Water–Energy–Food Security Nexus—Estimating Future Water Demand Scenarios Based on Nexus Thinking: The Watershed as a Territory

Abstract

Water, energy, and food security are essential for maintenance and human survival. This research applied an approach for the water–energy–food system to a watershed in the Atlantic Rainforest in Southern Brazil. It is based on a WEAP system that was developed and evaluated future water demand scenarios for the 2020–2050 timespan. The Sustainability Index (SI) was used to assess the scenarios to propose an alternative to compare the current development trend. The results indicated that by 2050, the necessary flow for different water uses would be 78.75% greater than in 2020 while maintaining the present scenario (C_REF). Substituting part of the hydroelectric energy by photovoltaic solar energy (C_EAL), implementing watershed action masterplans as a water resource instrument (C_GES), and integrating these scenarios (C_INT) will reduce the current demand ranging from 8.96% to 24.59%. The average flow would decrease by 28.44% and 29.07%, respectively, by evaluating water availability in climatic change scenarios (C_MCL) for the RCP 4.5 and 8.5 scenarios. Compared with the reference scenario C_REF, C_INT presented an improved SI (0.1231), while C_MCL 8.5 presented the worse (−0.0047). Therefore, resources must be generated holistically so that one segment does not negatively impact the others. The findings highlight the pressing need for Santa Catarina State to implement effective management strategies targeting specific sectors, particularly the industrial and human consumption supply sectors. It is imperative to propose adaptation plans and implement actions that foster the reduction in water demands while also providing subsidies and guidance to the industrial sector on responsible water usage to prevent any negative impact on their productivity in the future. Comprehensive plans and policies should be formulated to ensure integration and coherence among various sectors involved in water resource management.

1. Introduction

The population growth associated with the great demand for natural resources has promoted dysfunctions in urban environmental dynamics [1]. It has been estimated that by 2050, around 70% of the world’s population will live in urban zones, increasing the demands for water, energy, and food [2]. The absence of these resources can trigger environmental, social, and political crises, causing the disestablishment of the political systems inside countries as well as beyond national borders [3,4,5]. In response, the water–energy–food nexus (WEF nexus) has arisen as an approach based on understanding the existing relationship related to water, energy, and foods, providing the consonance among these sectors to integrate them physically, socially, and politically [3,6,7].

The global forecast demand for water and food will increase by 55% and 60% by 2050, respectively. Furthermore, the energy demand will increase by 60% in the next ten years [8]. Factors such as the rapid urbanization process linked to economic growth and changes in consumption patterns have resulted in demands on the availability of water, energy, and food supply [9]. The high degree of modification and exploitation of the environment has contributed to degrading ecosystems, and the complex relationship among these resources has exerted increased stress on the importance of developing integrated management of them [3].

Classic management theories on linear causalities and suppositions on unlimited resources do not offer more solutions to the current problems; therefore, other alternatives are demanded that can address multiple causes, interdependence, and the interconnections of nexus sectors [10]. Water is present in most processes involved in managing energy and food, playing a core role in the stability of these sectors [3,11]. It is necessary to consider the existing connections among them so they can be understood to promote the security and availability of these resources [12]. Thus, quantifying the interconnections is a major step in modeling and managing large-scale solutions [6,13].

Moreover, inclusion and promotion of the involvement of stakeholders have been premises in developing intervention measures based on nexus [4,14]. Measures for environmental sustainability and resources were proposed only for technocratic approaches in the past. However, deficits have originated from deficient dialog among the supply chain’s stakeholders utilizing these resources [15]. Holistic approaches enable a broad view of the essential necessities of the nexus. Collaborative interventions provide decision-makers with subsidies for assessing the context and formulation of solutions for guaranteeing water, energy, and food security [6]. Governance is still being approached in an isolated fashion despite being intrinsically linked to the interests of the sectors contemplated by nexus, thereby compromising the existing connections among these [16].

Due to the current climate change scenario and high resource demand, the importance of developing research involving the three sectors has been further emphasized [17]. Brazil must invest in research studies considering the three nexus pillars, not only water and energy, as most studies have proven that necessity [7]. By the end of the 21st century, small watersheds (<500 km²) will face extreme events (droughts, floods, and temperature variations), demanding that authorities adopt sustainable resource management measures [18].

The Water Evaluation and Planning System (WEAP) is a practical computerized tool capable of simulating and planning integrated water systems, enabling the analysis of supply and resource allocation policies for correct and sustainable water management [19,20]. It makes it possible to analyze current and future scenarios (climatic changes, anthropogenic variations, changes in resource-use patterns, and others), proposing to aid, but not to substitute, the decision-maker [19,21]. Hydrological processes can be modeled using process-based hydrologic models or recent computational techniques. Both can present pros and cons, while hydrologic models, such as WEAP, need enormous data entry and computational techniques (CTs) as data-driven machine learning models. As promising tools, CTs are growing in their applications because of their flexibility and the possibility of identifying complex relationships, among other things. However, they still need further studies to validate the transferability and scalability [22] or the influence of local mechanisms [23].

Authors have utilized WEAP on regional scales, especially at the watershed scale, for diverse applications as an alternative instrument for managing and planning water resources [21,24,25,26,27]. Related to the water–energy–food nexus, as focused on by Momblanch et al. (2018), Guan et al. (2020), and Liu et al. (2021) [28,29,30], where future scenarios were proposed considering such premises as socioeconomic growth, the impact from climatic changes, utilizing measures to reduce agricultural water consumption in the studied regions, as well as a combination between proposed scenarios seeking to maximize water availability. Finally, indexes were used to measure the efficacy of the proposed measures in the respective scenarios.

Studies focused on developing sustainable practices in small watersheds are necessary due to the vulnerability of these regions [18] and the remaining challenges to addressing the WEF nexus in water resources management [31]. The Brazilian National Water Agency and Basic Sanitation in Brazil (ANA) have stated that water resource management is a set of structured and organized actions with the participation of society to control and regulate water use, seeking to ensure its supply for the next generations [32]. In Brazil, water is defined as a public asset. Law # 9433, dated 8 January 1997, the Brazilian water resource policy (PNRH), implemented guidelines connected to water resources [33]. Although it was implemented in 1997, all the instruments laid down in PNRH still need to be implemented. Furthermore, the domain of the Brazilian bodies of water does not make all watersheds subject to PNRH. State-owned watersheds are subject to the management policies of state water resources, and each state has its own policy, making evaluations and comparisons more complex among different administrative spheres.

This article seeks to identify the relations that permeate the water–energy–food nexus based on the context of the Cubatão River Watershed in Santa Catarina State, in Southern Brazil. It is a small watershed with industrial characteristics, a high GDP, and is located in the region of the Atlantic Rainforest. The WEAP planning tool was used for that reason and to prepare for future water demand scenarios for the 2020–2050 period. The sustainability of different proposed actions was analyzed using the Sustainability Index (SI). Finally, the impacts will be evaluated on the best representation of user sectors to allocate water and its implications for managing water resources.

2. Materials and Methods

2.1. The Study Area

The Cubatão River Watershed (BHRC) (Figure 1), 26°16′51.09″ S 48°45′54.80″ W, in Southern Brazil, covers an area of 492 km² [34], the altitude varies from 0 to 1500 m above the mean sea level. Regarding the climate, based on the Köppen classification, its climate is Cfa (mesothermic, humid, and no dry season) in the plainand Cfb (mesothermic, humid, and a mild summer) in the upland regions [35,36,37], and the average annual precipitation is 2182.6 mm. It is responsible for the public water supply of 75% of the population of the Joinville municipality, or approximately 450,000 inhabitants [38,39]. The watershed is in the Atlantic Rainforest domain; vegetation is present in the medium/advanced phase in most of the area (around 76% of the region is composed of a forest formation) [34,40]. Economically, the service-providing and industrial sectors are the most important to the region [41]. The industrial sector contributed 25.88% of the region’s GDP, in Brazilian Real, BRL 34,528,619,000.00 (about USD 10,824,018,500.00) in 2019 [39]. Therefore, the industrial sector contribution was about BRL 8,936,125,000.00 in 2019.

2.2. WEAP Software

The Water Evaluation and Planning System is a computerized water resource integrated management tool developed in 1988 by the Stockholm Environment Institute (SEI) [20,21]. It is a flexible, integrated, transparent tool for evaluating sustainability and water management. It enables the association among physical hydrological processes, demand management, and the installed infrastructure. Besides that, it is possible to add the forecast and analysis of multiple scenarios involving climatic changes, regulations, available infrastructure, ecosystems, and anthropogenic actions [20,21,24].

To reproduce a watershed, the WEAP employs methods for simulating such processes as evapotranspiration, surface water runoff, groundwater, and deep percolation. Five methods are available in the platform; the Soil Moisture Method is the most complete and broad in scope for representing the watershed with two soil layers. It makes it possible to characterize its use in the hydrologic process [21,42]. Diverse authors have used the WEAP as a management and planning tool for water resources in various applications [24,25,26,27,28,29,30]. Furthermore, it provides an alternative for economically managing water resources for developing countries since the permission to utilize them is cost-free for these [43].

2.3. Data Collection

WEAP requires the implementation of climatic and hydrological parameters. The simulation area is shown in (Figure 1); this study corresponds to the upstream region of the “Estação de Tratamento de Água” (Water Treatment Station) (ETA) of “Rio Cubatão” (Cubatão River), where the Pirabeiraba Fluviometric and Pluviometric Station is located (codes 2648033 and 82270050 for the fluviometric and pluviometric station data, respectively). Monthly data were collected regarding precipitation, flow, climate, usage, soil occupation, vegetal coverage, hydrogeological, energy production, and water demand from the study region, as stated in the sources listed in

Table 1. The collected data set for simulation in WEAP.

Data Set	Source
Rainfall	[44]
Flow	[44]
Temperature, Relative Humidity, and Wind Speed	[45,46]
Soil Use and Occupation	[40,47,48]
Vegetation Cover	[39,41,49,50,51]
Water Demands	[52,53]
Hydrogeological Data	[54]
Energy Production	[55,56]

.

2.4. Calibration and Validation

The calibration and validation process of the model seeks to obtain a set of data that resembles data series [21]. The BHRC was simulated using the Soil Humidity Method in WEAP, which obtained it as a modeled flow compared to the observed data in the utilized fluviometric station (Pirabeiraba Station—ANA code 82270050).

The parameters regarding the soil (Crop Coefficient, Retainment Capacity, Water Conductivity in the Radicular and Deep Zones, Resistance Runoff Factor, Preferential Flow Direction, and Storage Fraction at the Beginning of the Simulation of the Radicular and Deep Zones) were initially calibrated through an autocalibration tool integrated with WEAP, the Parameter Estimation Tool (PEST) [57], jointly with a manual adjustment approach [58,59,60]. Based on the estimates obtained from PEST, the values were adjusted to represent behavior similar to the observed data [61,62].

Seventy percent of the flow data were used from the Pirabeiraba station (1987–2006) for the calibration step and 30% (2007–2014) for validation. Finally, the Nash–Sutcliffe Efficiency (NSE), Percentage of Bias (P_bias), Determination Coefficient (R²), and the Willmott Concordance Index (d) were used for evaluating the model performance [63,64]. Model simulation can generally be judged when NSE > 0.50, Pbias < ±25%, R² > 0.50, and d-value results are near 1.0.

2.5. Developing Water Demand Scenarios

After the model was calibrated and validated, five scenarios were developed and applied to analyze the future water demand and availability for 2020–2050. Population growth rates were observed for the reference scenario (C_REF) according to Engecorps (2011) [65] at 1.57% per year, the increase in the Industrial Gross Domestic Product (GDP) was 4.43% per year according to Sidra/IBGE (2021) [39], and the growth dynamics of the agricultural areas in the past years according to Mapbiomas (2021) [40] was −0.50% per year. The forecasts for the next decades were obtained based on those data (

Table 2. The growth premise evolution for the study horizon (2020–2050).

Premises	2020	2030	2040	2050
Population Growth (inhab.)	448,244	523,804	612,101	715,282
Industrial GDP (Thousand R$)	6,665,888.22	10,117,658.81	15,356,846.14	23,309,021.18
Agricultural Areas (ha)	9547	9083	8641	8220

). Cubatão water supplies 71.55% of the industries in Joinville [66].

The C_EAL scenario proposes the substitution of a portion of the hydroelectric source production by photovoltaic energy. Currently, 63.86% of the energy produced in Santa Catarina State is from a hydroelectric source, and only 1.35% is from a photovoltaic source [56]. BHRC has no hydroelectric power plants installed in its region; however, a great amount of energy is consumed in performing its activities. A portion of the water retained from hydroelectric power generation is lost by net evaporation. Hence, based on the data on energy consumption in the municipality of Joinville [55], the scenario proposes that 20% of the water demand for generating power (evaporated water in the hydroelectric production reservoirs) will be substituted using photovoltaic energy production. The result will decrease the evaporation loss from the hydroelectric generation, increasing the water availability for other uses. The Stephens and Stewart (1963) equation was used for calculating the liquid evaporation [67,68,69], considering the energy production reservoirs in the eastern region of Santa Catarina State. Finally, the ratio between the evaporated volume and the energy produced resulted in a net loss for each kilowatt-hour generated (m³/KWh), characterizing the water loss (demand) by net evaporation in the reservoirs, as well as the water demand for producing the energy consumed in this region of the state. The calculations resulted in a value coefficient of 0.022 m³/KWh.

The Action Plan present in the Cubatão River Water Resource Watershed Management Master Plan [49] and the Santa Catarina State Water Resource Plan [70] were used for the C_GES scenario. In Santa Catarina State, a water grant (an official authorization for the use of water) is the only authorized management instrument cited in the State Water Resource Policy [71]. However, in the National Water Resource Policy, in addition to Watershed Plans, classifying water bodies, the charge for using water resources, the compensation to municipalities, and the water resource data system are characterized as management instruments [33]. Hence, water-saving and reuse measures were adopted in the basic sanitation, industrial, and agribusiness sectors. For sanitation, a hypothetical 50% reduction was proposed in losses in the supply systems of BHRC, which currently are 42.19% [38]. Thirty percent water reuse was proposed for the industrial sector [72,73]. A 20% reduction in demand was proposed for the agribusiness sector by improving process efficiency involving that sector [74].

The C_MCL scenario proposes the analysis of water availability based on the scenario of future emissions of greenhouse gases (GHG) for the Representative Concentration Pathways (RCPs) 4.5 and 8.5 (intermediate and pessimistic scenarios, respectively) of the Intergovernmental Panel on Climate Changes (IPCC) [75]. It utilizes monthly data on total precipitation, average temperature, relative humidity, and wind speed from the Brazilian Climatic Forecast Portal (http://pclima.inpe.br/, accessed on 30 August 2022) from the National Institute for Space Research (INPE) for the climatic model ETA-MIROC5 [76,77]. The flow data were simulated in WEAP, compared to the reference flow scenario values C_REF and the evaluations regarding availability compared to the maximum grantable flow (50% of Q98) [78]. The water demand for energy production was considered for the water–energy–food nexus analysis. There are no hydroelectric power plants installed in the BHRC territory. However, a great amount of energy is consumed in performing its activities [55]. The water demand for generating energy compared to water evaporation from generating energy in the studied area was applied to account for the impacts caused by the energy consumption related to the nexus. Notice that this was not accounted for in the BHRC water balance, but it was considered in this work for analyzing the nexus as to how there is a sectorial interconnection [13,79,80].

Finally, the C_INT scenario utilized the same proposed measures in the C_EAL and C_GES scenarios (the substitution of hydroelectric energy generation by photovoltaic and using the actions proposed by the Water Resource and Watershed Plans) to evaluate the integration of the sectorial measures and analyzing the reduced water demands compared to the C_REF reference scenario and the others.

2.6. The WEF Sustainability Index

According to Guan et al. (2020), Liu et al. (2022), and Momblanch et al. (2018) [28,29,30], based on the approach developed by Daher and Mohtar (2015) and Daher et al. (2019) [80,81], an SI Sustainability Index was developed (Equation (1)) for the comparison among the proposed scenarios. Four variables were considered for water, energy, and food: (i) the total use of water in the demand sectors (m³); (ii) the cumulative refilling from groundwater (m³); (iii) the energy used for water (kWh/m³), basd on Moraes-Santos et al. (2021) [82]; and (iv) the total production of food (kg) considering rice as the reference crop and productivity according to Sidra/IBGE (2021) [39,41].

$$I{S}_C={{\displaystyle \sum }}_{i=1}^{4}IR{S}_{S,i}$$

(1)

where IRS_S,i is the sustainability of an individual resource i, calculated according to Equation (2):

$$IR{S}_{S,i}=\frac{R_{S,i}}{R_{REF,i}} w_i$$

(2)

where RS,i is the i resource in the scenario being evaluated; R_REF,i is the i resource in the reference scenario; and w_i is the coefficient of the adopted weight. The adopted w weight was 0.25 for all scenarios. It is used for the 4 variables, and considering that the sustainability of resources was considered equally important [28,29,30], the same weight was used (the summation of variables equal to 1.0). The IRS_S,i was multiplied by −1 for items i and iii to reflect the negative impact on local sustainability. Finally, the SI with the highest and positive value will display the most sustainable scenario.

3. Results and Discussion

3.1. Model Construction

The monthly water demand from BHRC totaled 7,865,790.29 m³/monthly based on the data added to the model and according to SDS (2011) and SIOUT (2021) [52,53]. The following Figure 2 shows the model study area.

3.2. Model Calibration and Validation

The first simulation was performed based on the data obtained in the literature and other data sources (item 2.3). The simulated values do not correspond to the observed values, presenting statistical coefficients that differ from them (Moriasi et al., 2007) [63], as it is necessary to perform model calibration and validation. Hence, the parameters were calibrated using the PEST tool and then adjusted manually to improve the statistical performance (NSE, P_bias, R², and d). The following hydrograph (Figure 3) displays the observed flows (blue) and simulations (red) for BHRC and the statistical model performance (

Table 3. Performance evaluation of the model simulation (calibration and validation) of the BHRC.

Simulation Step	Precision Statistics
	NSE	Pbias	R2	D
Calibration	0.602	0.545	0.618	0.882
Validation	0.488	−8.501	0.796	0.909

).

The simulated flows abide by similar trends to the observed data based on the hydrograph (Figure 3). However, some presented periods display underestimated or overestimated flows, especially for the maximum ones. The model showed satisfactory performance for the three performance statistics, except for NSE (Table 3). The Pbias presented good performance (Pbias < ±10%) [63], presenting the underestimation and overestimation bias in the calibration and validation steps, respectively. The R² results were satisfactory (collinearity between the simulated and observed data). For “d”, the results were 0.882 and 0.909 during the model calibration and validation, respectively.

Tena et al. (2021), Dill et al. (2022), and Khalil et al. (2018) [58,83,84] also achieved similar results to thos obtained in this work for the indexes mentioned above. NSE is influenced by high flows, leading to low coefficients [85], which can be considered for this work; some periods presented overestimated flows based mainly on the validation (Figure 3). The low degree of precision of peak season flows can be explained by the existing difficulties of measuring when there is flooding [86]. Operational problems during these events, such as flooding at the measurement stations and damaged equipment, cause records gaps, affecting the data’s reliability. Abbaspour (2022) [64] confirms that the quality of the input data set in a model directly affects the NSE and R². These parameters have no definite values to define whether a simulation is efficient. In the BHRC modeling, Dill et al. (2022) [83] also observed NSE values lower than those recommended by Morasi et al. (2007) [63] during the calibration and validation. The authors pointed out the presence of only one fluviometric station in the study area and the presence of a rock riverbank protection (belonging to the municipal sanitation company) being constantly modified, and the absence of information on these changes as factors that directly influenced the precision of flow data, consequently compromising the performance of the model. The values for the method parameters on the calibrated soil humidity for the BHRC are presented in

Table 4. WEAP parameter values after the BHRC calibration.

WEAP Parameter	Values	Units
Preferred Flow Direction—F	0.68	dimensionless
Crop Coefficient—Kc	0.40–1.20	dimensionless
Runoff Resistance Factor—RRF	2.00–3.00	dimensionless
Soil Water Capacity—Sw	480	mm
Initial Z1	13	%
Root Zone Conductivity—Ks	188	mm.month−1

.

3.3. Water Demand Future Scenarios

The water demand scenarios were estimated based on item 2.5 (the population growth, Industrial GDP, and agricultural areas) for the 2020–2050 time horizon. Ten-year periods were simulated for the BHRC flows (Figure 4) for each one of the proposed scenarios (C_REF, C_EAL, C_GES, C_MCL 4.5, C_MCL 8.5, and C_INT), considering the proposed modifications in each scenario regarding water usage (substitution of the hydroelectric energy portion by photovoltaic, utilization of water resource management instruments, the effect of climatic changes, and the integration among scenarios), which were analyzed compared to current growth trends (C_REF scenario).

3.3.1. Water Demand for the Reference Scenario (C_REF)

The BHRC presented an increase of 76.77% for the entire period, or 1.92% increased monthly water demand flow, according to Figure 4, for the reference scenario (C_REF). The observed population growth and industrial GDP for the next few years, shown in Table 2, contributed to the increased water demand in these sectors. By the end of the simulation period (2020–2050), the total demand for this scenario will be 3834.43 hm³. These values were utilized to compare the proposed alternative scenarios. The results show that mainly due to the increased population and the industrial sector, the demand for water resources will increase by 2050 in the BHRC. Consequentially, energy and food consumption will also increase in the region, causing a dispute for resources in the activities linked to these sectors.

3.3.2. The Water Demand Scenario for Alternative Energy Sources Demands (C_EAL)

There was an 8.99% reduction loss by liquid evaporation from the reservoirs that generate hydroelectric energy for the BHRC by substituting the hydroelectric production with photovoltaic related to C_REF in the C_EAL scenario (Figure 4). It is worth emphasizing and accounting for the impacts caused by energy consumption in the water–energy–food nexus. Water demand is based on the amount of water that evaporated from generating energy and is considered as what was consumed in the study area. However, it must not be counted for the water balance, as it is considered only for nexus analysis purposes since it is considered a sectorial interconnection. The water demand will be 3520.13 hm³ by the end of the simulation period, presenting a reduction of 314.30 hm³ compared to the C_REF scenario.

Such studies as Guan et al. (2020) and Mounir et al. (2019) [28,87] showed that alternatives in the diversification of the energy matrix by using photovoltaic can reduce the water demand, as well as carbon dioxide emissions (CO2) due to energy generation from fossil fuels. The region presents different socioeconomic activities and water users characterized by the user sectors and reflected in their activities. However, despite other interests, management must understand and relate to existing connections among sectors [11,12].

Photovoltaic solar energy has not been exploited very much in Brazil. In Santa Catarina State, only 1.35% of energy production originates from solar power [56]. Even though there is low irradiation present, Santa Catarina State generates more energy compared to a sunnier location in Germany, a country leading in installed generation capacity at 46 GW (the overall horizontal irradiation rates are 1534 KWh/m² and 1.241 KWh/m² for Santa Catarina State and Germany, respectively) [88]. However, there are no robust regulating initiatives focused on this sector. Therefore, the development of measures that bolster the photovoltaic energy market in all its aspects, ranging from research investments to incentives to suppliers, retailers, and potential customers (electric energy self-generators), makes it necessary to increase the utilization of this source [89,90].

Solar farms and parks are one of the existing interconnections for photovoltaic energy production in the water–energy–food nexus. These refer to large extensions of lands previously used for farming and now used for renewable energy production. The profitability has attracted landowners to substitute food production with these structures, which can threaten food security and the dispute for lands in both segments [91]. Thus, it is necessary to plan the space for its expansion adequately. This also shows the importance of intersectoral planning while considering the trade-offs of the nexus sectors, for example, the energy sector related to the others [92]. There are innovative initiatives to decrease these trade-offs, such as agrivoltaics, making it possible to use the land for both purposes, agriculture and photovoltaic energy generation, thereby increasing the land’s efficient use [93,94].

In the urban context, using photovoltaic panels on buildings is also associated with nexus interconnections. Due to the urbanization of cities, projects have been conceived for achieving good energy performance and have been presented as challenging. Cultivating the development of neighborhoods focused on maximizing solar energy production provides independence to existing buildings, spreading the concept of “Net Zero Energy Buildings” capable of producing all or even a surplus of the energy they consume [95].

3.3.3. Water Demand for the Scenario of Increased Usage of Water Resource Management Instruments (C_GES)

There was an average reduction of 16.32% in the flow of the water demand by adopting the C_GES scenario from adopting the water resource management practices proposed by the Watershed Master Plan [49] and by the State Water Resource Master Plan [70] described in item 2.5, compared with the C_REF scenario. By the end of the simulation period (2020–2050), the demand for the total volume will be 3205.18 hm³, a reduction of 629.25 hm³ compared to the C_REF scenario.

The importance of managing water resources is stressed by using Action Plans, focusing on maximizing the availability of water resources regarding supply and demand [96]. Currently, at the statewide level, only grant authorization is used as a water resource management instrument [71]. However, the National Water Resource Policy contemplates and authorizes another four instruments that can help manage, such as classifying bodies of water, the charge for using water resources, compensation to municipalities, and Information Systems on Water Resources [33]. Furthermore, policies and legislative measures linked to water resources must be periodically updated. The BHRC Watershed Master Plan was produced in 2007 [49]. Therefore, it was confirmed that revising and updating the plan is necessary to improve the understanding of the current context of the region (population increase, modification in soil use and occupation, waste product emissions, etc.).

The actions for reducing water consumption must be addressed holistically, considering socioeconomic, political, environmental, and technological factors and the involved entities [97]. The rapid urbanization of cities has demanded sustainable water management solutions to maintain these regions’ development due to water scarcity [98]. Despite the low degree of public acceptance related to reusing water, the recovery of these solutions, mainly for non-potable purposes, can significantly reduce the water demand and promote strategies for this resource’s sustainability [99]. The actions can go beyond just reducing losses from distribution in human supply. Measures such as varying the price of consumed water, offering discounts, rationing in periods of scarcity, constructing efficient projects, and educational and awareness campaigns can attract attention to the population and contribute to reducing water consumption [96].

All water demand sectors are interconnected regarding the existing interconnections between water, energy, and foods [13,100]. The BHRC needs water, especially for human supply, but also for industry, which includes diverse sectors, including textile, metalworking, appliance, and others. Hence, the importance of the integration and development of a “nexus thinking” focused on the region’s development is evident insofar as it breaks through the limitations of single resource management and helps to increase the security and efficiency of multiple resources [101]. Based on the three sectors, the proposed solutions must be developed in an integrated manner, disregarding the fragmented thought or as a “silo,” as generally occurs, for promoting the maximization of existing synergies in activities performed in different regions of the state, by developing more integrated and participative solutions [10,102]. In this case, the Watershed Committees play roles as mediators in managing the process based on a nexus thinking pattern. Their assemblies and representatives of the user sectors can debate and arrive at solutions, enabling increased sustainability in their territories.

3.3.4. Water Demand for the Climatic Change Scenario (C_MCL)

The results for the C_MCL scenario were obtained through the hydrographs for the RCP 4.5 and RCP 8.5 scenarios [75]. These were compared to the C_REF hydrograph scenario. The water demand flow was considered the same as for the C_REF scenario, as the water supply was analyzed based on the increase in the BHRC. Visually analyzing the hydrographs for the C_REF, C_MCL 4.5, and C_MCL 8.5 scenarios (Figure 5) shows that the Cubatão River flows for the RCP 4.5 and 8.5 scenarios will be lower compared to the foreseen flows for the C_REF scenario.

The average flows for the hydrologic year in the C_MCL 4.5 and 8.5 scenarios will be 27.81%, respectively (

Table 5. Monthly average flows (m³/s) for the 2020–2050 period in the C_REF, C_MCL 4.5, and C_MCL 8.5 scenarios for the BHRC and the percentile difference compared to the C_REF scenario.

Month	Flow (m3/s)			Difference (%)
	C_REF	C_MCL 4.5	C_MCL 8.5	REF—4.5	REF—8.5
January	33.99	21.22	18.45	−37.57%	−45.72%
February	32.80	19.85	19.22	−39.48%	−41.40%
March	27.33	16.67	16.88	−39.00%	−38.24%
April	17.32	11.88	12.57	−31.41%	−27.42%
May	16.20	8.95	9.20	−44.75%	−43.21%
June	14.38	8.22	8.43	−42.84%	−41.38%
July	14.04	8.79	8.95	−37.39%	−36.25%
August	11.94	12.07	10.79	1.09%	−9.63%
September	17.91	16.93	14.86	−5.47%	−17.03%
October	20.48	20.48	19.18	0.00%	−6.35%
November	22.67	18.40	19.65	−18.84%	−13.32%
December	21.55	17.45	21.13	−19.03%	−1.95%
Entire Hydrologic Year	20.88	15.08	14.94	−27.81%	−28.45%
Wettest Season	24.90	19.06	18.75	−23.47%	−24.71%
Driest Season	16.87	11.10	11.14	−34.22%	−33.98%

), according to an analysis of the simulated flows for the entire period, which are 27.81% and 28.45% lower compared to the C_REF reference scenario. There was only an increased flow in the C_MCL 4.5 scenario related to C_REF (1.09%) in August. There was no increase or reduction in the average flow for C_MCL 4.5 in October. The average estimated flows for the wettest season (September–February) and driest season (March–August) were lower compared to C_REF. The reductions were 23.47% and 34.22% for the C_MCL 4.5 scenario for the wettest and driest seasons, respectively. The wettest and driest seasons were 24.71% and 33.98% for the C_MCL 8.5 scenario, respectively.

There are no more available water resources in the BHRC based on the monthly flow evolution of water demand compared to the grantable flow from the Cubatão River Mouth (50% of the Q98) (Figure 6). No more uses of water resources can be authorized in the BHRC. According to SDS (2008) and SIOUT (2021) [52,53], the authorized flow for surface runoff water is currently 2.130 m³/s; thus, the remaining is only 0.072 m³/s, which can be authorized in the region of the Cubatão River Mouth, 2.202 m³/s [103]. The authorized current flow for the groundwater is 0.007 m³/s [52,53]. Recall that the water removed for generating energy in the BHRC was not considered for the water balance, however, to perform the water–energy–food nexus analysis and consider the premise that the nexus analysis is based on integrated systems [3,6].

Dill et al. (2021) [104], when evaluating the climatic change scenarios for BHRC, used the MIROC-ESM-CHEM model and presented the reduced flow for the RCP 4.5 and 8.5 scenarios. The percentage reductions compared to the historical scenario were 41% and 47.8% for the RCPs 4.5 and 8.5, respectively, for the entire hydrological year. According to the authors, the lower flow results can be explained using the climatic variables from the respective RCPs, such as precipitation and temperature, the historical series, and the hydrological simulation software that can under- or overestimate the simulation results regarding the observed values.

The monthly total precipitation, average temperature, relative humidity, and wind speed are the climatic variables used for simulating the flow in WEAP. The RCP 4.5 and 8.5 scenarios presented the highest precipitation and temperature indexes in their series that simulate lower flows when associated with other variables [104]. The water usage conflicts would be installed, putting pressure on the management system to arbitrate them due to the decreased flows in the drier seasons [105].

The effect of climatic changes will change the river flow regime, as the precipitation and temperature scenarios of climatic changes are factors that affect the water runoff regime [83,105,106]. Insufficient water for the coming years is a problem that already needs to be faced, and climate change is a situation that is worsening. Brazil has already faced some degree of water stress. There is increasing uncertainty about the future availability of water and the impact climatic changes will have on Brazilian water resources. For example, in 2000–2002, Brazil faced the worst hydroelectric crisis due to water scarcity that impacted hydroelectric generation. The Alternative Sources of Electric Energy Incentive Program (PROINFA) was created because of the crisis in 2002 that propelled the diversification of renewable energy sources in Brazil [92].

Thus, it is necessary to employ water resource management instruments and the C_GES scenario to face the condition of extreme hydrological events [107], and PNRH indicates in its foundations that in situations of water scarcity, “the priority use of water resources is for human and animal consumption” [33]. The importance of the Santa Catarina State Water Resource Plan is evident in its third objective, which addresses “Increasing the resilience related to critical hydrological events” [107]. The general target for this objective is to “reduce the average number by 25% of those affected by floods and droughts 2017–2027 compared to 1991–2016”. As well as activities focused on management, sectorial actions must also be performed to maximize water availability in quantity and quality [107]. The plan includes steps for monitoring the quali–quantitative aspects of surface and groundwater, reuse, supply system efficiency, conservation techniques, optimization of agribusiness and industrial uses, the construction and analysis of multiple usages of reservoirs, and the recovery of degraded areas [70].

3.3.5. Water Demand for the Integrated Scenario (C_INT)

There was an average reduction of 24.57% in the C_INT scenario, using the C_EAL and C_GES scenario proposals compared to the C_REF scenario. By the end of the simulation period (2020–2050), the total water demand will be 2890.87 hm³, presenting a reduction of 943.56 hm³ compared to the C_REF scenario. The BHRC consumption flow was maximized by integrating the C_EAL and C_GES scenario practices. Kou et al. (2018) and Amin et al. (2018) [25,108], when incorporating the scenarios based on the previously proposed proposals in their studies, also achieved significant reductions compared to their reference scenarios. However, combining multiple strategies can reduce water consumption, increase water availability for diverse purposes, and minimize the periods of possible scarcity.

Furthermore, incorporating measures to reduce water consumption in diverse sectors is associated with proposed integrated planning using the water–energy–food nexus approach [6,7]. It is possible to offer joint solutions and improve the sustainability and security of resources in general by analyzing existing connections between the water demand sectors [12]. Concepts such as the circular economy and life cycle are directly linked to nexus. They can be attributed to the development of multisectoral solutions, thereby reducing wastage and adding value to products and materials [109,110].

3.3.6. Evolution of Water Demand Based on the Proposed Scenarios

Finally, Figure 7 shows the evolution of water demands (m³) for the horizon of this study (2020–2050) in each of the proposed scenarios. By 2050, the BHRC will present a demand of 13.90 Mm³ (millions of m³), considering the current growth rate (C_REF scenario). Based on the applications of the C_EAL, C_GES, and C_INT scenarios, that demand will present a reduction of 1.08 Mm³, 2.37 Mm³, and 3.46 Mm³, respectively.

According to the Sidra/IBGE data (2021) [39], in the year 2019, the GDP in the municipality of Joinville was BRL 34,528,619,000.00. The contribution from the industrial sector to the GDP is 25.88% of the municipality. Therefore, the impact of water availability for the coming years, by implementing the water demand scenario, can guarantee that the water user sectors continue using water resources in their production processes, contributing to the development of their regions [111].

3.4. Sustainability Index

Four variables were considered for calculating the WEF Sustainability Index (SI): (i) the total use of water in the demand sectors (m³), (ii) the cumulative refilling of groundwater (m³), (iii) the energy used for the water (kWh/m³), and (iv) the total production of foods (kg) for the reference crops used during the entire simulation period (2020–2050). The resultant values for the IRS of each variable for this simulation are displayed in

Table 6. The Sustainability Index (SI) parameter results by the end of the simulation period (2020–2050).

WEF—BHRC Sustainability Index Parameters
Scenario	Parameter I 1	Parameter II 2	Parameter III 3	Parameter IV 4
C_REF	3,834,443,824.00	29,990,510,592.00	2,684,103,676.80	222,367,574.15
C_GES	3,205,182,576.00	29,990,510,592.00	2,243,627,803.20	222,367,574.15
C_EAL	3,520,126,752.00	29,990,510,592.00	2,464,088,726.40	222,367,574.15
C_INT	2,890,875,480.00	29,990,510,592.00	2,023,612,836.00	222,367,574.15
C_MCL_4.5	3,834,433,824.00	29,450,862,592.00	2,684,103,676.80	222,367,574.15
C_MCL_8.5	3,834,433,824.00	29,436,542,976.00	2,684,103,676.80	222,367,574.15

¹ Parameter I: Total water use in the demand sectors (m³); ² Parameter II: Cumulative Refilling of the groundwater (m³); ³ Parameter III: Energy used for the water (kWh/m³); ⁴ Parameter IV: Total food production (kg).

. Figure 8 shows the Sustainability Index (SI) for the proposed scenarios.

The index is defined as zero for the C_REF scenario, as the other scenarios will be compared related to the current growth scenario or business as usual. The implementation of management actions is focused on the sectors promoting water demand reduction strategies that produce a positive SI result (C_GES scenario) and, therefore, is a more sustainable scenario than the current situation. The substitution of hydroelectric generation by photovoltaic (C_EAL scenario) also presented a positive SI; however, a little lower when compared to the others and higher compared to the climatic change scenarios.

The climatic change scenarios (C_MCL 4.5 and C_MCL 8.5) presented a negative SI, considering that the water availability will be lower when compared to the C_REF scenario because of the climatic changes in the study area. According to the INPE data (2022) [76] simulated in the WEAP, for the RCP 4.5 and 8.5 scenarios in the southern Brazilian region, the foreseen flows are less than when compared to the C_REF scenario. Although there is a small difference between the results (−0.0045 and −0.0046 for RCPs 4.5 and 8.5, respectively), the scenario for the RCP 8.5 emissions presented a lower index than the RCP 4.5 emissions, as it is characterized as more unfavorable.

Finally, the C_INT scenario presents a higher positive SI, which can be explained due to the integration between the C_GES and C_EAL scenario actions. After implementing the sectorial management actions and using photovoltaic energy as a portion of the hydroelectric generation, the demand for water resources decreased, bringing about a higher SI.

Based on the proposed measures in the scenarios, the SI measured how sustainable a scenario would be compared to the reference. The C_INT Integrated Scenario presented the highest coefficient (0.1230) and, therefore, presented as the most sustainable panorama; meanwhile, C_MCL 8.5 presented the lowest coefficient (−0.0046); consequently, it was the least sustainable scenario. Hence, using the proposed practices in the evaluated scenarios, the integration between the proposals produced a more sustainable solution when compared with the reference scenario, while the effect from the low levels of precipitation and temperature for the RCP 8.5 scenario resulted in the lowest availability of water resources.

For example, the International Energy Agency (IEA, 2022) [112] emphasized understanding the water–energy–food nexus to avoid unintentional consequences and focus on the importance of considering water use in decisions in politics and other sectors. A great number of countries have already faced some kind of water stress or scarcity. Hydroelectric energy generation plays a key role in the Brazilian renewable electricity matrix and is especially vulnerable to climatic variability.

Previous studies by Guan et al. (2020) and Liu et al. (2021) [28,30], who used sustainability indexes in their works, also achieved similar results when they adopted sustainable energy usage and management practices. When they simulated water scarcity scenarios, the authors obtained negative SI results due to the possible insufficient water to supply the current demands.

Hence, SI can be a tool for managing water resources by involving decision-makers and entities. Based on the analysis of future scenarios, it is possible to list the gains and the existing obstacles so that the developed proposals can effectively sustain and guarantee resources [26]. SI not only makes performance analysis possible but also makes a comparison between policies, considers alternative measures, and quantitatively analyzes the existing dynamics. Hence, it is possible to consider how adverse impacts can be reduced and consequently achieve more sustainable conditions supplying the water needs to human beings and the environment [27,113].

4. Conclusions

This research sought to identify and understand the dynamic operation of a watershed in the Atlantic Rainforest domain in Santa Catarina State, Southern Brazil. It is based on the water–energy–food nexus approach. It has sought to understand the interconnections between the sectors that utilize these resources through water demand analysis using the WEAP 2021.0203 software program. The modeling and analysis are based on future scenarios of water demand. The BHRC presented a predominantly industrial and human supply demand. The model’s performance was satisfactory, despite one of the metrics’ values being below the level recommended by the literature. This can be explained by the absence of a quality data set, which is characterized as a limiting factor. However, the model satisfactorily represented the observed flows based on the observed data throughout recent years.

The water demand is estimated to increase by 76.77% by 2050 compared to the current scenario. Implementing the future water demand scenarios shows that using water resource management actions can reduce a portion of the water demand, preserve and maximize resource utilization, and promote the assurance of these resources for the coming generations. Insufficient water for the next years is already a problem, and climate change scenarios are aggravating the situation. A possible socioeconomic drought condition can occur. This situation can be potentialized because the information supporting technical analyses and simulations is scarce. Improvements in monitoring points are needed. This leads to critical situations because decisions are made based on few data. Water management needs planning, technicians, human resources, and equipment. However, there the necessary financial resources to maintain this structure are lacking. This has important consequences, because the Cubatão River watershed has high production of goods and services, and the population is growing. It is necessary to act preventively and provide the legal provisions to arbitrate the conflicts that will intensify further.

Utilizing the Sustainability Index can contribute to supporting decision-makers in evaluating proposals and analyzing the effectiveness of the implementation of scenarios. The results have revealed that Santa Catarina State needs to implement management actions focused on mainly the industrial and human consumption supply sectors. It is necessary to propose adaptation plans and actions promoting the reduction in water demand and subsidies to guide the industrial sector on conscientious water usage to not affect their future productivity. Thus, policies dedicated to integrating water resource usage in these sectors must be developed to maximize the synergies and reduce existing trade-offs. There must also be plans and policies considering the interdependency of sectors so that the proposals can be formulated to be more integrated. Managing only one sector is not an option, due to the context of a climatic crisis. Water, energy, and food must be managed holistically, and intersectoral planning is necessary so that the solutions focused on only one sector do not negatively impact other sectors. This work summarizes the application of WEF nexus thinking in a useful way to promote systemic management. This approach promises to bridge gaps by reducing fragmentation, not only among water management sectors but also by going beyond and supporting more coherent and integrative policies.