Introduction
⌅Rice consumption in Cuba has been estimated at approximately 70 kg per capita annually, meaning the country demands approximately 700,000 tons of rice (Gutiérrez & Lau, 2019GUTIÉRREZ, S.D.; LAU, S.B.: “Metodología para la formación del precio del arroz en Cuba”, Revista Cubana De Finanzas Y Precios, 3(1): 91-101, 2019.). Average paddy rice production for the years 2018-2022 was 297,281 tons, equivalent to approximately 134,000 tons of rice for consumption. For the same five-year period, the average annual rice import was 477,480 tons ONEI-Cuba (2022)ONEI-CUBA: Anuario Estadístico de Cuba 2022, La Habana, Cuba, 2022, ISBN: 978-959-7119-62-3., this means that the country has needed to import an average of 78% of its rice consumption. The above figures indicate the importance of this crop for the country and the need to increase its production in order to minimize such a high volume of imports. This dilemma highlights the potential future impact of climate change on rice production in Cuba, especially in terms of whether or not it will be able to meet its water demand.
The Third National Communication to the United Nations Framework Convention on Climate Change (CITMA-Cuba, 2020CITMA-CUBA: Tercera Comunicación Nacional a la Convención Marco de las Naciones Unidas sobre Cambio Climático, Ed. Ministerio de Ciencia, Tecnología y Medio Ambiente, República de Cuba, La Habana, Cuba, 2020, ISBN: 978-959-300-170-0.), predicts that among the main impacts of climate change on rice cultivation in the country are a reduction in the cultivated area, resulting in lower water availability, lower-quality harvests, and lower overall magnitude. Furthermore, according to Hervis et al. (2019)HERRERA-PUEBLA, J.; HERVIS-GRANDA, G.; GONZÁLEZ-ROBAINA, F.; DUARTE-DÍAZ, C.: “Estudio sobre el balance hídrico del arroz en Cuba”, Ingeniería Agrícola, 9(3), 2019, ISSN: 2227-8761., the most significant potential impacts on agricultural development focus on climatic variables such as temperature and precipitation, at different time scales and depending on the region. Future local changes will increase the average annual temperature by 1.0°C and the average annual minimum temperature by 2.0°C. The IPCC (Intergovernmental Panel on Climate Change), in its sixth report, defines shared socioeconomic pathways (SSPs), which replace the so-called Representative Concentration Pathways (RPCs), these shared socioeconomic pathways (SSPs) as modeled scenarios, which are used to explore future emissions, climate change, impacts, possible mitigation and adaptation strategies, in addition, the corresponding future projection of greenhouse gas emissions and land use change is linked under the argument of the reference SSP as a new and improved version of the RCPs (Environment and Climate Change Canada, 2023ENVIRONMENT AND CLIMATE CHANGE CANADA: CMIP6 and Shared Socioeconomic Pathways overview, ‘en línea“, ClimateScenarios.Canada, 2023, Disponible en: https://ClimateScenarios.Canada.ca/?Page=cmip6-OverviewNotes.).
In particular, SSP1-2.6 is a scenario starting in 2015 with low GHG emissions and decreasing CO2 emissions to net-zero emissions around or after 2050, followed by varying levels of net-negative CO2 emissions (Januta, 2021JANUTA, A.: “Que significan los cinco futuros del informe de la ONU sobre el clima”, es.euronews.com, 2021.). It is an optimistic scenario developed to simulate developments compatible with the 2°C target (global warming of 2°C is unlikely to exceed, estimated temperature: 1.5°C), on the way to a sustainable and green world (IPCC, 2021IPCC: Cambio climático 2021. Bases físicas. Resumen para responsables de políticas, Contribución del Grupo de Trabajo I al Sexto Informe de Evaluación del Grupo Intergubernamental de Expertos sobre el Cambio Climático, La Habana, Cuba, 40 p., 2021, ISBN: 978-92-9169-358-0.).
In studies on future water demand for rice crops in various countries (Gilanipour & Gholizadeh, 2016GILANIPOUR, J.; GHOLIZADEH, B.: “Prediction of rice water requirement using FAO-CROPWAT model in North Iran under future climate change”, Preprints, 2016.; Kyu & Truong, 2018KYU, S.L.; TRUONG, A.D.: “Predicting future water demand for Long Xuyen Quadrangle under the impact of climate variability”, En: Acta Geophysica, 2018, DOI: https://doi.org/10.1007/s11600-018-0176-4.; Hastarai et al., 2022HASTARAI, A.A.; WIBOWO, A.; TAMBUNAN, M.P.; PUTRO, D.A.: “Projection of The Effect of Climate Change on Crop Water Requirements for Rice Plants in Majalengka Regency”, En: IOP Conference Series: Earth and Environmental Science, Ed. IOP Publishing, vol. 1111, p. 012021, 2022, ISBN: 1755-1315.; Agrawal et al., 2023AGRAWAL, A.; SRIVASTAVA, P.K.; TRIPATHI, V.K.; MAURYA, S.; SHARMA, R.; DJ, S.: “Future projections of crop water and irrigation water requirements using a bias-corrected regional climate model coupled with CROPWAT”, Journal of Water and Climate Change, 14(4): 1147-1161, 2023, ISSN: 2040-2244. among others), the CropWat program developed by Smith (1992)SMITH, M.: CROPWAT: A computer program for irrigation planning and management, Ed. Food & Agriculture Org., 1992, ISBN: 92-5-103106-1., has been widely used in its most modern version CropWat 8.0 (FAO, 2008FAO: CropWat 8.0 for windows 2000-2008, Food Agricultural Organization (FAO), Rome, Italy, 2008.). The CropWat program calculates potential evapotranspiration using the Penman-Monteith equation, for which it requires data on maximum and minimum temperature, relative humidity, wind speed, and sunshine; from these data and using the crop coefficient, it can calculate crop evapotranspiration for the selected site. In addition, it requires rainfall and soil data, the latter data, in the case of rice, demands a different treatment than the rest of the crops, taking into account that rice is grown in most of the world using flood irrigation. All of this data is processed in different subroutines, which the program then combines to calculate irrigation demand and schedule irrigation (FAO, 2009FAO: Cropwat 8.0 for windows user guide, Food Agricultural Organization (FAO), Rome, Italy, 2009.).
With this in mind, this study aimed to study the effect of rainfall variation using the SSP1 2.6 climate scenario and three regional climate models (Hadgem3-gc31-ll-SSP1 2.6, Mpi-esm1-2-hr-SSP126, and Mri-esm2-0-SSP126), with climate variables up to the year 2050. CropWat 8.0 FAO (2008)FAO: CropWat 8.0 for windows 2000-2008, Food Agricultural Organization (FAO), Rome, Italy, 2008.. was used to calculate irrigation demand.
Materials and Methods
⌅Table 1 shows the location of the study sites. The geographic coordinates shown correspond to the location of the meteorological station used.
| Site | Province | Coordinates | Altitude (m) | |
|---|---|---|---|---|
| North | West | |||
| Los Palacios | Pinar del Río | 22°33’47” | 83°18’26” | 47 |
| Corojal | Artemisa | 23º 29’ 4,1” | 83º 24’ 6,26” | 38 |
| Sur del l Jibaro | Sancti Spiritus | 21°46´18,03" | 79°16´6,22" | 41 |
| Vertientes | Camagüey | 21° 52´68 | 78° 22´58 | 32 |
| Jucarito | Granma | 20°40´6,76" | 76°33´10,12" | 37 |
| Yara | 20°13´38,59" | 76°57´3,40" | 41 | |
Rainfall
⌅In order to organize the data required by the different subroutines of the CropWat program, an analysis was conducted of rainfall in the country's main rice-growing regions (Table 1) for the period 2022-2050. This analysis was performed for three models and the same SSP1 2.6 scenario:
For the rainfall analysis, according to Aleman et al., 2020ALEMÁN, L.A.; MENESES, J.; MARTINEZ, J.; POLÓN, R.; HERRERA, J.; LEÓN, O.: Manejo del agua., Ed. Ministerio de la agricultura, Instituto de Granos, Instructivo técnico del arroz ed., vol. I, La Habana, Cuba, 2020, ISBN: 978-959-285-065-1.), three rice growing cycles were considered: Cold (December-May, 150-day cycle), Pre-spring (May-September, 140-day cycle), and Spring (July-October, 110-day cycle). The rainfall data were sorted by month within each year, and the rainfall for each crop cycle was calculated for the country's main rice-growing regions over the 29 years studied. This data is shown in Table 1.
The rainfall data, sorted as described above, were analyzed for subsequent use in calculating crop water demand, following the procedure outlined by FAO (2016)FAO: Cropwat 8.0 for windows user guide, Food Agricultural Organization (FAO), Rome, 2016.. for use in the CropWat 8.0 program.
Soils
⌅Martínez et al. (2017)MARTÍNEZ, J.; HERRERA, J.; MENESES, J.: Los suelos para el cultivo del arroz y su relación con el riego. En: Manejo integrado del riego en el cultivo del arroz, Inst. Instituto de Investigaciones del Arroz- Instituto de Ingeniería Agrícola, Informe de proyecto, La Habana, Cuba, 2017., distributed the predominant soil types in the different rice-growing areas in each province, which is shown in Table 2.
Ascanio et al. (1980)ASCANIO, O.; BOUZA, H.; HERNANDEZ, A.; AGAFONOV, O.: “Principales propiedades fisicas e hidrofisicas de los suelos arroceros de Cuba.”, Ciencia y Tecnica en la Agricultura. Serie Riego y Drenaje, 3, 1980. based on the Second Genetic Classification of Cuban Soils, grouped Cuban rice soils into 7 types, as shown in Table 2. Martínez et al. (2017)MARTÍNEZ, J.; HERRERA, J.; MENESES, J.: Los suelos para el cultivo del arroz y su relación con el riego. En: Manejo integrado del riego en el cultivo del arroz, Inst. Instituto de Investigaciones del Arroz- Instituto de Ingeniería Agrícola, Informe de proyecto, La Habana, Cuba, 2017., distributed these soil types across different rice-growing areas in each province.
| Types | soils | Rice areas | Soils infiltration groups according to Samaké (1998)SAMAKÉ, M.: Familias de curvas de infiltración para los suelos arroceros cubanos y sus aplicaciones en el diseño de sistemas de riego para estos suelos, Instituto Superior Agropecuario de la Habana (ISCAH)-Instituto de Investigaciones de Riego y Drenaje (IIRD), Tesis en Opción al Grado de MSc., Provincia Habana, Cuba, 1998. |
|---|---|---|---|
| I | Dark gleyed and humic plastics, typical gley (heavy clay texture) | Matanzas, Sancti Spiritus Granma. | I |
| II | Dark gleyed and gleyzous plastics | Artemisa y Mayabeque | II |
| III | GleyFerraliticos | Pinar del Río y Camagüey | III |
| IV | Dark gleyzoid plastics (granular structure up to 40 cm) | Matanzas, Sancti Spiritus Granma. | I |
| V | Typical humicgley (loamy texture) | Granma | I |
| VI | Yellowish quartzite gley, laterized | Pinar del Río, Camaguey | III |
| VII | Typical yellowish quartzite | Pinar del Rio, Artemisa | III |
Based on the data of rice´s soils physical properties existing in the IAgric database, and the works of Simeón (1979)SIMEÓN, F.: “Características de las propiedades hidrofísicas de los principales suelos agrícolas de Cuba”, Voluntad hidráulica, 16(49-50): 16-23, 1979., Ascanio et al. (1980)ASCANIO, O.; BOUZA, H.; HERNANDEZ, A.; AGAFONOV, O.: “Principales propiedades fisicas e hidrofisicas de los suelos arroceros de Cuba.”, Ciencia y Tecnica en la Agricultura. Serie Riego y Drenaje, 3, 1980. and Cid et al. (2012)CID, G.; LÓPEZ, T.; GONZÁLEZ, F.; HERRERA, J.; RUIZ, M.E.: “Características físicas que definen el comportamiento hidráulico de algunos suelos de Cuba”, Revista Ingeniería Agrícola, 2(2): 25-31, 2012, ISSN: 2306-1545., the file of properties of these soils was prepared according to each site, where the predominant soil types in each one were taken from the 1:25 000 soil map of each of the provinces where the study sites are located. According to the soil distribution shown in Table 2, and the three categories into which Samaké (1998)SAMAKÉ, M.: Familias de curvas de infiltración para los suelos arroceros cubanos y sus aplicaciones en el diseño de sistemas de riego para estos suelos, Instituto Superior Agropecuario de la Habana (ISCAH)-Instituto de Investigaciones de Riego y Drenaje (IIRD), Tesis en Opción al Grado de MSc., Provincia Habana, Cuba, 1998., grouped them according to its infiltration capacity, Table 3 was prepared, which shows the soil parameters used in the soil module of the CropWat program for each site considered in the study.
| Parameter | Group I | Group II | Group III |
|---|---|---|---|
| Matanzas, Sancti Spíritus, Granma | Artemisa, Mayabeque | Pinar del Río, Camagüey | |
| Total Available water (TAW) | 270 | 250 | 170 |
| Maximum Infiltration rate (mm/day) | 17 | 17 | 43 |
| Maximum rooting depth (cm) | 50 | 50 | 50 |
| Initial soil moisture depletion (% de ADT (%) | 100 | 100 | 100 |
| Initial available soil moisture (mm/m) | 0 | 0 | 0 |
| Drainable Porosity () % | 5 | 9 | 9 |
| Critical depletion for puddle cracking(fraction) | 0,6 | 0,6 | 0,6 |
| Maximum percolation rate after puddling(mm/day) | 2,6 | 2,6 | 2,6 |
| Water availability at planting (% of saturation) | 0 | 0 | 0 |
| Maximum water depth (mm) | 100 | 10 | 10 |
Crop coefficients
⌅Crop coefficients (KC) were calculated for the study sites based on the Kc proposed by Herrera-Puebla et al. (2020)HERRERA-PUEBLA, J.; MENESES-PERALTA, J.; DUARTE-DÍAZ, C.; GONZÁLEZ-ROBAINA, F.; HERVÍS-GRANDA, G.: “Determination of Crop Coefficients for Estimating Evapotranspiration in a Paddy Field in Cuba”, Revista Ciencias Técnicas Agropecuarias, 29(3): 05-20, 2020, ISSN: 1010-2760., for the western region of Cuba, using the methodology proposed by Allen et al. (2006)ALLEN, R.G.; PEREIRA, L.S.; RAES, D.; SMITH, M.: “Evapotranspiración del cultivo: guías para la determinación de los requerimientos de agua de los cultivos”, Roma: FAO, 298(0), 2006.. To calculate the initial Kc, in their methodology, Allen et al. (2006)ALLEN, R.G.; PEREIRA, L.S.; RAES, D.; SMITH, M.: “Evapotranspiración del cultivo: guías para la determinación de los requerimientos de agua de los cultivos”, Roma: FAO, 298(0), 2006., established that for rice grown in flooded fields with a water depth of 0.10-0.20 m, the ETc value during the initial stage will consist primarily of evaporation from the water surface. The Kc value included by the aforementioned authors is 1.05 for a sub-humid climate with light to moderate winds. However, for the conditions of western Cuba, Herrera-Puebla et al. (2020)HERRERA-PUEBLA, J.; MENESES-PERALTA, J.; DUARTE-DÍAZ, C.; GONZÁLEZ-ROBAINA, F.; HERVÍS-GRANDA, G.: “Determination of Crop Coefficients for Estimating Evapotranspiration in a Paddy Field in Cuba”, Revista Ciencias Técnicas Agropecuarias, 29(3): 05-20, 2020, ISSN: 1010-2760., found an initial Kc value of 0.8, which was used to extrapolate the mean and final coefficients for the sites where irrigation demand will be calculated. Table 4 shows the crop coefficients for each study site and their comparison with the Kc proposed by Allen et al. (2006)ALLEN, R.G.; PEREIRA, L.S.; RAES, D.; SMITH, M.: “Evapotranspiración del cultivo: guías para la determinación de los requerimientos de agua de los cultivos”, Roma: FAO, 298(0), 2006..
| Allen et al. (2006)ALLEN, R.G.; PEREIRA, L.S.; RAES, D.; SMITH, M.: “Evapotranspiración del cultivo: guías para la determinación de los requerimientos de agua de los cultivos”, Roma: FAO, 298(0), 2006. | Paso Real y Corojal | La Sierpe y Sur del Jibaro | Vertientes | Jucarito | Yara y Veguitas | Kc average | |
|---|---|---|---|---|---|---|---|
| Winter planting | |||||||
| Kcinitial | 1,1 | 0,80 | 0,80 | 0,80 | 0,80 | 0,80 | 0,80 |
| Kcmedium | 1,2 | 1,40 | 1,30 | 1,30 | 1,30 | 1,13 | 1,29 |
| Kcfinal | 1,05 | 1,3 | 1,18 | 1,19 | 1,23 | 1,23 | 1,23 |
| Spring planting | |||||||
| Kcinitial | 1,1 | 0,80 | 0,80 | 0,80 | 0,80 | 0,80 | 0,80 |
| Kcmedium | 1,2 | 1,33 | 1,40 | 1,30 | 1,31 | 1,21 | 1,31 |
| Kcfinal | 1,05 | 1,23 | 1,30 | 1,20 | 1,21 | 1,17 | 1,22 |
Results and Discussion
⌅Rainfall Analysis
⌅Figure 1 shows the behavior of annual and seasonal rainfall for each scenario at every studied sites. It can be observed that, regardless of the model, rainfall decreases from west to east, with average values for the three models of 1644, 1467.5, and 1239.1 mm yr-1 for the western, central, and eastern zones, respectively. The coefficient of variation (C.V.) for annual rainfall is also much lower in the western and central zones than in the eastern zones, indicating greater consistency in the average values for this region. In the western zone, scenario 2 (Mpi-esm1-2-hr-SSP126) showed the highest average annual rainfall values (1715,2 mm year-1), while in the eastern region, the highest values corresponded to Hadgem3-gc3SSP126 model (1287,2 mm). In the central zone, scenarios 1 and 2 had similar annual rainfall values (1456,5 mm year-1); in all the studied sites, the mean annual rainfall was 1456,5 mm year-1.
Also during the rainy season, rainfall in the west and center, as well as the CV, show higher values than the eastern region in all three models studied. However, for the dry season, only in model 1 (model Hadgem3-gc31-ll-SSP1 2.6) is there a clear superiority in rainfall values for the western region, although it also presents the highest CV values.
The behavior of the average monthly rainfall for all the sites studied and for each scenario is shown in Figure 2.
For all models, the lowest rainfall values are found in the months of November to April, corresponding to the dry season, and the highest in the months of May to October. A decrease in rainfall can be observed in the months of July and August, equivalent to 30%, 20%, and 20% for models 1, 2, and 3, respectively, relative to the total rainfall that would occur during the remaining months of the rainy season. This period, known as the cold season, according to Carrazana et al. (2013)CARRAZANA, J.A.; MARTÍNEZ, J.; COLOM, L.; FONTOVA, M.: “Uso Eficiente del agua en el Cultivo del Arroz”, En: VIII Seminario Internacional del Uso Integral del Agua. INRH, Cuba, Ed. Instituto Nacional de Recursos, Hidráulicos, La Habana, Cuba, 2013., is characterized by the longest growth cycle and expression of yield potential. These factors are influenced by the behavior of temperatures and solar radiation. It also corresponds to the highest water demand due to a longer growth cycle and the rainy season.
Figure 3 shows the rainfall values for each planting period, scenario, and site; As expected, rainfall values in period 1 (Figure 3a, sowing in December and cycle length of 150 days), which correspond to the dry season, show the lowest total rainfall, regardless of site or model. Period 2 (Figure 3b), where crop growth coincides with the rainy season, shows the highest accumulated precipitation, also independent of site and scenario. Period 3 (Figure 3c) shows an intermediate situation between the two previous periods.
Table 5 shows the calculated values for the probability of rainfall for the western and central regions and the eastern region. As Table 5 shows, there are differences, albeit slight, between the coefficients for each site and planting season. This is because Kc depends on both climatic conditions (potential evapotranspiration) and crop development, which determines crop evapotranspiration.
From the results obtained, in terms of rainfall parameters, crop coefficients and soil properties of the different sites where the future water demand for rice will be determined according to the SSP1 2.6 scenario and the 3 models studied, the great variation in rainfall between regions is observed, where it is shown that regardless of the model there is a decrease in rainfall values from west to east, with average values for the three scenarios 1644, 1467,5 and 1239,1 mm year-1 for the western, central and eastern areas respectively, which together with the variations in Kc and soil types between sites will impose variations in the irrigation demand of the crop.
| Sitio | Época | Meses | P (%) | Modelos | |||||
|---|---|---|---|---|---|---|---|---|---|
| Hadgem-gc31-II SSP-1.26 | Mpi.esn-1-2-hr SSP-1.26 | Mpi.esn-2-0-hr SSP-1.26 | |||||||
| Año | Lluvia total (mm) | Año | Lluvia total (mm) | Año | Lluvia total (mm) | ||||
| Región Occidental | |||||||||
| Los Palacios | 1 | dic-mayo | 25 | 2040 | 556,1 | 2029 | 599,9 | 2040 | 556,1 |
| 1 | 50 | 2033 | 462,5 | 2035 | 507,7 | 2033 | 462,5 | ||
| 1 | 75 | 2041 | 407,7 | 2037 | 453,8 | 2041 | 407,7 | ||
| 2 | mayo-sept. | 25 | 2029 | 985,3 | 2039 | 1113,8 | 2046 | 1057,8 | |
| 2 | 50 | 2036 | 878,5 | 2032 | 1002,2 | 2048 | 962,1 | ||
| 2 | 75 | 2047 | 816,1 | 2037 | 936,9 | 2033 | 906,2 | ||
| 3 | julio-nov | 25 | 2042 | 852,9 | 2029 | 956,3 | 2049 | 861,8 | |
| 3 | 50 | 2023 | 744,7 | 2039 | 855,8 | 2047 | 794,5 | ||
| 3 | 75 | 2045 | 681,5 | 2042 | 797,0 | 2041 | 755,2 | ||
| Corojal | 1 | dic-mayo | 25 | 2028 | 557,6 | 2029 | 603,8 | 2040 | 550,4 |
| 1 | 50 | 2037 | 482,1 | 2035 | 514,3 | 2042 | 473,5 | ||
| 1 | 75 | 2024 | 437,9 | 2037 | 462,0 | 2050 | 428,5 | ||
| 2 | mayo-sept. | 25 | 2039 | 1038,3 | 2048 | 1171,1 | 2032 | 1093,3 | |
| 2 | 50 | 2036 | 931,5 | 2041 | 1057,4 | 2050 | 1002,5 | ||
| 2 | 75 | 2033 | 869,1 | 2044 | 990,9 | 2042 | 949,4 | ||
| 3 | julio-nov | 25 | 2042 | 899,9 | 2027 | 998,3 | 2026 | 909,6 | |
| 3 | 50 | 2024 | 791,7 | 2045 | 897,8 | 2048 | 835,4 | ||
| 3 | 75 | 2047 | 728,5 | 2042 | 839,0 | 2045 | 792,0 | ||
| Región Central | |||||||||
| Sur del Jibaro | 1 | dic-mayo | 25 | 2043 | 503,5 | 2039 | 478,5 | 2029 | 499,9 |
| 1 | 50 | 2026 | 449,4 | 2030 | 437,6 | 2047 | 445,9 | ||
| 1 | 75 | 2038 | 417,7 | 2041 | 413,7 | 2036 | 414,4 | ||
| 2 | mayo-sept. | 25 | 2039 | 1076,4 | 2027 | 1189,5 | 2050 | 1073,5 | |
| 2 | 50 | 2026 | 979,3 | 2041 | 1016,9 | 2036 | 983,4 | ||
| 2 | 75 | 2032 | 922,6 | 2047 | 915,9 | 2023 | 930,7 | ||
| 3 | julio-nov | 25 | 2028 | 800,5 | 2034 | 854,6 | 2037 | 756,7 | |
| 3 | 50 | 2042 | 700,7 | 2027 | 691,7 | 2029 | 679,8 | ||
| 3 | 75 | 2048 | 642,3 | 2041 | 596,4 | 2032 | 634,8 | ||
| La Sierpe | 1 | dic-mayo | 25 | 2043 | 503,5 | 2039 | 477,2 | 2039 | 499,9 |
| 1 | 50 | 2026 | 449,4 | 2038 | 436,0 | 2047 | 445,9 | ||
| 1 | 75 | 2038 | 417,7 | 2041 | 411,9 | 2046 | 414,4 | ||
| 2 | mayo-sept. | 25 | 2039 | 1076,4 | 2027 | 1189,5 | 2050 | 1073,5 | |
| 2 | 50 | 2026 | 979,3 | 2041 | 1016,9 | 2036 | 983,4 | ||
| 2 | 75 | 2032 | 922,6 | 2047 | 915,9 | 2023 | 930,7 | ||
| 3 | julio-nov | 25 | 2028 | 800,5 | 2029 | 854,6 | 2037 | 756,7 | |
| 3 | 50 | 2042 | 700,7 | 2027 | 691,7 | 2027 | 679,8 | ||
| 3 | 75 | 2048 | 642,3 | 2041 | 596,4 | 2032 | 634,8 | ||
| Region Oriental | |||||||||
| Vertientes | 1 | dic-mayo | 25 | 2043 | 521,9 | 2046 | 506,1 | 2048 | 486,3 |
| 1 | 50 | 2038 | 461,7 | 2027 | 444,2 | 2040 | 443,4 | ||
| 1 | 75 | 2036 | 426,5 | 2042 | 407,9 | 2035 | 418,2 | ||
| 2 | mayo-sept. | 25 | 2033 | 937,1 | 2025 | 1054,9 | 2026 | 919,5 | |
| 2 | 50 | 2050 | 823,4 | 2032 | 845,6 | 2036 | 819,7 | ||
| 2 | 75 | 2036 | 756,9 | 2024 | 723,1 | 2047 | 761,3 | ||
| 3 | julio-nov | 25 | 2047 | 723,1 | 2027 | 811,6 | 2029 | 646,5 | |
| 3 | 50 | 2041 | 609,4 | 2032 | 604,3 | 2048 | 579,6 | ||
| 3 | 75 | 2038 | 542,9 | 2043 | 483,1 | 2030 | 540,5 | ||
| Jucarito | 1 | dic-mayo | 25 | 2032 | 455,5 | 2032 | 432,8 | 2034 | 440,8 |
| 1 | 50 | 2033 | 395,5 | 2031 | 364,2 | 2026 | 390,6 | ||
| 1 | 75 | 2036 | 360,4 | 2039 | 324,1 | 2035 | 361,2 | ||
| 2 | mayo-sept. | 25 | 2041 | 828,9 | 2029 | 919,5 | 2029 | 823,8 | |
| 2 | 50 | 2045 | 736,7 | 2034 | 730,9 | 2038 | 728,1 | ||
| 2 | 75 | 2036 | 682,8 | 2026 | 620,6 | 2040 | 672,2 | ||
| 3 | julio-nov | 25 | 2024 | 623,5 | 2023 | 733,7 | 2039 | 572,0 | |
| 3 | 50 | 2032 | 533,4 | 2032 | 555,6 | 2049 | 509,5 | ||
| 3 | 75 | 2045 | 480,7 | 2043 | 451,4 | 2033 | 473,0 | ||
| Yara | 1 | dic-mayo | 25 | 2025 | 444,1 | 2041 | 426,8 | 2033 | 416,8 |
| 1 | 50 | 2045 | 389,4 | 2026 | 364,9 | 2026 | 379,7 | ||
| 1 | 75 | 2026 | 357,3 | 2036 | 328,6 | 2035 | 358,0 | ||
| 2 | mayo-sept. | 25 | 2041 | 878,9 | 2023 | 882,5 | 2026 | 835,4 | |
| 2 | 50 | 2050 | 752,1 | 2033 | 754,3 | 2037 | 732,1 | ||
| 2 | 75 | 2036 | 677,9 | 2048 | 679,3 | 2038 | 671,7 | ||
| 3 | julio-nov | 25 | 2042 | 694,2 | 2035 | 679,3 | 2026 | 641,9 | |
| 3 | 50 | 2035 | 568,0 | 2031 | 572,5 | 2030 | 549,7 | ||
| 3 | 75 | 2026 | 494,2 | 2044 | 510,1 | 2025 | 495,8 | ||
The variations in rainfall probabilities between years, regions, and models demonstrate the appropriateness of choosing rainfall from the period in which the crop is planted rather than the climatic or hydrological year.
Since the planting period, corresponding to the so-called "cold period," is when the highest yields are obtained and where the greatest water demands also occur, it was decided, when calculating future crop water demands, to work only with this period, since any water demand planning for a rice irrigation system project would include the demand for this cycle.
Water Balance
⌅The water balance shown in Figure 4 was calculated from the ETo values calculated using the Penman-Monteith equation (Allen et al., 2006ALLEN, R.G.; PEREIRA, L.S.; RAES, D.; SMITH, M.: “Evapotranspiración del cultivo: guías para la determinación de los requerimientos de agua de los cultivos”, Roma: FAO, 298(0), 2006.) and the monthly rainfall for each model and site.
As can be seen in Figure 4, in all models under the same SSP1 2.6 scenario, there is a marked imbalance between rainfall and ETo in the months corresponding to the dry season. While in the western and central regions this deficit disappears starting in May, in the eastern region, Models 1 and 3 maintain this throughout the year, while in Model 2, it disappears from August to November.
Crop Evapotranspiration
⌅According to Bouman et al. (2012)BOUMAN, B.A.M.; HAEFELE, S.M.; HIZZI, G.; PENG, S.; HSIAO, T.C.: Respuesta del rendimiento de los cultivos al Agua, Ed. International Rice Research Institute, Estudio FAO Riego y Drenaje, FAO ed., vol. 56, Roma, Italia, 2012, ISBN: 971-22-0219-4., the ETc for a rice crop cycle varies between 400 and 700 mm in the tropics and 800 to 1100 mm in temperate regions, representing 56 to 53% of the total water delivered to the crop, according to Renault (2004)RENAULT, D.: Rice is Life. International Year of Rice, ‘en línea“, FAO, Rome, Italy, 2004, Disponible en: www.rice.org.. Halfway through the crop cycle, when the crop is fully covered, rice evapotranspires at a rate slightly higher than the reference evapotranspiration (ET0), with average daily ET rates of 4-5 mm day-1 in the tropical wet season and 6-7 mm day-1 in the tropical dry season, but which can reach 10-11 mm day-1 in arid regions (Bouman et al., 2012BOUMAN, B.A.M.; HAEFELE, S.M.; HIZZI, G.; PENG, S.; HSIAO, T.C.: Respuesta del rendimiento de los cultivos al Agua, Ed. International Rice Research Institute, Estudio FAO Riego y Drenaje, FAO ed., vol. 56, Roma, Italia, 2012, ISBN: 971-22-0219-4.),
For Cuban conditions, Herrera-Puebla et al. (2019)HERRERA-PUEBLA, J.; HERVIS-GRANDA, G.; GONZÁLEZ-ROBAINA, F.; DUARTE-DÍAZ, C.: “Estudio sobre el balance hídrico del arroz en Cuba”, Ingeniería Agrícola, 9(3), 2019, ISSN: 2227-8761., reviewed studies on this parameter in Cuba and showed values ranging from 657 to 1173 mm/crop cycle. These authors also compared the results of lysimeter studies and data estimated using the CropWat program for the Los Palacios region (Pinar del Río).
Figure 5 shows the average crop evapotranspiration (ETc), in each model, for the total of a 130-day cycle in rice grown in the so-called cold season.
The values in Figure 5, show that rice´s ETc for the study period is within the values obtained both globally and nationally. The Mpi-esm1-2-hr-SSP126 model (model 2) showed the lowest ETc values, 3,9% and 8,3% lower than models 1 and 3, respectively.
Comparing the study sites, Figure 6 shows that regardless of the model, the highest ETc values are always found in the central region, with average values for the three models exceeding the values obtained in the western and eastern regions by 18,9% and 6,4%, respectively.
Water balance in rice cultivation
⌅Rice irrigation water demands are closely related to the crop's unique water balance. The rice water balance can be briefly represented by the water balance formula (Ding et al., 2017DING, Y.; WANG, W.; SONG, R.; SHAO, Q.; JIAO, X.; XING, W.: “Modeling spatial and temporal variability of the impact of climate change on rice irrigation water requirements in the middle and lower reaches of the Yangtze River, China”, Agricultural water management, 193: 89-101, 2017, ISSN: 0378-3774.):
Where ΔH is the variation in water stored on the terrace, I is the water applied as irrigation, and R, E, T, P, and D represent rainfall, evaporation, transpiration, percolation, and surface drainage, respectively.
Figure 7 shows the components of the crop water balance according to the three climate models studied. Crop consumption losses plus percolation losses constitute the majority of the water that irrigation must replenish. As Figure 7 shows, rainfall accounts for a contribution of 15 to 17%, depending on the model considered.
Crop irrigation demand
⌅As noted above, the amount of water to be applied as irrigation in rice crops depends on the ETc and percolation losses, since for agro technical reasons, the crop is maintained almost constantly with a water depth of 5-10 cm above the soil. The magnitude of this depth, expressed as the net irrigation rate, depends on the length of the crop cycle, climatic demand, and the soil infiltration capacity. Table 6 shows the gross irrigation demand values by site, for a 130-day crop cycle and cold sowing, i.e., sowing in December and harvesting in early May.
| Sites | Climatic Models | |||||
|---|---|---|---|---|---|---|
| Hadgem3-gc31-ll-SSP126 | Mpi-esm1-2-hr-SSP126 | Mpi-esm2-0-ssp-126 | ||||
| Irrigation needs (mm/crop cycle) | Irrigation needs (mm/crop cycle) | Irrigation needs (mm/crop cycle) | ||||
| Net | Gross | Net | Gross | Net | Gross | |
| Los Palacios | 1026,8 | 1466,9 | 971,8 | 1388,3 | 1025,8 | 1465,4 |
| Corojal | 891,4 | 1273,4 | 982,4 | 1403,4 | 655,8 | 936,9 |
| La Sierpe | 840,9 | 1201,3 | 736,8 | 1052,6 | 951,1 | 1358,7 |
| Vertientes | 1061,2 | 1516,0 | 1049,7 | 1499,6 | 952,3 | 1360,4 |
| Yara | 962,0 | 1374,3 | 843,1 | 1204,4 | 849,1 | 1213,0 |
| Jucarito | 835,1 | 1193,0 | 747,4 | 1067,7 | 938,1 | 1279,1 |
| Average | 1337,5 | 1269,3 | 1279,1 | |||
To calculate the gross irrigation, need as shown in Table 7, an overall irrigation system efficiency coefficient of 0,7 was used, equivalent to the efficiency required by an well designed and maintaining system according to INRH Standard 287/2015 (INRH-Cuba, 2015INRH-CUBA: Resolución 287/2015, Anexo 2. ÍNDICES DE CONSUMO: Normas de Riego Netas Totales para los Cultivos Agrícolas, Inst. Instituto Nacional de Recursos Hidráulicos, Presidencia del INRH, La Habana, Cuba, 2015.).
The same Standard establishes net irrigation requirements for rice ranging from 1128 to 1186 mm by cycle, which reach gross values, taking into account a system efficiency of 70%, of 1611 to 1694 mm, well above the actual requirements of the crop. Therefore, for the purposes of the national water balance, a gross irrigation requirement of 1400 mm has been established.
Recent studies by Cisneros et al. (2023)CISNEROS, E.; HERRERA, J.; CUN, R.; GONZÁLEZ, F.; CHATERLAN, Y.; DOMÍNGUEZ, C.: Estimación de las normas totales netas de riego en tres polos arroceros de Cuba, Inst. Instituto de Investigaciones de Ingeniería Agrícola (IAgric), Informe técnico de Introducción de resultados de Investigación, La Habana, Cuba, 10 p., 2023. for the sites Los Palacios, Corojal, and La Sierpe, using a 13-year historical rainfall and ETc series from 2008 to 2020 and rainfall occurrence probabilities within the crop cycle (130 days, December-May), produced the values shown in Table 7.
| Sitio | Precipitation Probability (%) | Rainfall (mm) | ETc (mm) | Total Net Irrigation Requirement (mm) | Total Percolation Losses (mm) |
|---|---|---|---|---|---|
| Los Palacios Pinar del Rio | 75 | 256,3 | 493,4 | 1023,88 | 580,4 |
| EL Corojal Artemisa | 75 | 287,0 | 488,0 | 891,80 | 415,5 |
| La Sierpe Santi Espíritus | 75 | 150,9 | 540,1 | 1027,78 | 411,8 |
Comparing the ETc values shown in Table 7 with those in Figure 6, it can be seen that the ETc for the period studied in the three models increased on average by 32, 27, and 33% for Los Palacios, Corojal, and La Sierpe, respectively, while rainfall decreased by 4,4% and 15,2% in Los Palacios and Corojal and increased by 36,3% for La Sierpe.
The above values for these three sites suggest that, under the studied climate scenarios, an increase in crop ETc of approximately 30% can be expected, related to the expected increases in temperatures.
However, this increase in ETc does not always lead to an increase in crop irrigation demand, which, as shown in Figure 8a, is closely related to rainfall behavior (Figure b) in each model.
As shown in Figure 8a, in model 1 (Hadgem3-gc31-II-SSP1.26), irrigation demand decreases for the Corojal and La Sierpe sites, although rainfall is only higher at the La Sierpe site. In model 2 (Mpi-esm1-2hr-SSP126), for the Los Palacios site, there is a decrease in irrigation demand despite a slight decrease in rainfall, while for a greater difference in rainfall, irrigation demand increases at the Corojal site and at La Sierpe, in correspondence with the increase in rainfall, irrigation demand also decreases. Model 3 (Mri-esm2-0-1.26) shows an increase in the amount of rainfall for all sites and, correspondingly, the amount of water demanded for irrigation also decreases. These variations in irrigation demand are influenced not only by the total rainfall during the crop's growth period, but also by its distribution within this cycle, as well as by soil type.
Acharjee et al. (2017)ACHARJEE, T.K.; LUDWIG, F.; VAN HALSEMA, G.; HELLEGERS, P.; SUPIT, I.: “Future changes in water requirements of Boro rice in the face of climate change in North-West Bangladesh”, Agricultural water management, 194: 172-183, 2017, ISSN: 0378-3774. in Bangladesh reported that ETo increased in the future, primarily due to rising temperatures, while potential rice water requirements (ETc) decreased by 6,5% and 10,9% for the RCP 4.5 and 8.5 scenarios, respectively, in 2050, and by 8,3% and 17,6% for these respective scenarios in 2080, compared to the 1980-2013 baseline.
Conclusions
⌅-
Independently of the climatic model use for the future rainfall´s evaluation, there is a diminution in the total annual rainfall´s values from west to the east, with average for the three models of 1644, 1467,5 and 1239,1 mm year-1 for the occidental, central and oriental zone, respectively.
-
From the global climatic models study under SSP1-2.6 scenario, rice water demand was higher in the Hadgem3-gc31 model with an average gross water requirement for the 6 study sites of 1337,5 mm in a 130 days’ rice growth cycle, it was 5,8 and 4,3% superior to the models Mpi-esm1-2hr-SSP126 y Mri-esm2-0-126, respectively.
-
The expected effect of the increase in temperatures predicted in climate change studies for Cuba will undoubtedly lead to an increase in potential crop water consumption (ETc); however, the uncertain behavior of rainfall between sites and models for the SSP1-2.6 scenario leads to variability in crop irrigation demand, which is strongly influenced by the amount of rainfall within the cycle and by its distribution.
-
Another possible source of variation in future crop irrigation demand, not studied in this result but reported by different authors, mainly in Asia, is the potential decrease in the crop cycle due to increased temperatures, which, in addition to decreasing water demand, could also lead to a decrease in crop yield, which should be the subject of future studies in Cuba