Anghileri, D., Mocotti, M., & Weber, E. (2013). //Rapporto D3 - Il Modello complessivo del sistema.// Italy: Dipartamento di Electrom¡nica e informazione of Politecnico di Milano. Retrieved from http://www.progettostrada.net/media/report_conclusivi/Report_Tecnici_Azione_2/d3.pdf
- Betti, E., Cellina, F., Passoni, A., Soncini-Sessa, R., & Weber, E. (2013). //The Verbano Project: sustainable water management in a dynamic and transboundary context.//
- Caminola, A. (2015). //Presentation slides of Interactive multi-objective visualizations class.// Como, Italy.
- Cannata, M., Antonovic, M., Molinari, M., & Pozzoni , M. (27-28 de February de 2013). called istsos, sensor observation management system: a real case application of hydro-meteorological data for flood protection. Padua, Italy. Obtenido de http://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XL-5-W3/111/2013/isprsarchives-XL-5-W3-111-2013.pdf
- Coluccino, M., & Strigaro, D. (2012). INTERREG-Provincia del Verbano-Cusio-Ossola PROGETTO STRADA AZIONE 6.3 SCENARI DI EVENTO E DI VULNERABILITA(per fenomeni di esondazione del Lago Maggiore a Verbania). Lugano.
- Engelhardt, A. (n.d.). Retrieved 2015, from http://baobab.elet.polimi.it/TwoLeWeb/verbano/html/cap01/184.108.40.206/cornice.htm
- FOEN. (2015). Retrieved 2015, from Federal Office for the Environment Topic Hydrological foundations and data: http://www.hydrodaten.admin.ch/warnungen-vorhersagen/en/index.html?lang=en
- Guariso, G., Laniado, E., & Rinaldi, S. (1985). //The management of Lake Maggiore Conflict Analysis and the Price of the Noncooperation.//
- Jasper, K., Gurtz, J., & Lang, H. (2002). //Advanced flood forecasting in Alpine watersheds by coupling meteorological observations and forecasts with a distributed hydrological model.// Switzerland: Journal of Hydrology.
- Kampf, L., Brauer, A., Dulski, P., Lami, A., Marchetto, A., Gerli, S., . . . Guillizzoni, P. (2012). //Detrital layers marking flood events in recent sediments of Lago Maggiore and their comparison with instrumental data.// Italy: Freshwater Biology. Retrieved from http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2427.2012.02796.x/abstract
- Kollat, J. B., Woodruff, M. J., & Lewis, R. w. (2013). Discovery Decide Deliver. DiscoveryDV software Version 0.27.
- Poretti, L. (2010). Flood hazard analysis for river systems. Comunità Montana Valtellina di Tirano, Italy. Recuperado el 28 de May de 2015, de https://boa.unimib.it/.../Phd_unimib_033261.pdf
- Premazzi, G., Dalmiglio, A., Cardoso, A. C., & Chiaudani, G. (2003). //Lake management in Italy: the implications of the Water Framework Directive.// Lakes & Reservoirs: Research and Management.
- MathWorks, Inc. © 2015 MATLAB and Simulink are registered trademarks of The MathWorks, Inc. Natick, 2009
- Stockholm Environmental Institute SEI. (2007). //WEAP Water Evaluation and Planning System User Guide for version 3.2.// Retrieved from http://www.weap21.org
Can Delice, M.S Student, Politecnico di Milano, 2015 Como, Italy, firstname.lastname@example.org
Gao He, M.S Student, Politecnico di Milano, 2015 Como, Italy, email@example.com
Gina Alexandra Cabas Rosado, M.S Student, Politecnico di Milano, 2015 Como, Italy, firstname.lastname@example.org
Seda Karatas, M.S Student, Politecnico di Milano, 2015 Como, Italy, email@example.com
Appendix A: Other Stakeholders’ performance indicators evaluated in our scenarios
Table 4. Performance indicators of agriculture stakeholder
Table 5. Performance indicators of Hydropower stakeholder
Table 6. Performance indicators of Navigation stakeholder
Table 7. Performance indicators of Environmental stakeholder