Grandi Laghi Lombardi - AGRICULTURE
GRANDI LAGHI LOMBARDI: MAIZE CROP IRRIGATION STUDY
Authors: Johanna Garcia Jerez (firstname.lastname@example.org), Milica Predragovic (email@example.com), Carlo Luini (firstname.lastname@example.org)
This work contains a group project developed for Spring 2015 in Natural Resources Management and Integrated Water Resources Management (NRM+IWRM) at Politecnico di Milano (Polo Territoriale di Como) regarding the water system of Grandi Laghi Lombardi. This system consists of lakes, rivers and channels mostly located in the Italian region of Lombardy and where different stakeholders operate: namely the agricultural, hydropower, flood protection, tourism and environmental ones. Among these, the present wiki page focuses on the agriculture and its main aim, which is to analyse how efficiently the actual water system meets the maize crop irrigation demand in the region of Lombardy, examine which actions can be undertaken to improve the system without generating conflict with other stakeholders.
The model used to describe the Grandi Laghi Lombardi system and maize irrigation process is implemented in Water Evaluation and Planning (WEAP) software (Stockholm Environment Institute, 2011), suitably combined with MABIA software (Jabloun & Sahli, 2012). This combination is mainly based on adding four catchments (Maize Regina Elena, Maize Industriale, Maize Villoresi, Maize Adda) to existing Agricultural Demand Sites (Regina Elena Irrigation, Industriale Irrigation,Villoresi Irrigation, Adda Irrigation District) within original model, and subtracting maize crop water demand from original agriculture water demand. Specific data were assigned to each catchment (Land area, Type of crop, Irrigation schedule, Type of soil) and the distribution rule associated to the three channels (Regina Elena, Industriale and Villoresi) in the model scheme were modified to take in account water demands calculated for catchments in previous time step. Three scenarios were simulated by changing the fraction of maize crop area that it will be irrigate in each catchment, evidently keeping constant the total maize crop area. Performance indicators considered to pursue the objective and analyse the results are reliability of water system to meet maize crops water demand, vulnerability (average non irrigated area), and other stakeholder’s reliability.
Results reveal that the best scenario is when catchments irrigate maize crops in the following area distribution: Maize Villoresi 9652 [ha], Maize Adda 149606 [ha], Maize Industriale 149606 [ha] and Maize Regina Elena 173736 [ha]. In this case the overall reliability of the system is improved and average non-irrigated area is minimized. Besides, this alternative did not show conflict with other stakeholders since their performance metrics varied ±0.23 [%].
The Grandi Laghi Lombardy water system represents an interactive system of lakes, rivers, reservoirs, stakeholders and demand sites. The model of this system with all its contains was provided to students as a starting point for project development using software Water Evaluation and Planning (WEAP), (Stockholm Environment Institute, 2011). The project is based on dividing class into five groups each representing different stakeholders with their own interest and objective: hydropower, flood protection, tourism, environmental and irrigation stakeholder. Each group defined their goals as well as way how to evaluate their development by specifying certain performance indicators. This process includes making infrastructure changes of the original model (according to stakeholder needs). After all individual preparations were finished, modifications were merged into one unique model (BEEP/Natural Resources Management/DOCUMENTI E MEDIA/Group Project/WEAP model/GrandLaghiLombardi_MonthlyIntegrated.rar). By running this model, each group was able to gain insight into their performance indicators and interaction with other stakeholders (by computing their performance indicators for each change done in model).
This paper is based on presenting Agricultural stakeholder, and more specifically maize crops irrigation in above mentioned Grandi Laghi Lombardi system. First, some main background about Lombardy region and its water system is given. Then the stakeholder and its objective: Maximize the ability of Grandi Laghi Lombardi water system to fulfill the requirements of maize crop irrigation, without causing a conflict with other stakeholders, are introduced followed by performance metrics indicators: (i) reliability [%] (ability of system to meet maize crops water demand), (ii) vulnerability [ha] (average non irrigated area) and (iii) reliability of other stakeholders [%].
Hereafter, main characteristics of Agricultural sector of Lombardy region and more closely specific details about maize crops cultivation and water needs are presented. Further on it is stated which modifications to initial WEAP model are performed and tools used (software MABIA). These modifications are divided into two groups (i) Schematic modification (building catchments and accompanying elements e.g. transmission links, groundwaters) and (ii) Data modification (adjusting initial data to changes made). All of this lead to final model set and scenarios simulation. Three scenarios are built, varying the irrigation area four implemented catchments (//Maize Regina Elena, Maize Industiale, Maize Villoresi, Maize Adda//) are supposed to provide water for. Finally, the explanation how performance indicators are computed (combining WEAP and Matlab code) is given, obtained results are presented and discussed and according to them recommendations and final conclusions are derived.
Grandi Laghi Lombardi water system
"Lombardy region is one of the 20 regions of Italy located in the north of this country. It is the most populated Italian region, with nearly 10 million inhabitants, mostly concentrated in lowlands and foothills (Figure 1). In these areas, the population density exceeds 600 inhab/km. Lombardy has a total surface of 23,862 m2, 7.9% of national area, and it is characterized by lowlands (47%), hills (12.4%) and mountains (40.5%). Land cover shows a prevalence of agricultural areas (43.7%), followed by woodland (24.5%) and populated areas (14.1%)", (INEA, 2012).
Figure 1. Location of Lombardy region in Italy (INEA, 2012).
The model representing the Grandi Laghi Lombardi water system implemented in software WEAP (Stockholm Environment Institute, 2011) ,provided by teacher in the beginning of the project, is evident on Figure 2. Two major basins compose the given Grandi Laghi Lombardi system: the Lake Maggiore and the Adda basin, each one governed by a regulation authority. Two reservoirs are present in Lake Maggiore basin: Lake Ceresio and Lake Verbano, whereas one reservoir in Adda basin: Lake Como. This hydrometric network contains three rivers: Ticino River, Tresa River (Lake Maggiore basin) and Adda River (Adda basin). Lake Maggiore basin contains three channels: Regina Elena, Industriale and Villoresi channel. Along first two channels, four demand sites are located: Regina Elena hydropower, Regina Elena irrigation, Industriale hydropower, Industriale irrigation, one demand site along Villoresi channel: Villoresi irrigation, and one demand site on Adda River: Adda irrigation district. Other elements of system can be visible in Figure 2 legend.
Figure 2. Original WEAP model scheme of Grandi Laghi Lombardy water system.
The agriculture stakeholder, and more specifically maize growing stakeholder represent only one part of immense, complex and interactive water system of lakes, rivers, reservoirs, and other stakeholders of Lombardy (hydropower, tourism, environmental and flood protection) govern by the authorities and water managers. All of these elements are sharing the same water resources of Lombardy region, and therefore are in close relationship and interaction with one another.
Objective and performance metrics
The aim of this project is to analyze how efficiently the current water system meets the needs of the maize crop fields irrigation in the region of Lombardy, examine which changes can be made to improve the productivity of the system and present the results and constructive recommendations driven from the analysis. In order to conduct this, certain modifications of model infrastructure are required and discussed below. In this sense an objective of representative stakeholder can be defined: Maximize the ability of Grandi Laghi Lombardi water system to fulfill the requirements of maize crop irrigation, without causing a conflict with other stakeholders. To describe and quantify the objective, three performance indicators are used:
- Reliability [%] - a fraction of time when a system is in satisfactory state (Stockholm Environment Institute, 2011). In this case a satisfactory state means: water demand is met and no shortages occurred (section Reliability).
- Vulnerability [ha] - for each demand site is the average non irrigated area (when shortage occurs) with respect to the total area dedicated to the maize crop (section Vulnerability):
Equation 1. Performance Indicator: Reliability
- Reliability of other stakeholders [%] (section Reliability of other stakeholders).
In Mediterranean countries, agriculture sector is one of the main elements in water management. In Italy, irrigation accounts for about 50% of total water use and around 60% of Italian agricultural exports are produced by irrigated methods (Bartolini, Bazzani, Gallerani, Raggi, & Viaggi, 2005). The average yearly precipitations are 300 billion m^3 corresponding to about 1000 mm/year. About 18% of the 300-billion m^3/year of precipitations are used for civil, agricultural or industry needs. The water availability limitation is mainly due to the irregular distribution, in space and time, of the rainfall (Orlandini, Marta, & Natali, 2009). In Lombardy region most of the agricultural land is arable, drained by artificial ditches, and irrigated during the summer. According to Eurostat (2012) major relevant crops are cereals (especially maize) with an average production of 8.83 ton.ha-1 during 1999 – 2007 (Bocchiola, Nana, & Soncini, 2013). The irrigated area is more present in the North of Italy (70%) and it is spread in plain territory. The water is distributed using sprinkler irrigation (37, 5% of irrigated surface), followed to surface irrigation (30,2%), drip irrigation (20,6%), flooding (8,8%) and other methods (3,8%). The 45.4% of farms (corresponding to 733.775 ha) takes water from wells; the 40.4% (corresponding to 1.452.335 ha) is irrigated from land-reclamation authority (Orlandini, Marta, & Natali, 2009).
Maize crop water demands
In the region maize is the major crop cultivated as a function of Utilized Agricultural Area (UAA) where grain maize covers 34% of the area total area (ERSAF, 2014). Moreover, it is important to take in account the amount of maize cultivated in crop rotation (maize-wheat, maize-alfalfa, etc.) as is presented in Figure 3. Total area used in this work is 482600 [ha] taking in account Meadows-Grain Maize, Grain Maize, Silage Maize, Maize-Wheat, Alfalfa-Maize-Wheat (Zopounidis, Kalogeras, Mattas, Dijk, & Baourakis, 2014).
Figure 3. Major crops cultivated as a function of the total Utilized Agricultural Area (UAA) in the Lombardy region (ERSAF, 2014)
Maize crop requires significant amount of water for production, i.e. rainfall and irrigation. With available water in the soil, plants have enough water to meet their potential transpiration. On the contrary, if precipitation decreases and it is not compensated by irrigation there will be a permanent damage in the plants (Bocchiola, Nana, & Soncini, 2013). Maize crop is potentially the highest yielding grain crop due to efficient uses of water in terms of total dry matter and among cereals. A medium maturity grain crop requires between 500 and 800 mm of water for maximum production depending of the climate (FAO, 2013). Water requirements are supplied by rainfall and irrigation, and to obtain a good stand and rapid root development, the root zone should, where feasible, be wetted at or soon after sowing. Where rainfall is low and irrigation water supply is restricted, irrigation scheduling should be based on avoiding water deficits during the flowering period (2), followed by yield formation period (3), (Figure 4). Under conditions of marginal rainfall and limited irrigation water supply, the number of possible irrigation applications may vary between 2 and 5 (FAO, 2013).
Irrigation water requirement is the amount of water that has to be applied in addition to rainfall to serve crop water requirements (Wriedt, Velde, Aloe, & Bouraoui, 2008). According to Agricultural Census in Italy (2012) Lombardy region is the most intensively irrigated, as on average 8.182 cubic meters of water were used for each hectare of its Utilized Agricultural Area (UAA), (Eurostat, 2012). In the case of maize crop, water demand is in general 500 – 600 mm during the cultivation period. About half of this amount is applied through rainfall, whereas the other half (250 – 300 mm) must be applied through irrigation (Union, 2012). This value should be provided by water basin during the cultivation period to ensure maize crops development.
Figure 4. Schematic graph of the growth periods of maiz (FAO, 2013)
Model Infrastructure Adaption
To be able to examine the interaction between Grandi Lagi Lombardi water system and maize crops irrigation and calculate listed performance metrics it is necessary to make several infrastructural changes to given WEAP model. These changes are based on introducing catchments into infrastructure of a model, which simulate maize crops irrigation needs, in combination with software package WEAP-MABIA which is used to determine monthly water demand for each catchment separately according to specific parameters that affect the dynamics of irrigation process (see section Catchment and WEAP-MABIA). Further on, more details about adaptations made in WEAP are given.
To extract maize crops irrigation from other crops irrigation, four catchments (Maize Regina Elena, Maize Industriale, Maize Villoresi, Maize Adda) are added next to existing Agricultural Demand Sites (Regina Elena Irrigation, Industriale Irrigation,Villoresi Irrigation, Adda Irrigation District). Each Catchment is linked with corresponding river using transmission link located into the same withdrawal node as demand site. To be able to operate with catchments it is necessary to introduce a ground water next to each catchment and connect the catchment with Runoff/Infiltration link both with ground water and corresponded river due to software requirements (Figure 7). Four Groundwaters are added to initial model named GWi (i=1,2,3,4). Priority of catchments is set equal to priorities of all demand sites (20), meaning that available amount of water is split evenly between those users, and that in case of drought, maize crops and other crops will have the same restrictions of water delivery. This assumption is made due to a fact that available documentation does not specify if maize crops, or any other crops have higher priority of water delivery.
A catchment is a unit within the water basin scheme that can simulate specific processes, for example precipitation, runoff, evapotranspiration and irrigation. When the catchment is added in the schematic, in this case for irrigation, it is necessary to create transmission links from a supply to the catchment and to input some parameters related to the crop (Stockholm Environment Institute, 2011). It is possible to choose one of four method to simulate catchment processes: (1) Rainfall Runoff; (2) Simplified Coefficient Approach, (3) Soil Moisture Method; and (4) MABIA Method. This selection depends of the available information and the results that are expected. During this project MABIA Method was used to simulate maize crops irrigation.
The software package WEAP-MABIA is used for modelling crop water requirements and the components in the water balance. It takes in account natural rainfall; irrigation scheduling and crop yield reduction. Also, it simulates runoff, infiltration and percolation due to agricultural processes (Jabloun & Sahli, 2012). Dr. Ali Sahli and Mohamed Jabloun developed MABIA software at the “Institut National Agronomique de Tunisie”. MABIA is based on dual Kc method like FAO Irrigation and Drainage Paper No. 56 indicated (Stockholm Environment Institute, 2011). Essentially, it is a daily simulation of irrigation requirements and scheduling that includes modules for estimating evapotranspiration and soil capacity. For each WEAP timestep (e.g., monthly), MABIA would run daily in that timestep and combined its results (evaporation, irrigation requirements, runoff, and infiltration) to that timestep. Furthermore, WEAP-MABIA package contains a “Crop Library” with the crop information of more than 100 crops from different regions and climates in the world. The crop information relates specific parameters like length of growth stage, total available water, rooting depth and evapotranspiration (see Figure 5).
Figure 5. Graph illustrating the different parameters considered in the crop module (Jabloun & Sahli, 2012)
Additionally, there is a “Soil Library” that allows the estimation of the average soil water capacity over several profiles and soil horizons. The parameters described in the tool for each type of soil are saturation, field capacity and wilt point. MABIA Method used two vertically stratified “buckets” (compartments) to compute the water balance (Stockholm Environment Institute, 2011). The bucket 1 is in the top and it represents the rooting zone where is placed the available water and infiltration process; else bucket 2 is in the bottom and the water goes to groundwater from it.
Using MABIA software, specific features (listed below) which simulate maize crops irrigation, are assigned to all catchments. This way, after running the results, it is possible to have an insight into maize irrigation water demands on a monthly base (Figure 6). Following the path /Data/Demand Sites and Catchments/Name of desired Catchment, these properties were introduced for all four catchments:
- Land area - presenting a specific portion (depending on Catchment) of total irrigable maize area 482600 [ha], (Zopounidis, Kalogeras, Mattas, Dijk, & Baourakis, 2014).
- Type of crops -(Crop Scheduling Wizard/ “Maize (sweet) Mediterranean”).
- Irrigation schedule: Irrigation/Irrigation Scheduling Wizard/Maize (sweet) Mediterranean, May 5 – Jul 23.
- Type of soil: Soil Water Capacity/Soil Library/Sandy clay loam, (Stockholm Environment Institute, 2011).
Water demand varies depending upon parameters listed above which are fix for each scenario. Acquired monthly demand and corresponding Bucket 1 water saturation profiles [%] (section WEAP-MABIA) obtained using software MABIA and WEAP are shown for the catchment “Maize Regina Elena”, (Figure 7). This example is sufficient to understand the dynamics of water demands variations for all catchments, since they follow the same shape of Water Demand an Bucket 1 Water Saturation profiles. During years when current Bucket 1 Water Saturation is varying between Field Capacity and Readily Available Water, water demand for all Catchments is equal to zero (Figure 6, 1976-1979). Observing the profile of monthly water demands, it is evident that the highest water demand occurs in time steps when the saturation of the soil drops very close to wilt point (absence of water). Presented profile of water demands ensures enough moisture of the field where maize crop is growing, and provides conditions necessary for stable and healthy crop growth (see section Maize crop water demands).
Figure 6. Maize Regina Elena monthly water demands [million cubic meters]
Figure 7. Bucket 1 water saturation [%]
Irrigation demand sites
When maize crops irrigation are extracted using catchments, it is necessary to adapt given monthly demand for Irrigation Demand Sites to changes made. In Data/Name of the Demand Site/Water Use/Monthly Demand/Expression Builder/ from current water demand, function PrevTSValue (returns results calculated in previous time steps) related to water demand of a relevant catchment is subtracted. This way the portion of water maize catchments are using every month is deducted from the rest of the water related to other crops irrigation, and can be examined separately. Whereas water demand for maize catchments is computed using MABIA software, water demands for rest of the crops is changing according to catchments requirements, which means that it is based on difference between fixed values of monthly demands (provided in original WEAP model) and maize crops monthly demands for previous year.
Provided WEAP model of Grandi Lagi Lombardi water system has its water management policy determined by water managers, which is represented through distribution rule associated with three channels in model infrastructure: Regina Elena, Industriale and Villoresi channel. This distribution rule is described using formula of distribution, specified for each channel separately and integrated in original model provided for project development. These distribution rule formula for each channel is modified in such a way that it takes into account water demands calculated for catchments in previous time step and modified to be expressed in suitable units [m^3/s]. To see the whole modified formula follow the path: GrandLaghiLombardi_MonthlyIntegrated/Schematic view/Right click o “Channel name”/ Edit data/Maximum Diversion/Expression Builder.
Final model set
After implementing all infrastructure changes and adaptations discussed above, it is necessary to integrate and combine the changes other stakeholders have made as pre requirements to their project development into a single model. This involves four stakeholders (besides agriculture) who represent: hydropower, tourism, flood protection and environmental. Final model set is obtained (Figure 8) and capable of simulating water system of Grandi Laghi Lombardi, in terms of providing data needed to compute performance metrics for all stakeholders and their interest in specific outcomes of system simulation.
Figure 8. Modified WEAP model of Grandi Laghi Lombardi water system
Once finishing all pre requirements described above, it is possible to create different case scenarios depending on amount of area catchments should irrigate. According to Figure 9 different types of maize crops are spread all over the Grandi Laghi Lombardi area. Whereas no details about area division among irrigation sites were found during background investigation, a specific portion of land for irrigation is assigned to each catchment. This is done by defining Land area (see section Catchments data) in such a way that it represents a portion of whole irrigable land related to maize crops.
Figure 9. Representative cropping system of the Utilized Agricultural Area (UAA) of the agrarian regions in Lombardy plain, (Zopounidis, Kalogeras, Mattas, Dijk, & Baourakis, 2014). (PP: Permanent meadows, PVM: Meadows-Grain Maize, MG: Grain Maize, MF: Silage Maize, F: Wheat, M_F: Maize- Wheat, MG_L: Grain maize + cover, R: Rice, A: Crops trees, O: Open field vegetables, Alfa: Alfalfa – Grain Maize)
To obtain a certain portion of land it is enough to implement the total amount of area to irrigate into Land area (482600 [ha]) and multiply it with a number between 0 and 1, to obtain a fraction of total area. In this sense, mentioned number will be called area distribution coefficient. Area distribution coefficient can be manually implemented and changed to simulate irrigation of different portions of total area. The sum of four coefficients introduced to a model has to be equal to 1, to maintain the fixed amount of area catchments are supposed to irrigate (482600 [ha]). According to this, three scenarios are simulated:
- Base case scenario is based on assumption that all four catchments provide water for the same amount of area [ha]. Since the base case assumes identical portions of land to irrigate, area distribution coefficient is equal for all catchments, and amounts 0.25.
- Second scenario refers to irrigation demand sites location. Three irrigation demand sites (Regina Elena, Industriale and Villoresi irrigation) are located in the West of Grandi Laghi Lombardi area (Figure 10), whereas Adda irrigation is the only irrigation site for middle and East of the region. This is why in second scenario area distribution coefficient assigned to Maize Adda catchment amounts 0.4, and for the rest of the catchments is equal to 0.2.
- Third scenario After running the results of base and second scenario, and investigating reliability for these two cases, third scenario is created by assigning area distribution coefficient in such a way that a portion of land is lowered in catchment which has lower reliability (Maize Villoresi) and increased for the rest of the catchments (see section 6.4. Results, base case and second case reliability). Area distribution coefficient in this scenario is:
- Maize Adda 0.31
- Maize Industriale 0.31
- Maize Regina Elena 0.36
- Maize Villoresi 0.02
Figure 10. Lombardy region lakes (BlogSpot, 2012).
Running the Model to generate Results
Running results after finishing all steps described above allows us to have the insight into data necessary to compute our performance metrics.
It can be directly read from WEAP by choosing the path: Results/(Drop down menu)Demand/Reliability and by selecting desirable Demand Sites, in this case Maize Regina Elena, Maize Industriale, Maize Villoresi, Maize Adda. Same results can be obtained by running MATLAB (MathWorks, 2014) code which uses as input data the monthly time series (covering the period 1974-2010) for unmet demands of each demand: Results/(Drop down menu)Unmet Demand/(desirable Demand site). The time series of unmet demands has initially to be exported into an Electronic sheet (e.g. Microsoft Excel) then transposed, and finally copied into a .txt file. The .txt file can therefore be loaded into Matlab and put into a vector. The Matlab code used, basically analyses the time series of unmet demands within the vector and, through a Boolean function, assigns the value of 1 every time the unmet demand is equal to 0, 0 in the other case; this series of 0 and 1 is put into a new vector. The reliability is then computed by making the sum of the elements of the Boolean vector and dividing this sum by the length of the time series.
Through WEAP, as seen for computing the reliability, the monthly time series of unmet demands is provided. Moreover, also the data related to the monthly time series of water demands is available (Results/Drop down menu) Unmet Demand/(desirable Demand site)); water demand in WEAP represent the need of a demand site before considering losses and reuse (Stockholm Environment Institute, 2011); the monthly time series taken into account is always the one going from year 1974 to 2010. The vulnerability indicator is computed only by using Matlab, which means that vectors containing monthly time series of unmet demands for each demand site and, in this case, also the ones containing the series of water demands, have to be loaded into the software. The Matlab code generated to compute vulnerability makes a proportion between average unmet demand with respect to average water demand and average non irrigated area with respect to total area. Both Matlab codes that compute reliability and vulnerability are uploaded to site BEEP under the names ReliabilityCode and VulnerabilityCode and available for further usage.
Reliability of other stakeholders
To study how creating new case scenarios and changing specific parameters (i.e. area distribution coefficient, see section Scenarios Simulation) affect other stakeholders (environmental, flood protection, tourism and hydropower), reliability for each stakeholder was computed and examined. To do this, Matlab codes, provided by stakeholders, available on BEEP are downloaded, input data from WEAP model results (specified by stakeholder) is extracted and modified (if necessary) to appropriate form.
Using the described methods reliability and vulnerability are computed for four maize irrigation catchments in case of all three scenarios simulated (Table 1).
Table 1. Reliability [%] and Vulnerability [ha] for four catchments in case of three scenarios
Note that the value 0 related to vulnerability of Maize Regina Elena catchment amounts less than 0.003 [ha] which can be neglected.
Affect on other stakeholders performance indicators (Reliability)
Results obtained are indicating that modifications made to simulate second and third scenario cause reliability to differ from base scenario in tolerance ±0.23[%] in case of all stakeholders (Table 2). This means that maize crops irrigation stakeholder does not face conflicts with other stakeholders when adapting one of new alternatives (scenarios created).
Table 2. Reliability of other stakeholders for all scenarios
Note that tourism stakeholder is not included in analyzes since input data for their performance metrics is "Workspace_COMOandMAGGIORE" – file extracted from the Excel files that comes along WEAP Model, which indicates the Historical Levels observed for two lakes, and therefore fixed and cannot be affected by changes made.
Results discussion and Recommendations
After computing performance indicators reliability [%] and vulnerability [ha] for maize irrigation stakeholder in case of three scenarios created, as well as analyzing the effect on other stakeholders it is evident that Third scenario gives the best results. Reliability of Maize Adda (96.85 [%]) and Maize Industriale (97.97 [%]) are same in all scenarios, but also really close to the 100 [%] reliability. Reliability of Maize Regina Elena decreased just 0.67 [%], whereas vulnerability stays 0 [ha]. On the other hand, Maize Villoresi reliability is 6.31 [%] improved and average non-irrigated area (vulnerability) dropped 42 [ha] comparing with base case scenario, (Table 1). Another important characteristic of Third scenario is that the total amount of average non irrigated area when taking into account all four catchments is significantly lower in comparison with base scenario (36 [ha] less) and second scenario (15 [ha] less). As mentioned before, none of the scenarios affect other stakeholders performances and do not cause conflict. This is why third scenario is the best way to increase the overall efficiency of maize crops irrigation, and minimize average non irrigated area of maize crops in Grandi Laghi Lombardi. Following distribution of area among four irrigation sites is recommended:
Table 3. Area distribution recommendation among maize irrigation sites
Since maize crop cultivation represents 75 [%] (including all types of maize) of total crops grown in Lombardy region agriculture sector, and it is potentially the highest yielding grain, ensuring a reliable irrigation policy is of great importance. Managing the amount of area assigned to each maize irrigation site can help improving systems reliability to provide desirable amount of water to this consumer and minimize the average non irrigated area in cases shortages occur. If Adda and Industriale irrigation districts dedicate 149606 [ha] each, of maize irrigation area, Regina Elena irrigation 173736 [ha] and Villoresi irrigation 9652 [ha] of area, the overall reliability of the system is improved and average non irrigated area is minimized comparing to other area division approaches. Moreover, recommended alternative has another important property (along with other two alternatives) – the conflict with other stakeholders is avoided since proposed alternatives do not change their performance matrices more than ±0.23 [%].
Bartolini, F., Bazzani, G., Gallerani, V., Raggi, M., & Viaggi, D. (2005). Water Policy and Sustainability of Irrigated Farming Systems in Italy. European Association of Agricultural Economists, 23-27.
BlogSpot. (2012). Retrieved May 2015, from Lombardia mapa de la ciudad: http://4.bp.blogspot.com/-cwgHJZZwrik/TydAe-DoTvI/AAAAAAAABA0/2IeG7Pt2AT0/s1600/lombardia-mapa.png
Bocchiola, D., Nana, E., & Soncini, A. (2013). Impact of climate change scenarios on crop yield and water footprint of maize in the Po valley of Italy. Agricultural Water Management, 116, 50-61.
ERSAF. (2014, 09 08). Regione Lombardia. Retrieved 04 20, 2015, from ERSAF - Ente Regionale per i Servizi all’Agricoltura e alle Foreste: http://www.ersaf.lombardia.it/upload/ersaf/gestionedocumentale/Visita_catalunya_ERSAF_784_20154.pdf
Eurostat. (2012, 11). Agricultural census in Italy. Retrieved 03 2015, from Statistic Explained: http://ec.europa.eu/eurostat/statistics-explained/index.php/Agricultural_census_in_Italy
FAO, F. A. (2013). FAO Land and Water Division. Retrieved 04 2015, from Crop Water Information: http://www.fao.org/nr/water/cropinfo_maize.html
INEA. (2012). Lombardy agriculture in figures. Retrieved April 2015, from Regione Lombardia: http://www.agricoltura.regione.lombardia.it/shared/ccurl/900/873/Lombardy_2012.pdf
Jabloun, M., & Sahli, A. (2012, January). Groundwater. Retrieved March 2015, from Bundesanstalt für Geowissenschaften und Rohstoffe (BGR) in Hannover: http://www.bgr.bund.de/EN/Themen/Wasser/Projekte/abgeschlossen/TZ/Acsad_dss/tutorial_weap-mabia.pdf?__blob=publicationFile&v=2
Mathworks. (2014). MATLAB R2013b. Retrieved from www.mathworks.com
Orlandini, S., Marta, A. D., & Natali, F. (2009). Water use in italian agriculture: analysis of rainfall patterns, irrigation systems and water storage capacity of farm ponds. Italian Journal of Agrometeorology, 9, 55-59.
Stockholm Environment Institute. (2011, May). User Guide. Retrieved March 2015, from WEAP - Water Evaluation And Planning system: http://www.weap21.org/downloads/WEAP_User_Guide.pdf
Union, E. -E. (2012, 04). TRANSNATIONAL INTEGRATED MANAGEMENT OF WATER RESOURCES IN AGRICULTURE FOR EUROPEAN WATER EMERGENCY CONTROL (EU-WATER). Retrieved 03 2015, from Southeast-Europe: http://www.southeast-europe.net/document.cmt?id=254
Wriedt, G., Velde, M. V., Aloe, A., & Bouraoui, F. (2008). European Comission. Retrieved 04 2015, from Water Requirements for Irrigation in the European Union: http://agrienv.jrc.ec.europa.eu/publications/pdfs/JRC46748_Report_Irrigation_EUR_23453_EN.pdf
Zopounidis, C., Kalogeras, N., Mattas, K., Dijk, G., & Baourakis, G. (2014). Agricultural Cooperative Management and Policy : New Robust, Reliable and Coherent Modelling Tools. Springer - Editor.
Milica Predragovic. M.S. Student Politecnico di Milano, 2015. Como, Italy.
Johanna Garcia. M.S. Student Politecnico di Milano, 2015. Como, Italy.
Carlo Luini. M.S. Student Politecnico di Milano, 2015. Como, Italy.