Impact of climate change on the irrigation water demand of rainfed crops in the Southern and Steppe regions of Rio Grande do Sul, Brazil

Abstract

Proper management of natural resources is essential to improve agricultural productivity and reduce potential losses. The PROJETA platform provides global climate models regionalized by the ETA model, enabling more accurate climate projections for specific regions. This study aimed to evaluate the influence of climate change on irrigation water demand using three global climate models (CANESM2, HADGEM2-ES, and MIROC5) regionalized by the ETA model for the baseline period (1961–2005) and two future scenarios (RCP 4.5 and RCP 8.5). The scenarios were subdivided into three future periods and analyzed for 27 municipalities located in the Southern and Steppe regions of Rio Grande do Sul using the SWAP (Soil, Water, Atmosphere and Plant) model. The results indicate variations in irrigation water demand among models and scenarios. Some projections suggest an increase in irrigation requirements in certain municipalities, while others indicate a reduction when compared to the baseline period. The projected increase in water demand is mainly associated with rising temperatures, which may intensify crop transpiration and soil water evaporation. These findings highlight the importance of climate adaptation strategies and improved water resource management to support sustainable agricultural production under future climate conditions.

Key words
Climate Models; RCP 45 and 85; SWAP model; Water Availability

INTRODUCTION

Climate change is increasingly present in our daily lives and can be observed through the occurrence of extreme events that take place around the world, such as the increased frequency of hurricanes, heat waves, droughts, and floods (Alcântara et al. 2020). According to Assad et al. (2020), these possible changes will directly affect Brazil and the world, causing impacts in several areas, where the energy sector, urban infrastructure, and agriculture, among others stand out.

Agriculture has a direct relationship with the climate, which affects the growth and development of plants in their various stages (Souza et al. 2018), which creates serious problems for farmers. Climatic variations are determining factors for the development of cultures in a region, since they directly affect their development and, consequently, their productivity, and it is essential to understand their behavior in order to carry out prior and adequate planning to reduce possible negative impacts (Costa & Queiroz 2021).

The state of Rio Grande do Sul (RS) has agriculture as one of the main pillars of its economy and has long been recognized for its significant contribution to national agricultural production, supplying both domestic and international markets. The state is also considered the largest contributor to the Gross Value Added (GVA) of Brazilian agriculture, occupying a strategic position in the national supply of several products, such as tobacco, soybeans, and wheat (Feix Júnior 2019). However, in 2020 the economy of Rio Grande do Sul experienced a decline in economic activity, largely influenced by the rainfall shortage observed during the first half of the year and by the COVID-19 pandemic. Although these factors are external to the economic system itself, they directly affected the state’s economic performance (Black 2020).

In addition, Rio Grande do Sul plays a key role in the national food supply and has long been known as the “Brazilian granary,” due to its significant contribution to both domestic markets and exports. The state occupies a strategic position in the production of major crops such as soybeans, tobacco, and wheat. It is also the leading national producer of tobacco, particularly in smallholder systems, with municipalities such as Canguçu, São Lourenço do Sul, and Venâncio Aires standing out. Furthermore, corn production in the state shows considerable growth potential, highlighting the importance of public policies aimed at expanding irrigated areas of rainfed crops, which have already experienced significant growth over the past decade (Leusin Júnior et al. 2025).

The IPCC (Intergovernmental Panel on Climate Change) is an international scientific body that provides data on climate change and also publishes reports addressing possible future impacts and risks. The report (AR5), considers that global warming is a reality and addresses future climate scenarios as Representative Concentration Pathways (RCP) (IPCC 2013). More recently, the Sixth Assessment Report (AR6), published in 2023, provided a more comprehensive evaluation of the climate system and reinforced the evidence that global warming may exceed an increase of 1.5 °C during the 21st century if mitigation measures are not implemented (IPCC 2023). The process of developing the Seventh Assessment Report (AR7) began in 2025 and is currently ongoing (IPCC 2025).

In light of the above, in addition to the original climate variables, the estimated variables such as, for example, reference evapotranspiration, real evapotranspiration, water deficit, and radiation balance, among others, are of paramount importance for carrying out work on climate characterization and agrometeorological modeling (Cordeiro et al. 2019).

The SWAP model (Kroes et al. 2008) is a one-dimensional model that allows the simulation of several intrinsic variables in the relationship between the soil, plant, water, and atmosphere system. Through equations such as Richards and Mualem-Van Genuchten, it is possible to project the movement of water in the soil, defining different boundary conditions.

Thus, the present work aims to evaluate whether, given two climate scenarios (RCP 4.5 and RCP 8.5), the irrigation water demands of rainfed crops in the Southern and Steppe regions of RS will be different in the future (2006 to 2099) when compared to the baseline period (1961-2005).

MATERIALS AND METHODS

Study area

The Southern and Steppe regions (Figure 1) cover 29 municipalities (FEE 2015), and their main activity is agriculture. The Southern region contributes with 7.7% of the state’s total agricultural production, and the Steppe region with 17.2% (SEPLAN 2015). For the development of this study, 27 municipalities were considered, since in a preliminary analysis of the climate projection data available on the PROJETA platform (https://projeta.cptec.inpe.br/), from the CPTEC/INPE, two municipalities (Aceguá and Chuí) did not present precision in their data, and the points of the climate models were in an oceanic area. Therefore, they were excluded from the present study.

Figure 1
Map of RS and Southern and Sttepe Regions.

Input meteorological data required by the SWAP model

The SWAP model, an open-source agro-hydrological simulation software, has been widely used in studies aimed at evaluating irrigation water demand, water use efficiency, and other analyses that relate climatic conditions to crop performance, such as those conducted by Behizadi et al. (2026), Heistermann et al. (2026) and Huang et al. (2026). The model allows users to select input variables according to the objectives of the study. In the present work, the following climatic variables were used as input data: precipitation (mm), relative humidity (U), minimum air temperature (T_min), maximum air temperature (T_max), wind speed (V), and solar radiation (R). These variables were selected because they are required for the estimation of reference evapotranspiration and crop water demand, which are key components of the soil water balance simulated by the SWAP model. The soil water balance within the soil profile is computed through the numerical solution of the nonlinear Richards equation (Richards, 1931), expressed as:

$\frac{\partial \theta }{\partial t}=\frac{\partial }{\partial z}\left[k \left(\theta \right)\left(\frac{\partial h}{\partial z}+1\right)\right]-S_a\left(h\right)- S_d\left(h\right)- S_m$ (1)

Where S_a(h) represents water uptake by plant roots, S_d(h) denotes water outflow through subsurface drainage systems, and S_m(h) accounts for the exchange of water between the soil matrix and macropores associated with preferential flow.

The Richards equation is numerically discretized using an implicit finite difference scheme, which allows the simultaneous simulation of water flow in both saturated and unsaturated zones of the soil profile.

Since the present study aimed to simulate water demand exclusively for rainfed crops, and assuming that water uptake by plant roots is equivalent to plant transpiration, the terms representing drainage and preferential flow through macropores in the Richards equation were disregarded. Therefore, the Richards equation applied in this study was expressed as:

$$\frac{\partial \theta }{\partial t}=\frac{\partial }{\partial z}\left[k \left(\theta \right)\left(\frac{\partial h}{\partial z}+1\right)\right]-S_a\left(h\right)$$ (2)

For the estimation of future water demand (I_d), it was assumed that the entire water requirement of the crop would be supplied through irrigation. Thus, irrigation demand was calculated as:

$$I_d= T_p- T_r$$ (3)

Where I_d represents the irrigation water depth required by the crop (mm), T_p is the potential crop transpiration (mm) under non–water-limited conditions, and T_r is the actual crop transpiration (mm), i.e., under water stress conditions. Therefore, when actual transpiration values are lower than potential transpiration values, the crop experiences water deficit conditions and irrigation become necessary. It should be noted that the average I_d represents the mean irrigation water depth required by the crops, calculated for each analyzed period.

The time series of these variables were obtained using data generated by CPTEC/INPE and made available on the PROJETA Platform, which are data from GCM regionalized by the ETA model that allow the prediction of climate phenomena with greater detail when compared to the models of general circulation (Chou 1996). The global models regionalized by the ETA model used were: CANESM2-ETA, HADGEM2-ES-ETA, and MIROC5-ETA, with a resolution grid of 20 km, and for each one data series were obtained for each municipality, which simulate the baseline period (1961 to 2005) and projections of the two future climate scenarios (RCP 4.5 and RCP 8.5) from 2006 to 2099.

For a better comparison of the future periods with the baseline period, the future was divided into three subperiods, being: F1 (2006-2040), F2 (2041-2070), and F3 (2071-2099).

Soil data required by the SWAP model

The soil classes with the greatest representation within the municipality were considered, also observing whether they are suitable for agriculture. This selection was performed with the help of the Quantum GIS 2.18.22 software, where shapes of soil classes provided by Santos et al. (2011) were plotted, thus generating the map of soil classes of the study region. From this, data from the water retention curve in the soil and saturated hydraulic conductivity (K_SAT) of the soil were sought, which are necessary for the simulation.

The data for these soil hydraulic variables, namely the parameters of the soil water retention curve and the saturated hydraulic conductivity (K_SAT), were obtained from the literature by searching for studies conducted in the municipalities of the study region. For locations where such information was not available, data were obtained from HYBRAS (Hydrophysical Database for Brazilian Soils), a database of Brazilian soils developed by the Hydrology Department of the Geological Survey of Brazil (CPRM), formerly known as CPRM (http://www.cprm.gov.br).

The soil types considered in the modeling correspond to the most representative classes in the study region and those suitable for agricultural activities. The following soil classes were included in the simulations: Eutrophic Haplic Planosols, Eutrophic Litholic Neosols, Haplic Litholic Neosols, and Dystrophic Red-Yellow Argisols.

To start the SWAP model, data on crop coefficient (k_c) (Table I), root depth (Table I), number of days each crop development stage lasts (Table II) and the leaf area index (LAI) (Table III) were necessary. In the simulation, for the number of days in the crop cycle, up to half of the period of the final crop development stage was considered, since generally most of them can develop without receiving water during the middle of this last period.

Table I
Values of the k_c and the average depth of the roots for the crops that stand out the most in the Southern and Steppe regions of RS.

Crop		Kc 1		Average depth of roots2
	Kc begin	Kc mid	Kc end	Begin	Mid	End
Onion	0.70	1.00	1.00	0.1	0.5	0.4
Barley	0.30	1.15	1.25	0.3	1.0	1.0
Bean	0.50	1.05	0.90	0.1	0.8	1.0
Tobacco	0.50	1.10	0.80	0.1	0.4	0.8
Cassava	0.30	0.80	0.30	0.1	0.4	0.8
Corn	0.70	1.20	0.60	0.1	0.8	1.4
Peach	0.80	1.15	0.85	0.1	0.4	0.8
Wheat	0.30	1.15	0.25	0.3	1.0	1.0
Soybean	0.50	1.15	0.50	0.1	0.5	1.0
Sorghum	0.70	1.00	0.55	0.1	0.8	1.4

• Source: Allen et al. 1998 ¹; FAO 2021 ¹; FAO (2021)²; Silva et al. (2018)²; Strategic R&D Project n° 010/2008 of ANEEL².

Table II
Crop development stages and sowing period.

	Development stages (day)
Crop	Begin	Develop	Mid	End	Total	Period
Onion1	30	55	55	40	180	Jul/Aug3
Barley1	20	50	60	30	160	May/Jul3
Bean1	15	25	25	10	75	Aug/Sep4
Tobacco2	20	30	30	30	110	Oct/Dec3
Cassava1	20	40	90	60	210	Sep/Nov5
Corn1	30	40	50	50	170	Oct/Dec4
Peach1	30	50	130	30	240	Jul/Dec6
Soybean1	20	35	60	25	140	Oct/Jan4
Sorghum1	20	35	40	30	130	Oct/Jan4
Wheat1	20	50	60	30	160	May/Jul

• Source: Allen et al. 1998 ¹; FAO 2021 ²; EMBRAPA & SEBRAE 2010³; CONAB 20174; Maluf et al. 20115; Raseira et al. 19986.

Table III
LAI of the main agricultural crops in the study area and their respective seeding and harvesting days.

Crop		LAI		Seeding Day	Harvest Day
	Begin	Mid	End
Onion1	0.5	3.5	1.5	09/17	03/16
Barley1	0.1	4.7	1.7	06/21	11/28
Bean2	0.1	4.0	2.0	08/24	11/07
Tobacco1	0.1	2.0	1.6	10/09	01/27
Cassava1	0.2	6.0	5.0	09/21	04/19
Corn2	0.1	4.8	2.0	10/07	03/26
Peach1	0.1	2.5	2.5	12/10	08/06
Soybean2	0.1	7.0	1.5	10/17	03/06
Sorghum1	0.1	4.8	2.0	10/08	03/06
Wheat2	0.1	2.7	1.7	07/11	12/18

• Source: Strategic R&D Project n° 010/2005 of ANEEL¹; Silva et al. 2018 ².

Anomalies

Anomalies in irrigation water demand (I_d) were also analyzed using the three GCMs regionalized by the ETA model under the RCP 4.5 and RCP 8.5 scenarios. The analysis considered the same division of future periods previously adopted: F1 (2006–2040), F2 (2041–2070), and F3 (2071–2099).

To determine the anomalies, the average annual I_d value was calculated for each future period (F1, F2, and F3) for each climate model and scenario. The anomaly values were then obtained using the following equation:

$$A_i=X_i- X_b$$ (4)

where A_i is the anomaly for future period i, X_i is the annual average for future period I, and X_b is the annual average for the baseline period.

In this study, the interpretation of irrigation water demand anomalies (I_d) was defined based on their relation to crop water requirements. Negative anomalies indicate an increase in irrigation water demand compared to the baseline period, reflecting conditions in which precipitation and soil water availability are insufficient to meet crop water requirements, thus requiring irrigation. In contrast, positive anomalies indicate a reduction in irrigation water demand, suggesting that climatic conditions are more favorable for crop development, with sufficient water availability to meet crop needs without the need for irrigation.

RESULTS AND DISCUSSIONS

Irrigation water demands (1961-2005)

The average I_d using the three climate models for the baseline period is shown in figure 2, where it is possible to observe that there is variation in the irrigation water demand values between them. This variability among climate models is expected, as different General Circulation Models (GCMs) adopt distinct physical parameterizations and climate sensitivities, resulting in differences in simulated precipitation and temperature patterns (IPCC 2013). It should be noted that the average I_d is nothing more than the calculation value of the average I_d for each analyzed period, that is, baseline period, F1, F2, and F3. The climate model CANESM2-ETA indicates that the municipalities of Santa Vitória do Palmar and Rio Grande have the lowest I_d values (less than 100 mm), and for the latter, the result is repeated in the projection of the MIROC5-ETA model. CANESM2-ETA also indicates that the average I_d will be 100 to 300 mm for the municipalities in the Sttepe region and the municipalities of Santana da Boa Vista, Pedras Altas, Jaguarão, Pedro Osório, Capão do Leão, Arroio do Padre, and São José do Norte. Canguçu has a higher average I_d, ranging from 600 to 900 mm, and the other municipalities from 300 to 600 mm.

Figure 2
I_d averages for the baseline period using the three climate models for the Southern and Sttepe regions a)CANESM2-ETA; b) HADGEM2-ES-ETA; c) MIROC5-ETA.

The HADGEM2-ES model estimated an average of water demand values between 100 and 300 mm for the municipalities of Bagé, Candiota, and Caçapava do Sul, in the Steppe region, in addition to Jaguarão, Pedro Osório, Capão do Leão, Arroio do Padre, Pedras Altas, Rio Grande, Santa Vitória do Palmar, and São José do Norte. A study conducted by Kulman et al. (2014), which analyzed drought occurrences at the municipal level in Rio Grande do Sul between 1981 and 2011, identified the municipality of Bagé as one of the most affected, with 22 recorded drought events. For the other municipalities in the Steppe Region, the model estimated an average I_d of 300 to 600 mm. For the municipalities of Pelotas, Morro Redondo, Canguçu, São Lourenço do Sul, and Tavares, HADGEM2-ES estimated average I_d values between 600 and 900 mm, while for the other municipalities in the Southern region average I_d values between 300 and 600 mm were estimated.

The average I_d estimated by the MIROC5 model for the baseline period is similar to the range of I_d estimated in the CANESM2 model for the municipality of Rio Grande, also presenting values smaller than 100 mm. The same similarity was observed in the municipalities of Arroio Grande, Amaral Ferrador, Cerrito, Morro Redondo, Turuçu, and Tavares, where the average I_d was between 300 and 600 mm. For some municipalities, the three climate models estimated ranges of equal average I_d values for the baseline period (between 100 to 300 mm; 300 to 600 mm, and 600 to 900 mm). The MIROC5 model estimated the highest average I_d value for the municipality of Canguçu (between 1100 and 1500 mm), which was the highest estimated value among municipalities and the three models. This result may be related to the fact that Canguçu is the municipality with the greatest diversity of cultures, and these have different development characteristics. According to Allen et al. (1998), crop water requirements vary significantly depending on crop type, growth stage, and climatic conditions, which may contribute to differences in irrigation demand. Concomitant to this, there may have been lower rainfall regimes in the summer of the baseline period, a season where most of the crops are developing and thus causing a greater demand for irrigation. In a study conducted by Collischonn et al. (2011) for the Quaraí River Basin, simulations under different water consumption scenarios indicated that water demand exceeds availability within the basin.

Irrigation water demands (CANESM2-ETA)

The CANESM2 model for the RCP 4.5 scenario for the three evaluated futures projected great similarity in the average values of I_d, as can be seen in figure 3. An average I_d value in the range of 100 to 300 mm until the end of the century was designed by the CANESM2 model for the Sttepe region and for the cities of Jaguarão, Pedro Osório, Capão do Leão, Tavares, and Santana da Boa Vista, which are part of the southern region. As for the municipality of Rio Grande, the model indicates, for the three future periods, an average I_d of less than 100 mm. For Arroio Grande and Pelotas, an average I_d of 600 to 900 mm is projected, while for Piratini, Pinheiro Machado, Herval, and São Lourenço do Sul the projection of I_d is in the range of 300 to 600 mm. According to the model, the average value of I_d in Canguçu will be in the range of 600 to 900 mm for periods F1 and F2, changing to 900 to 1100 mm in the last period (F3). For Santa Vitória do Palmar, the projection indicates an average I_d of less than 100 mm for F1, increasing to 100 to 300 mm in the last two futures. According to Pieper et al. (2024), the municipality of Canguçu experienced significant crop losses during the 2021/22 growing season due to increased water demand by crops.

For the RCP 8.5 scenario, the CANESM2 model (Figure 3) has the same average values of average I_d for the future F1, except for the municipality of Amaral Ferrador, which presented a higher average I_d and Turuçu and Capão do Leão, which have the highest I_d smaller average. In the future F2, there is an increase in I_d in the municipalities of Santa Vitória do Palmar, Rio Grande and Capão do Leão (from less than 100 mm to 100 to 300 mm). Pedras Altas and Piratini increased from 300 to 600 mm to 600 to 900 mm of I_d, and Pedras Altas from 100 to 300 mm to 300 to 600 mm. The others remain within the same range. In the last analyzed period, Santa Vitória do Palmar, Rio Grande, Bagé, Candiota, Piratini, Capão do Leão and Pedro Osório did not show an increase in the average I_d when compared to F2. The other municipalities all increase in the average I_d range, highlighting Pelotas and Canguçu, which go from 600 to 900 mm to 1100 to 1500 mm.

Figure 3
I_d averages using the CANESM2 climate model for the Southern and Sttepe regions of Rio Grande do Sul for: a) 2006 to 2040 for RCP 4.5; b) 2006 to 2040 for RCP 8.5; c) 2041 to 2070 for RCP 4.5; d) 2041 to 2070 for RCP 8.5; e) 2071 to 2099 for RCP 4.5; f) 2071 to 2099 for RCP 8.5.

The higher irrigation demand observed in Canguçu during the baseline period represents an interesting spatial pattern when compared to neighboring municipalities such as Capão do Leão and Pedro Osório, which present considerably lower Id values. This difference may be associated with a combination of climatic, soil, and agricultural factors. Although these municipalities are geographically close, Canguçu is located in the Serra do Sudeste region, which presents more complex topography and may influence local precipitation distribution, increasing the occurrence of short dry periods during the summer growing season.

In addition, soils in this region tend to present greater heterogeneity and, in some cases, lower soil water storage capacity than the lowland soils found near the coastal plain. Since the SWAP model simulates soil water balance based on soil hydraulic properties and atmospheric forcing, differences in soil water retention and precipitation distribution may directly affect the simulated soil moisture dynamics and, consequently, crop transpiration and irrigation demand. Soil hydraulic properties play a fundamental role in controlling water retention and movement in the unsaturated zone (Van Genuchten 1980), which directly influences water availability for crops.

Finally, the greater diversity of crops cultivated in Canguçu may also contribute to the higher simulated irrigation demand, as different crops present distinct water requirements and rooting characteristics, which influence the water uptake simulated by the model.

Irrigation water demands (HADGEM-ES-ETA)

The HADGEM2-ES model for the RCP 4.5 scenario (Figure 4) has projections for I_d for the future that comprises the period F1 with equal average values for some cities when compared to the baseline period. Others have larger I_d ranges, highlighting the municipalities of Pelotas and Canguçu (900 to 1100 mm).

Figure 4
I_d averages for the HADGEM2-ES climate model for the Southern and Sttepe regions of Rio Grande do Sul for: a) 2006 to 2040 for RCP 4.5; b) 2006 to 2040 for RCP 8.5; c) 2041 to 2070 for RCP 4.5; d) 2041 to 2070 for RCP 8.5; e) 2071 to 2099 for RCP 4.5; f) 2071 to 2099 for RCP 8.5.

In the analysis of the transition from F1 to F2 period, there are no large range increases or decreases as observed in the CANESM2 model. The municipalities of Canguçu and Pelotas continue to show the highest I_d, values, and the others largely remain within the projected range for the F1 period. Herval and Cerrito go from 100 to 300 mm to 300 to 600 mm and Caçapava do Sul is the only municipality that shows a decrease in the I_d, going from 600 to 900 mm to 300 to 600 mm. In the analysis of the F2 for the last period analyzed (F3), it is observed that the municipalities present values in the averages of I_d indicated in the F2 period.

For the RCP 8.5 scenario (Figure 4), the F1 period indicates an increase in the irrigation water demand for the municipalities of Caçapava do Sul, Dom Pedrito, Pinheiro Machado, Piratini, Pedras Altas, and São José do Norte. Pelotas and Canguçu also show an increase, being the municipalities with the highest irrigation water demand (900 to 1100 mm).

For the F2 period, the model estimates the same water demand, except in the municipality of Caçapava do Sul, where the I_d value is lower than in the F1 period. In F3, the municipalities have the same projected values in the F2 period, with Jaguarão, Cerrito, and Morro Redondo having higher values. Pelotas and Canguçu appear as the municipalities with the highest average values for irrigation water demand in this projection in almost the entire future period, in the two projected scenarios.

Irrigation water demands (MIROC5-ETA)

For the MIROC5 model, the RCP 4.5 scenario (Figure 5) showed, in the comparison between the baseline period and the first analyzed future period (F1), a decrease in irrigation water demand for only two municipalities (Bagé and Santana da Boa Vista). The increase in average I_d is also present for the cities of Amaral Ferrador, Arroio Grande, and Jaguarão. In Rio Grande, where the average I_d was in the lowest value range, there is an increase in I_d between 100 and 300 mm and Pelotas has the highest average I_d, value, which increased from 600 to 900 mm to 900 to 1100 mm. The other municipalities continue with the same average values of the baseline period.

Figure 5
I_d averages for the MIROC5 climate model for the Southern and Sttepe regions of Rio Grande do Sul for: a) 2006 to 2040 for RCP 4.5; b) 2006 to 2040 for RCP 8.5; c) 2041 to 2070 for RCP 4.5; d) 2041 to 2070 for RCP 8.5; e) 2071 to 2099 for RCP 4.5; f) 2071 to 2099 for RCP 8.5.

Analyzing the F1 and F2 periods, there is an increase in I_d for Dom Pedrito, which goes from 600 to 900 mm to 900 to 1100 mm, and Santana da Boa Vista, which from 300 to 600 mm presents an I_d of 600 to 900 mm. Amaral Ferrador is the only municipality projected to decrease the I_d, going from the range of 600 to 900 mm to 300 to 600 mm. The others maintain the average values observed for the F1 period. In the comparison of F2 with the last F3 period (2071 to 2099), there is a decrease for the municipalities of Dom Pedrito, which from the range of 900 to 1100 mm goes to 600 to 900 mm; and São Lourenço do Sul and Santana da Boa Vista, which from 600 to 900 mm go to the 300 to 600 mm range. The other municipalities remain at the same average values.

For the RCP 8.5 scenario, the MIROC5 model (Figure 5) projects that in the comparison between the baseline period with the future F1, there is an increase in I_d ranging from 300 to 600 mm to 600 to 900 mm in the municipalities of Arroio Grande and Amaral Ferrador; from 100 to 300 mm to 300 to 600 mm in Jaguarão; and in Pelotas, which goes from the average value of 600 to 900 mm to 900 to 1100 mm. Santana da Boa Vista and Herval decrease from the average I_d between 600 to 900 mm to 300 to 600 mm and the other municipalities remain with the same values. Rio Grande had the lowest average I_d value in the baseline period and in this future period, as well as Canguçu, which was the municipality that maintained the highest average value (1100 to 1500 mm).

In the analysis of future F1 and F2, there is an increase in I_d in Rio Grande, which goes from less than 100 mm to an average between 100 and 300 mm, and a decrease in the municipality of Cerrito, which from 600 to 900 mm of water demand for average irrigation goes to average between 300 to 600 mm of average I_d. Amaral Ferrador presents a decrease in I_d, since it passes from the range between 600 to 900 mm to 300 to 600 mm. The other municipalities show a tendency to remain in the average range of I_d observed for the F1 period.

Comparing the F2 period with the third analyzed period, F3, the MIROC5 climate model projects that there is an increase in I_d for two cities, namely Amaral Ferrador and São José do Norte. Only in the municipality of Cerrito there is a decrease in the values of irrigation water demand, and the others remain with the same averages. Canguçu appears as the city with the highest average irrigation water demand (1100 to 1500 mm) that will remain in the three futures, as in the RCP 4.5 scenario. Caliman (2024) used the HADGEM2-ES-ETA and MIROC5-ETA models to evaluate irrigation water requirements in the Araguaia Valley, western Bahia, the Central Plateau of Goiás, and in Santa Maria (RS), and found that supplementary irrigation will be required in all three future periods under both analyzed scenarios.

The results observed for the projections indicate that there is a difference in the values of average I_d in the municipalities that make up the study area. This may be due to the fact that the different types of crops exist in the municipalities, in addition to the fact that some have large areas of various crops, thus demanding more water. Still, most of the models show a certain similarity of the same municipality in their projections, such as, for example, Canguçu, which in all models is the municipality with the highest irrigation water demand, followed by Pelotas. Santa Vitória do Palmar appears in most projections as a municipality that maintains constant average of irrigation water demand, being among those that will demand less water (100 to 300 mm), either in the most optimistic scenario or in the most pessimistic scenario.

In a general analysis, the models do not show averages I_d with significantly higher values when compared to the baseline period, and higher values may occur, but they show the same trend between models identified for a given municipality, when compared to the others, in the baseline period analysis.

Therefore, the projections of the CANESM2-ETA, HADGEM2-ES-ETA and MIROC5-ETA models for the Southern and Sttepe regions of RS agree with Cardoso et al. (2022), who observed an upward trend in precipitation regimes for the future F1, F2, and F3 for the vast majority of municipalities when these were compared with the baseline period. However, Cardoso et al. (2022) also observed an increase in temperature for these regions, which corroborates the increase in I_d identified in this period, considering that these parameters are closely linked with this variable. It should be noted that such increases in average temperature were also observed by Bravo et al. (2011), when analyzing projections from 20 GCM’s considering two future periods (2030 and 2070) for the municipalities of Santa Vitória do Palmar and Rio Grande.

Therefore, it is possible to infer that the increase in precipitation volumes indicated for the regions, in almost its entirety, will not be enough to counterbalance the estimated temperature increase, as shown by Streck & Alberto (2006) who, when simulating the impact of climate change on the water available in the soil for wheat, soybean, and corn crops in the municipality of Santa Maria, in RS, observed that the climate will affect the water available in the soil for crops, and this fact will be due to the increase of temperature that can reach up to 6°C, and consequently will reduce the availability of water for summer crops.

Melo (2015) analyzed future irrigation water demands using five GCMs and five projections derived from the ETA regional climate model for the Northwestern region of Rio Grande do Sul, obtaining values lower than 10 cm. When compared to the results obtained in the present study for the Southern and Steppe regions, these values are considerably lower. Similarly, a study conducted by Kehl (2022), which evaluated the impact of climate change on irrigation water demand across four regions of Brazil, including the Northwestern region of Rio Grande do Sul, found no significant changes in irrigation water demand for this region. Studies such as Silva & Campos (2011) and Britto et al. (2008) indicate that, in addition to spatial variability across the state, rainfall regimes in Rio Grande do Sul tend to present higher volumes in the northern half compared to the southern half.

Therefore, there is agreement with the results of Melo (2015), since the Northwestern region has higher volumes of precipitation and, consequently, lower I_d, and the opposite was observed for the Southern and Sttepe regions, where the regimes are considered lower and the I_d higher.

The types of crops in each municipality also interfere with water demand, as crops have different water demands that must be met to ensure their development. Also noteworthy is the type of soil, which is another factor that interferes with water demands, since it is responsible for storing it and making it available for crops.

Anomalies

Figure 6 shows the anomalies for the F1 period of the RCP 4.5 scenario of the three climate models. For the CANESM2 model, it is possible to observe that there are negative and positive anomalies for the municipalities in the study area. The observed negative anomalies indicate that the water demand of the baseline period was greater than that of the future period, which may be an indication that the precipitation regimes of the first future period supplied most of the demand of the crops, corroborating with the precipitation volumes observed by Cardoso et al. (2022). For most of the municipalities in the Southern region for the same period, the same did not happen, that is, the demands of this future period were greater than those observed in the baseline period.

Figure 6
I_d anomalies for the F1 future period being: a) 2006 to 2041 for RCP 4.5 for CANESM2-ETA model; b) 2006 to 2041 for RCP 8.5 for CANESM2-ETA model; c) 2006 to 2041 for RCP 4.5 for HADGEM2ES-ETA model; d) 2006 to 2041 for RCP 8.5 for HADGEM2ES-ETA model; e) 2006 to 2041 for RCP 4.5 for MIRO-C5-ETA model; f) 2006 to 2041 for RCP 8.5 for MIRO-C5-ETA model.

The HADGEM2-ES model for RCP 4.5 indicates a decrease in water demand in the future period only for the municipality of Morro Redondo; the others indicated a positive anomaly. For MIROC5, RCP 4.5, there are positive and negative anomalies that occur in both regions of the study area. For the Sttepe region, only the municipality of Dom Pedrito indicates a positive anomaly. There is no similarity between the three models, and CANESM2 and MIROC5 present more similar results, mainly in terms of negative anomalies.

Figure 6 presents the anomalies for the same future period (F1), but for the RCP 8.5 scenario. The CANESM2 and MIROC5 models present a greater number of negative anomalies, that is, the water demand was higher in the baseline period than in the F1 future. For the CANESM2 climate model, only five municipalities have positive anomalies, namely: Jaguarão, Herval, Morro Redondo, Pelotas, and Turuçu. In the MIROC5 model, all municipalities have negative anomalies. The HADGEM2-ES model, on the other hand, indicates positive anomalies in almost its entirety, except for Arroio Grande.

For the F2 future period, scenario RCP 4.5 (Figure 7), there is great similarity in the results, where for CANESM2 and MIROC5, almost all municipalities present positive anomalies, except for four municipalities in the first model and seven in the second. In the projection of the HADGEM2-ES model, all municipalities indicate positive anomalies, that is, the irrigation water demands in the future period will be greater when compared to the past.

Figure 7
I_d anomalies for the F2 future period being: a) 2006 to 2041 for RCP 4.5 for CANESM2-ETA model; b) 2006 to 2041 for RCP 8.5 for CANESM2-ETA model; c) 2006 to 2041 for RCP 4.5 for HADGEM2ES-ETA model; d) 2006 to 2041 for RCP 8.5 for HADGEM2ES-ETA model; e) 2006 to 2041 for RCP 4.5 for MIRO-C5-ETA model; f) 2006 to 2041 for RCP 8.5 for MIRO-C5-ETA model.

In the RCP 8.5 scenario for the same F2 period, as shown in figure 7, most of the models present positive and negative anomalies. The municipalities of Dom Pedrito, Bagé, Hulha Negra, Lavras, Amaral Ferrador, and Amaral Ferrador have negative anomaly trends for both the CANESM2 and MIROC5 models. The municipality of São Lourenço do Sul also has a negative anomaly trend in the CANESM2 model. The others indicate positive anomalies. As for the MIROC5 model, in addition to those mentioned above, they present negative anomalies: Canguçu, Pedras Altas, and Santana da Boa Vista. In the HADGEM2-ES model, only Arroio has a negative anomaly. The other cities indicate that the water demand for the future period is greater than that for the baseline period, that is, they project positive anomalies. A study conducted by De Villa et al. (2024) found that the Campanha region, particularly the municipality of Bagé, presented the highest evapotranspiration deficit during the period from 2010/11 to 2020/21.

Figure 8 shows the results of irrigation water demand anomalies for the RCP 4.5 scenario for the three climate models in the F3 future. The CANESM2 model presents the same positive anomalies observed in F2, except for some municipalities in the sttepe region and the municipalities of Arroio do Padre and Amaral Ferrador. In the HADGEM2-ES model, all municipalities indicate positive anomalies. For the MIROC5 model, the greatest tendency is for negative anomalies, and six municipalities have positive anomalies, namely: Dom Pedrito, Herval, Morro Redondo, Pelotas, Rio Grande, and Turuçu. Therefore, it is observed that for the F3 period of this scenario, there is a great difference in the results of the climate models, which results in uncertainties.

Figure 8
I_d anomalies for the F3 future period being: a) 2006 to 2041 for RCP 4.5 for CANESM2-ETA model; b) 2006 to 2041 for RCP 8.5 for CANESM2-ETA model; c) 2006 to 2041 for RCP 4.5 for HADGEM2ES-ETA model; d) 2006 to 2041 for RCP 8.5 for HADGEM2ES-ETA model; e) 2006 to 2041 for RCP 4.5 for MIRO-C5-ETA model; f) 2006 to 2041 for RCP 8.5 for MIRO-C5-ETA model.

For the RCP 8.5 projection, F3 (Figure 8) indicates positive anomalies for the CANESM2 and HADGEM2-ES models, and in the latter, only the municipality of Arroio Grande resulted in a negative anomaly. The MIROC5 model differs from the others, with negative anomalies in 12 municipalities, which encompass the entire Sttepe region and some cities in the southern region.

The results of anomalies of irrigation water demands present great uncertainties in results and more precise conclusions; however, it is possible to observe some trends that predominate in the models and scenarios. In the RCP 4.5 projection for the F1 future, it can be noted that the HADGEM2-ES-ETA model is the only one that indicates that almost all of the southern and Sttepe regions present positive anomalies. On the other hand, the MIROC5-ETA model indicates a negative anomaly for most municipalities. In the RCP 8.5 projection for the same future period, the MIROC5-ETA model presents a negative anomaly in all municipalities, and the CANESM2-ETA model indicates a negative anomaly for the vast majority of municipalities. The opposite is observed in the HADGEM2-ES-ETA model, which indicates positive anomalies for most municipalities.

For the F2 future period, the three models indicate that all or most of the municipalities will present positive anomalies, both in the RCP 4.5 projection and in the RCP 8.5 projection. In the last future period (F3), the CANESM2-ETA model indicates a positive anomaly for all municipalities in the RCP 8.5 projection and for the vast majority in the RCP 4.5 projection. The HADGEM2-ES-ETA model presents a positive anomaly for all municipalities in the RCP 4.5 projection and for almost all municipalities in the RCP 8.5 projection. The MIROC5-ETA model showed opposite results for the two projections, with the model indicating a negative anomaly for most municipalities.

Therefore, it is observed that most of the models indicate a decrease in irrigation water demand for future periods, when this is compared with the past, and others project an increase in demand. This may be due to the fact that the other variables used in the modeling present values that are not favorable for the development of crops, making them require greater volumes of water.

In a general analysis, the observed increase in water demand may have a direct influence on the observed temperature increase for the three climate models, according to Cardoso et al. (2022), considering that this can cause greater plant transpiration, in addition to soil evaporation, that is, the soil will not have enough available storage to supply the plant and thus meet its demands, and then it will be necessary to use the irrigation technique to provide water for crops.

In addition, other traditional agricultural practices can be adopted to increase water availability for plants, such as direct planting, which can reduce erosion and bring benefits not only to the soil but also to crop development. In this technique, the soil is always covered with growing plants or plant residues, and there is no plowing or harrowing. With this, the cover protects the soil from erosion, decreases the temperature and consequently the evaporation of water from the soil, thus increasing the water available to the plants.

In recent decades, news points to the occurrence of water rationing in the Southern and Sttepe regions, mainly in the municipality of Bagé, which often declares an emergency and rations water resources since its reservoirs have levels below expectations. Consequently, there are many crop losses during these periods. According to Brondani et al. (2013), who applied surveys to the urban population of the municipality of Bagé to verify the community’s perception of the drought and its impacts, it was observed that a large part of the urban population considers that the phenomenon of drought occurred more frequently from the 1980s to the year of the research. The same type of phenomenon is commonly observed in the municipality of Pelotas, as in the 2019/2020 drought, which caused major impacts on the population (Fernandes et al. 2021).

It is observed that the precipitation regimes in RS have changed in recent decades, when these data are compared with the historical period, where large rainfalls were distributed throughout the year and are currently concentrated in a short space of time, being essential to carry out strategic planning and management of water resources (Pessoa 2015).

In a general context, aiming at the management of water resources, the increase in average temperature also culminates in greater evaporation of water present in surface water resources. It is known that it returns to the Hydrological Cycle, but not necessarily in the same place. With this, it is essential to manage the water resources in the region, aiming, as indicated by Law No. 9.433/1997, to supply the priority uses that are human consumption and animal watering, and to manage always seeking to meet the multiple uses.

CONCLUSIONS

According to the results obtained, it is possible to observe the existence of uncertainties in the modeling of water demand for agricultural crops; however, it is possible to observe some behavioral trends in the projections (RCP 4.5 and RCP 8.5) and in future periods. In general, there is a predominance of I_d ranging from 100 to 600 mm in both projections for all future periods, and the Sttepe region tends to have a lower I_d when compared to the Southern region.

Relating the future projections with the baseline period, there is a trend towards a greater number of municipalities with negative I_d anomalies in the CANESM2-ETA and MIROC5-ETA models in the F1, F2, and F3 futures. The HADGEM2-ES-ETA model presents the trend of positive I_d anomalies in the three future periods.

The modeling for irrigation water demand cannot be considered completely accurate either, since there is a high turnover in crops, in addition to the fact that the future period (F3) is very distant from the current one, making it impossible to infer which crops, techniques used for their planting and development, and the planted areas that will actually exist in the future. Nevertheless, the results obtained in this study provide relevant information for agricultural planning and water resource management in the study region. The identification of possible trends in irrigation water demand may support decision-making related to irrigation management and strategies for adapting agricultural production systems to future climate conditions.

• Data availability
  The data supporting the findings of this study are publicly available from the Agência Nacional de Águas e Saneamento Básico (ANA) and the PROJETA platform. The processed data generated during the study are available from the corresponding author upon reasonable request.

References

• ALCÂNTARA LRP, SILVA MER, NETO SMS, LAFAYETTE FB, COUTINHO AP, MONTENEGRO SMGL & ANTONINO ACD. 2020. Mudanças climáticas e tendências do regime pluviométrico do Recife. Res Soc Develop, 3(9): 1-21.
• ALLEN RG, PEREIRA, LS RAES D & SMITH M. 1998. Crop evapotranspiration: Guidelines for computing crop water requirements. Rome: FAO, FAO – Irrigation and Drainage Paper 56, 300 p.
• ASSAD ED, VICTORIA DC, CUADRA SV, PUGLIERO VS & ZANETTI MR. 2020. Efeito das mudanças climáticas na agricultura do Cerrado. In: Bolfe EL, Sano EE & Campos SK (Eds), Dinâmica agrícola no cerrado: análises e projeções. Brasília, DF: Embrapa, 1: 213-228.
• BEHIZADI N, FARAMARZMANESH S, AMERICANO M & GARMDAREH SHE. 2026. Calibration and evaluation of SWAP model under PRD irrigation: a case study, Pakdasht, Iran. Water Pract Technol 0: 1-20.
• BLACK C. 2020. A economia gaúcha no primeiro semestre de 2020: Desaceleração cíclica e dois choques. Rev Estud de Planej, 16: 2-25.
• BRAVO JM, MARQUES DM, TASSI R & CARDOSO A. 2011. Avaliação de projeções de anomalias de temperatura e precipitação em cenários climáticos futuros na região do sistema hidrológico do Taim, RS. In: Anais do XIX SBRH. Associação Brasileira de Recursos Hídricos.
• BRITTO FP, BARLETTA R & MENDONÇA M. 2008. Regionalização sazonal e mensal da precipitação pluvial máxima no estado do Rio Grande do Sul. Rev Bras Climatol Curitiba 3: 83-99.
• BRONDANI ARP, WOLLMANN CA & RIBEIRO AA. 2013. A percepção climática da ocorrência de estiagens e os problemas de abastecimento de água na área urbana do município de Bagé-RS. Rev Dep Geogr – USP 26: 214-232
• CALIMAN MA. 2024. Projeção das mudanças climáticas na demanda de água para irrigação, 77 p. Tese de Doutorado em Engenharia Agrícola, Universidade Federal de Viçosa, Viçosa.
• CARDOSO IP, SIQUEIRA TM, TIMM LC, RODRIGUES AA & NUNES AB. 2022. Analysis of average annual temperatures and rainfall in Southern region of the state of Rio Grande do Sul, Brazil. Rev Bras Ciênc Ambient 1(57): 58-71.
• CHOU SC. 1996. Regional Eta Model. In: Climanálise. Edição Comemorativa de 10 anos. Instituto Nacional de Pesquisas Espaciais. Cachoeira Paulista, SP.
• COLLISCHONN B, PAIVA RCD, COLLISCHONN W, MEIRELLES FSC, SCHETTINI EBC & FAN FM. 2011. Modelagem Hidrológica de Uma Bacia com Uso Intensivo de Água: Caso do Rio Quaraí-RS. Rev Bras Rec Hidr 4(16): 119-133.
• CONAB. 2017. Calendário de Plantio e Colheita de Grãos no Brasil, 73 p.
• CORDEIRO APA, ALVES RCM & ROCHA MB. 2019. Caracterização agroclimática de Bagé, RS. Agrometeoros 2(27): 293-309.
• COSTA RA & QUEIROZ AT. 2021. Definição da duração da estação seca e estação chuvosa e sua influência na agricultura no município de Ituiutaba – MG. Rev Bras Climat 28: 391-405.
• DE VILLA B, PETRY MT, MARTINS JD, MELO GL, TOKURA LK, MOURA MB, TONETTO F & GONÇALVES AF. 2024. Balanço hídrico do solo e a necessidade de irrigação do milho em diferentes regiões do Rio Grande do Sul. Rev Bras Climat 34: 406-431.
• EMPRESA BRASILEIRA DE PESQUISA AGROPECUÁRIA (EMBRAPA) & SERVIÇO BRASILEIRO DE APOIO ÀS MICRO E PEQUENAS EMPRESAS (SEBRAE). 2010. Catálogo Brasileiro de Hortaliças. Brasília, DF, 59 p.
• FEIX RD & JÚNIOR SL. 2019. Painel do Agronegócio do Rio Grande do Sul – 2019. Porto Alegre: SEPLAG, Departamento de Economia e Estatística, 54 p.
• FERNANDES VR, CUNHA APMA, PINEDA LAC, LEAL KRD, COSTA LCO, BROEDEL E, FRANÇA DA, ALVALÁ RCS, SELICHI ME & MARENGO J. 2021. Seca e os impactos na região Sul do Brasil. Revista Brasileira de Climatologia (Online) 28: 561-584.
• FAO - FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS. 2021. Crop Information. http://www.fao.org/land-water/databases-and-software/crop-information Accessed in 26 january 2023
  » http://www.fao.org/land-water/databases-and-software/crop-information
• FEE - FUNDAÇÃO DE ECONOMIA E ESTATÍSTICA. 2015. COREDES https://arquivofee.rs.gov.br/perfil-socioeconomico/coredes Accessed in 19 november 2024.
  » https://arquivofee.rs.gov.br/perfil-socioeconomico/coredes
• HEISTERMANN M ET AL. 2026. Soil moisture monitoring with cosmogenic neutrons: an asset for the development and assessment of soil moisture products in the state of Brandenburg (Germany) 26: 465-486.
• HUANG X, SHANG S, XIAOMIN M, LI J, BO L & ZHAO Y. 2026. Mapping and assessing maize water-efficiency dynamics using data assimilation and the SWAP model in Northwest China. Field Crops Res 336.
• IPCC - INTERGOVERNAMENTAL PANEL ON CLIMATE CHANGE. 2013. The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, USA, 1539 p.
• IPCC - INTERGOVERNAMENTAL PANEL ON CLIMATE CHANGE. 2023. Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate. Geneva, Switzerland, 184 p.
• IPCC - INTERGOVERNAMENTAL PANEL ON CLIMATE CHANGE. 2025. IPCC authors meet in Paris to begin drafting the Seventh Assessment Report. https://www.ipcc.ch/2025/12/01/ar7lam1pr/#:~:text=IPCC%20authors%20meet%20in%20Paris,the%20Seventh%20Assessment%20Report%20%E2%80%94%20IPCC Accessed in 14 march 2026.
  » https://www.ipcc.ch/2025/12/01/ar7lam1pr/#:~:text=IPCC%20authors%20meet%20in%20Paris,the%20Seventh%20Assessment%20Report%20%E2%80%94%20IPCC
• KEHL LR. 2022. Análise comparativa da demanda hídrica para irrigação na agricultura considerando as mudanças climáticas em quatro regiões do Brasil, 80 p. Trabalho de Conclusão de Curso (Graduação Engenharia Ambiental) – Universidade Federal do Rio Grande do Sul, Porto Alegre. (Unpublished).
• KROES JG, VAN DAM JC, GROENENDIJK P, HENDRIKS RFA & JACOBS CMJ. 2008. SWAP 3.2. Theory description and user manual. Wageningen, Alterra, Alterra Report 1649(02) – SWAP32 Theory description and user manual.doc., 262 p.
• KULMAN D, REIS JT, DE SOUZA AC, PIRES CAF & SAUSEN TM. 2014. Ocorrência de Estiagem no Rio Grande do Sul no Período de 1981 a 2011. Ciênc Nat (Online) Santa Maria 3(36): 441-449.
• LEUSIN JÚNIOR S, FEIX RD, PESSOA ML, RISCO G & BARBOSA AM. 2025. Painel do agronegócio do Rio Grande do Sul – 2025. Porto Alegre: DPGG.
• MALUF JRT, MATZENAUER R & MALUF DE. 2011. Zoneamento Agroclimático da Mandioca no Estado do Rio Grande do Sul – Uma alternativa para a produção de etanol. Porto Alegre: FEPAGRO. BOLETIM FEPAGRO, 22, 60p.
• MELO TM. 2015. Simulação Estocástica dos Impactos das Mudanças Climáticas Sobre as Demandas de Água para Irrigação na Região Noroeste do Rio Grande do Sul, 133 p. Tese (Recursos Hídricos e Saneamento Ambiental) – Universidade Federal do Rio Grande do Sul, Porto Alegre. (Unpublished).
• PESSOA ML. 2015. O Rio Grande do Sul corre o risco de enfrentar uma crise hídrica? Carta de Conjuntura FEE. 3, ano 24, p. 1.
• PIEPER MS, AREJANO LM, CARDOSO RC ROCHA LHS, GRACIOSE TVZF & TEDESCO GS. 2024. Desafios da agricultura familiar em Canguçu-RS: impactos da estiagem e necessidades estruturais. Rev Bras Eng Sustent, Pelotas, n. Esp. 13: 81-90.
• RASEIRA A ET AL. 1998. Instalação e Manejo do Pomar. In: Medeiros CA & Raseira MCBA (Eds), Cultura do Pessegueiro. Brasília: Embrapa, p. 130-160.
• SANTOS HG ET AL. 2011. O novo mapa dos solos do Brasil: legenda atualizada. Dados eletrônicos. Rio de Janeiro: Embrapa Solos, 67 p.
• SEPLAN - SECRETARIA DO PLANEJAMENTO E DESENVOLVIMENTO REGIONAL. 2015. Governo do Estado do Rio Grande do Sul. Perfis – COREDEs e Regiões Funcionais de Planejamento. Porto Alegre: Governo do Estado do Rio Grande do Sul, 82 p.
• SILVA MV & CAMPOS RJ. 2011. Anomalias decadais do regime hídrico do RS no período de 1977 a 2006. Ciênc Nat, Santa Maria, 1(33): 75-89.
• SILVA KK, SIQUEIRA TM, ADAM KN, CASTRO AS, CORRÊA LB & LEANDRO D. 2018. Future irrigation water requirements in the Ijuí River basin, RS. Rev Bras Eng Agríc Amb 1(22): 57-62.
• SOUZA PJOP, SANTOS CDM, SOUZA EB, OLIVEIRA EC & SANTOS JTS. 2018. Impactos das mudanças climáticas na cultura da soja no nordeste do estado do Pará. Rev Bras Agric Irrig 2(12): 2454-2467.
• STRECK NA & ALBERTO CM. 2006. Simulação do impacto da mudança climática sobre a água disponível do solo em agrossistemas de trigo, soja e milho em Santa Maria, RS. Ciênc Rural, Santa Maria, 2(36): 424-433.
• VAN GENUCHTEN MTA. 1980. Closed-form Equation for Predicting the Hydraulic Conductivity of Unsaturated Soils. Soil Sci Soc Am 44: 892-898.

Edited by

• Handling editor
  Mirco Solé

Data availability

The data supporting the findings of this study are publicly available from the Agência Nacional de Águas e Saneamento Básico (ANA) and the PROJETA platform. The processed data generated during the study are available from the corresponding author upon reasonable request.

Publication Dates

• Publication in this collection
  24 Aug 2026
• Date of issue
  2026

History

• Received
  11 Dec 2024
• Accepted
  17 May 2026