Impacts of climate change on water quantity,water salinity,food security,and socioeconomy in Egypt

Mohie El Din Mohame Omar ,Ahme Moustafa Ahme Moussa ,Reinhar Hinkelmann

a National Water Research Center,Fum Ismailiya Canal,Shoubra El-Kheima 13411,Egypt

b International Center for Agricultural Research in the Dry Areas,Cairo 2416,Egypt

c Nile Research Institute(NRI),National Water Research Center,El-Qanater 13621,Egypt

d Department of Water Engineering,Technische Universit¨at Berlin(TU Berlin)Campus El Gouna,El Gouna 84513,Egypt

Received 16 April 2020;accepted 30 August 2020 Available online 23 March 2021

Abstract Climate change might have direct impacts on water quantity in Egypt and lead to indirect effects on Mediterranean saltwater intrusion to groundwater,which exposes agriculture to vulnerability.This study investigated impacts of climate change on agriculture,with particular regard to food security and socioeconomy,and quantified the effectiveness of cropping pattern adaptation measures by integrating three mathematical models.The BlueM model was used for hydrological simulations of Nasser Lake under flooding scenarios to predict the water supply from the High Aswan Dam.The water and salinity balance(WB-SAL)model was adopted to estimate the water salinity in the Nile Delta.The simulated results from the BlueM and WB-SAL models were integrated with the agricultural simulation model for Egypt(ASME)to project cropping patterns,food security,and socioeconomy throughout the country.The results showed that future climate change will directly affect the total crop area;crop areas for 13 crop types;the self-sufficiency of wheat,rice,cereal,and maize supplies;and socioeconomic indicators.The proposed cropping pattern adaptation measures focus on fixing the crop areas of rice and orchards and providing half of the population with lentils,maize,onion,vegetables,milk,and meat.The adaptation measures have the potential to promote food security without causing deterioration of the socioeconomic situation.However,water availability has much more significant effects on food security and socioeconomy than cropping pattern adaptation measures do.Accordingly,the country should rationalize water use efficiency and increase water supply.© 2021 Hohai University.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords:Climate change;Water quantity;Salinity;Cropping pattern adaptation measures;Food security;Socioeconomy

1.Introduction

Egypt is almost entirely dependent on the water of the Nile River,which supplies an annual water volume of approximately 55.5 km3,accounting for 93% of the country"s conventional water resources.The annual total water demand is 81.3 km3,with approximately 86% for agriculture,2.5% for industry,and 11.5% for the domestic sector(Omar and Moussa,2016).There is a gap between water demand and water availability,which is compensated for by the reuse of drainage water,wastewater,and shallow groundwater.The uncertainty of climate change impacts is another challenge to the water resources system in Egypt.

Climate change affects air temperature and precipitation around the world.It tends to result in sea-level rise,thereby influencing groundwater hydraulics and causing greater seawater intrusion in many coastal aquifers(Ketabchi et al.,2016),including the coastal aquifers in Egypt.The global climate has become 0.5°C warmer over the last 100 years due to greenhouse gas emissions caused by human activities(WMO,2017).Air temperature rise and precipitation variation have the potential to change water flows and might lead to intensified extreme hydrological events.However,previous studies focusing on the uncertainty of water flow variation have produced contrasting results.Elshamy and Wheater(2009)used the bias-corrected statistical downscaling approach to downscale 17 general circulation models(GCMs)and projected that by the end of this century the Blue Nile flow would have a change rate ranging from－60% to 45%.Strzepek and McCluskey(2007)used five GCMs based on two emission scenarios and estimated the variations of the Nile flows entering Nasser Lake in 2050 and 2100 under 20 scenarios.Twelve reduced flow scenarios and eight increased flow scenarios were projected.Kotb(2015)presented six scenarios using the Quantifying Uncertainty in Model Projections(QUMP)model and the regional climate model(RCM),which showed a precipitation increase of 5%－11% and a flow increase of 7.15%－23.21% at the Dongola Station at the Nasser Lake entrance.Regarding the sea-level rise,Eissa et al.(2017)used an analytical model combining hydrogeological and geochemical characterization of aquifer geometry to provide saltwater and freshwater balances at the Ras El Hekma site on the Egyptian Mediterranean Sea coast.Sefelnasr and Sherif(2014)used the finite element subsurface flow system model and found that large areas in the coastal zone of the Nile Delta would be submerged in seawater.UNESCO(2013)quantified the regions vulnerable to sea-level rise in Egypt.The increase in shallow groundwater salinity and drainage water reuse would lead to a gradual increase in the salt content of irrigation water in the Nile Delta.In Egypt,possible risks induced by climate change include the following:(1)water scarcity due to the reduction of water supply from the Nile River and high water requirement for crops,and(2)salinity increase in irrigation water in the Nile Delta resulting from saltwater intrusion to shallow groundwater and water reuse.

Agriculture consumes approximately 86% of total water resources in Egypt.Irrigated crops contribute mostly to local food production and exports and therefore to the prevention of saltwater intrusion in the Northern Nile Delta by rice cultivation.Climate change in Egypt would decrease crop yields for most crops,with wheat yields expected to be reduced by up to 9% in 2030 and by close to 20% in 2060(Smith et al.,2013).Hence,it is necessary to comprehensively investigate future climate change impacts on the Nile River flow,evapotranspiration,and shallow groundwater salinity,and how those impacts will affect the agricultural sector,with regard to the vulnerability of food security and the socioeconomy.It is also important to quantify the effectiveness of adaptation measures.

The most direct and simplest adaptation measure is to encourage farmers to change cropping patterns.However,the consequences of changing cropping patterns might have adverse impacts on food security and the socioeconomy.Adhikari(2018)reported that farmers in Nepal abandoned rice,cereal crops,tubers,and sweet potatoes,and even abandoned agriculture on their lands throughout 15 drought events owing to climatic conditions in the period from 1972 to 2015.This negatively affected agricultural production and the socioeconomic situation throughout the country.Alabdulkader et al.(2016)reported adverse effects of climate change on the agricultural sector in Saudi Arabia,and presented changes in the date palm cropping pattern as an adaptation measure,which could increase net annual return and decrease water demand.Chebil et al.(2019)investigated the impacts of climate change on 21 strategic crops in Tunisia and found that the total cropped area and agricultural employment in Tunisia would be negatively affected.However,the study did not present the effectiveness of changing cropping patterns as an adaptation measure.

In Egypt,few previous studies have quantified the effectiveness of changing cropping patterns as climatic adaptation measures.To adapt to the possible impacts of climate change on agriculture in Egypt,including reduction in Nile River flows,air temperature rise,Mediterranean sealevel rise,and salinity increase,changing the cropping structure is required.Therefore,the main objective of this study was to investigate impacts of climate change on water quantity and salinity in Egypt and their consequent impacts on food security and the socioeconomy.In addition,the effectiveness of cropping pattern adaptation measures was quantified.To achieve these goals,climate change impact projections were conducted,with regard to the Nile River flow,air temperature,open water evaporation,evapotranspiration,precipitation,and aquifer salinity in the Nile Delta.Afterward,the impacts of climate projections and adaptation measures on food security indicators,self-sufficiency,and cropping intensity were investigated.Finally,the impacts of climate projections and adaptation measures on socioeconomic indicators,agricultural productivity,consumerproducer surplus,number of jobs,and water marginal value were quantified.

2.Materials and methods

In this study,data of climate change projections and demographic changes were collected.Afterward,the BlueM river basin management model was used to predict the water supply from the High Aswan Dam(HAD)based on future flooding scenarios and constraints of reservoir storage volume.Meanwhile,the water and salinity balance(WB-SAL)model was adopted to predict salinity changes in the Nile Delta region according to future HAD release from the BlueM model and the expected demographic and climatic conditions.Finally,the agricultural simulation model for Egypt(ASME)was used to predict the optimal cropping pattern,food security,and socioeconomic indicators in the future based on the results of the BlueM and WB-SAL models.Fig.1 shows the study area.

2.1.Data acquisition

Fig.1.Study area.

Three data sets were obtained.The first was the input data for the BlueM model,which included climate change projections of the Nile River flow at the entrance of Nasser Lake,meteorological conditions in the lake region,seepage losses,and constraints controlling the lake operation processes.This BlueM-based simulation estimated the water release from the HAD.The output data in addition to the second data set were used as input data for the WB-SAL model.The second data set mainly focused on the Nile Delta region and consisted of meteorological conditions,evapotranspiration,the Mediterranean seawater intrusion rate to shallow groundwater,volumes and salinities of water supplies,and salt emissions from agricultural crops and industrial sectors.The outputs of the WB-SAL model were the salinities of irrigation water and drainage water.The third data set included the data affecting agriculture throughout the country,including water resources,demographic changes,agricultural areas,and socioeconomic conditions.These data,in addition to the results of the BlueM and WB-SAL models,were used as the input data for the ASME model.

2.1.1.Climate change projections

Many studies have combined GCMs and RCMs with hydrological models to investigate the link between climate and water resources.GCMs estimate the effects of emissions on global climate and describe the physical processes in the atmosphere,in the oceans,and on the land surface.By contrast,RCMs add detailed information about future climate change on fine scales to the GCMs’large-scale projections.At the Dongola Station at the Nasser Lake entrance,Strzepek and McCluskey(2007)used GCMs to predict water flow in 2050.In terms of RCMs,Kotb(2015)used six perturbed physical ensembles developed by the QUMP model to downscale climate projections over the Nile River Basin.Because RCMs account for the regions with mountains,coastlines,and islands on the scale of 100 km or less,and RCMs are more realistic than GCMs,this study utilized the results of Kotb(2015)that all future flows would increase.This study also adopted the findings of Strzepek and McCluskey(2007)that in six out of ten cases,river flow would decrease.Additionally,flow reductions projected from GCMs were considered,because the risks of flow reduction in arid countries such as Egypt cannot be ignored.Therefore,this study constructed one scenario that represented an average river flow increase of 14.3% according to RCMs and another scenario with an average flow decrease of 11.8% according to GCMs.Table 1 presents various climate change projections for the Nile River flow,air temperature,precipitation,open water evaporation,evapotranspiration,and Mediterranean saltwater intrusion rate.

2.1.2.Future planning alternatives

The present study collected other data affecting water demand.The population increase,expansion of agricultural areas,and establishment of new communities were combined with different climate scenarios.In 2050,population would be expected to reach 159 956 808 in Egypt(UN,2019).Due to the agriculture expansion plan,Egypt is expected to increase agricultural area by 315 000 hm2(Omar and Moussa,2016).

2.2.Mathematical modeling

2.2.1.BlueM model

The BlueM model was used in this study to simulate the operational process of Nasser Lake considering both the decrease and increase of river flow entering the lake.The BlueM model was developed by the Darmstadt University of Technology,in Germany,for river basin management(Bach et al.,2009).The operation of Nasser Lake was described by the water balance equation under various constraints concerning storage volume,outflow from the lake,and water losses,and water balance calculation was performed on a monthly basis,as follows:

Table 1Changes in different parameters in Egypt in 2050 due to climate change.

whereItis the mean inflow in montht(m3);Qtis the outflow from the dam(m3);Mtis the flow released from the emergency spillway(m3);Dtis the water demand of the Toshka Project for reclamation of 226 800 hm2from Nasser Lake(m3);Ttis the volume of water released from the Toshka Spillway to empty the reservoir down to the level of 175 m before floods(m3);Stis the seepage losses from the lake(m3),which was computed using the approach of Ebaid and Ismail(2010);andEtis the mean open water evaporation from the lake(m3),which is expressed as follows:

whereAtandAt+1are the lake areas at the beginning and end of montht,respectively;andCtis the evaporation coefficient in montht,which was calculated using the method of Ebaid and Ismail(2010).Model calibration was conducted by comparing the simulated daily water level upstream of the HAD with the observations,with the absolute error as the error index.

2.2.2.WB-SAL model

Owing to the intensive drainage water reuse,salt fluxes from the Nile Delta to the Mediterranean Sea and coastal lakes are higher than the fluxes entering the delta.As climate change might result in severe water shortage and saltwater intrusion,the salinity of drainage,surface irrigation,and shallow groundwater in the delta region may increase in the future.This would threaten agricultural production and its related socioeconomic aspects in the delta region.Therefore,it is necessary to project the future salinity of water bodies in the delta region.In general,traditional water quality models simulate the concentration of materials in any water system as a function of space and time based on different water volumes and loads entering and exiting the system.This means that the salt loads or concentrations of agricultural,municipal,and industrial disposals should be collected,measured,or assumed as the model input.However,it is a difficult,time-consuming,and high-cost process to collect these data in large basins such as the delta region.Hence,the WB-SAL model was used to project future salt emissions of agricultural,municipal,and industrial disposals to water bodies.The loads of agricultural drainage were estimated according to cropping patterns.The loads of industrial wastewater were based on industry categories and their share percentages.Afterward,the model was used to estimate the salinity of mixed drainage and irrigation water,with the total dissolved solids(TDS)as the salinity indicator.In the WB-SAL model,the salt balance is calculated based on the fact that dissolved salts are conservative and not subject to transformation,sedimentation,or biomass removal.When water is evaporated,the dissolved salts are left.The salt balance equation in the WB-SAL model is expressed as follows:

whereVoandCoare the volume and salinity of drainage water that discharges to the Mediterranean Sea,respectively;ViandCiare the volume and salinity of the Nile River flow entering the delta region,respectively;VpandCpare the volume and salinity of precipitation,respectively;VRandCRare the volume and salinity of reused drainage,respectively;andVDandCDare the volume and salinity of water diverted for different demands,respectively,and are expressed as follows:

whereVAandCAare the volume and salinity of diverted water for agricultural demand,respectively;VMandCMare the volume and salinity of diverted water for municipal demand,respectively;VIandCIare the volume and salinity of diverted water for industrial demand,respectively;andVAqandCAqare the volume and salinity of diverted water for aquacultural demand,respectively.The salinity of irrigation water was estimated using the following equation:

whereVSandCSare the volume and salinity of surface water,respectively;andVSGandCSGare the volume and salinity of shallow groundwater,respectively.

In this study,drainage salinity was calculated as a mixture of agricultural drainage,industrial wastewater,treated wastewater,untreated wastewater,and aquaculture drainage.Some drainage water is reused,and the remaining drainage water is discharged to the Mediterranean Sea and northern lakes via drainage systems.Hence,the salinity of reused water was defined as drainage salinity.Salt loads from agricultural drainage and industrial wastewater were estimated from emission factors.For agricultural drainage,emission factors for different crops were adopted according to those from Deltares(2012).Based on current cropping patterns,salt loads were separately estimated using the WBSAL model.Similarly,loads from industrial wastewater were derived from the emission factors of various sectors and their percentage of the total industry.Industrial emission factors were adopted according to those developed by Wahab and Badawy(2004).The share percentages of various industrial sectors in each governorate were taken from Deltares(2012).The share percentages of various industrial sectors for the entire delta region were calculated.The emission loads were added to the original loads in the supplied water.The salinity from agriculture and industry was calculated by dividing the salt loads by the volumes of agricultural drainage and industrial wastewater,respectively.Salt emission loads were calculated separately and were linked to water and salt pathways as shown in Fig.2.

The salinities of treated and untreated wastewater were estimated according to the findings of El Gammal(2011),who collected water quality data of raw wastewater at seven locations in Egypt.This study defined the salinity required by the WB-SAL model as the average salinity at the seven locations in El Gammal(2011).The salinity of aquacultural drainage water was adopted from the findings of Emara et al.(2016),who analyzed the effluent of an aquacultural farm discharging to Burullus Lake in the northern Nile Delta on a seasonal basis,and found an average TDS of 3 000 mg/L.Based on the predictions from the BlueM model,factors affecting Nile River flows were altered.Due to both cultivation expansion and urbanization,the agricultural area will change.Total irrigation demand was estimated from the agricultural area as well as from water requirements for crops,which will increase due to climate change.Municipal water demand was dependent on population and the water consumption rate per capita.Treated and untreated wastewater volumes were estimated from wastewater discharge per capita and the ratio of the volume of treated wastewater to that of total wastewater.The calibration process was performed by comparing the simulated salinity of drainage water to the coastal lakes and the Mediterranean Sea with the observations.

Fig.2 shows the water and salt balances calculated by the WB-SAL model.The current annual conventional water resources in the delta region are 31.7 km3from the Nile River and 1.0 km3from effective rainfall.By contrast,the current annual water demand is 38.39 km3for agriculture,4.63 km3for municipal use,1.16 km3for industry,and 0.80 km3for aquaculture.The gap between water demand and supply is approximately 12.67 km3,which is compensated for by the recycling of drainage water and wastewater as well as utilization of shallow groundwater.

Fig.2.Water and salt balances calculated by WB-SAL model.

2.2.3.TDS measurements

Water sampling was conducted for the Nile River flow entering the delta.The measured TDS was used as the input data for the WB-SAL model.Drainage water sampling was also conducted at the drainage sites of Bahr Hadus,El Serw,Bahr El Baqar,El-Omoum,Nasser,East Burullus,and West Burullus before drainage water flowed into the coastal lakes and the Mediterranean Sea.The measured TDS concentrations at the drainage sites were used for comparison with the simulated TDS from the WB-SAL model.Water sampling was performed in both summer and winter.

2.2.4.ASME model

ASME is an optimization model that determines the cropping patterns needed to produce the highest agriculture-related welfare with a certain availability of water.The optimization mode of ASME has no restrictions on crop area but merely has rotational restrictions to ensure feasible cropping patterns.In this study,several restrictions were set up,including fixed crop areas for specific crops,controlled import rates,and securing part of the population with specific crops.These restrictions enabled ASME to predict the optimal cropping pattern to achieve optimal food security and socioeconomy under different climatic scenarios.In addition,they enabled ASME to quantify effective adaptation measures.

ASME is programmed with the General Algebraic Modeling System.The mathematical model was developed and applied at a governorate level using linear programming.The model is static when irrigation water is allocated for a year among different users.The objective function of ASME is to maximize the welfare of producers by aggregating gross margins from crop production with limited land and water resources.The objective function(Z)maximizes the total gross margins from all crops by selecting the optimal crop grouping subject to a set of constraints.In addition,the maximization of net return in the short run is equivalent to the maximum gross margin.The objective function at each level of mathematical analysis is as follows:

whereZrepresents the value of the objective function(USD),Pijis the unit price of cropjin governorateipaid to the producers(USD/t),Yijis the yield per unit area for cropjin governoratei(t/hm2),Cijisthe investment costper unitareafor cropjin governoratei(USD/hm2),andXijis the crop area for cropjin governoratei(hm2).Theoptimalareaofeachcropdependsonthe total amount of water and crop water demand.The maximization of gross margin per unit area is equivalent to the maximum gross margin per unit water.This helps to estimate the total quantity of water that should be used for a given crop:

whereWtis the total amount of water available for irrigation(m3),andWijis the crop water demand for cropjin governorateiin monthm(m3).

The ASME model decomposes the commodities into directuse crop commodities,processing-input crop commodities,processing-output crop commodities,processing-by-product crop commodities,and livestock commodities.The water system in ASME consists of the Nile River water from the HAD and irrigation canals.Variation of future water supply from the HAD due to climate change estimated from the BlueM model was considered in future scenarios.Evaporation from open water and fallow land was considered.Agricultural water use refers to evapotranspiration,part of which is compensated for by effective rainfall.Net of rain and losses abstracted from the Nile system werecalculatedatagovernoratelevel.Evaporationwasestimated at the national level and was proportionally distributed over the study area.Changes in rainfall,evaporation,and evapotranspiration arising from climate change were obtained and considered in future scenarios.Future salinity alteration in the delta region calculated by the WB-SAL model was considered in the delta governorates.Municipal,industrial,and agricultural drainage water returns to the Nile River,canals,sea,or lakes via gravity or pumping stations.Additionally,ASME calculated the deep groundwater balance in New Valley,Matruh,Red Sea,and South Sinai governorates.To calibrate the ASME model,the estimated volume of reused drainage was compared with the observations.

2.2.5.Measures in adaptation to climate change

Mitigation measures for emission reduction are not enough,because certain greenhouse gases will still exist in the atmosphere.Adaptation is essential in reducing these impacts.We enacted a set of cropping pattern adaptation measures as follows:

(1)Only 294 000 hm2of rice was cultivated in coastal governorates,such as Beheira,Kafr El Sheik,Dakhalia,Damietta,Sharkia,and Port Said.

(2)Orchard area was fixed at the 2017 level based on a questionnaire in this study in the five governorates of Fayoum,Qena,Damietta,Behira,and Sharkia.It was found that 96%of the farmers prioritized orchards in the case of water shortage.

(3)Half of the population demand for lentils,maize,onions,vegetables,meat,and milk was produced in each governorate to guarantee the food self-sufficiency of the rural population.

(4)According to FAO(2011),Egypt lost 13%－15% of available cereal between harvesting and final consumption.Hence,a loss of 14% was considered in this study.

2.2.6.Modeling scenarios

The current scenario and three future scenarios were established(Table 2).Due to the uncertainty of climate change projections of Nile River flows and water salinity in the delta region,the three future scenarios were set to investigate the impacts of all possible projections and the suggested cropping pattern adaptation measures on cropping patterns,food security,and the socioeconomy in Egypt.

Scenario 1 investigated the effects of the reduction in water supply from the HAD and the consequent increase of water salinity in the delta region.The reduced water supply from theHAD was estimated from the BlueM model based on the GCM-predicted average reduction percentage of the Nile River flow at the Nasser Lake entrance(11.8%)(Table 1).The change of water salinity in the delta region was estimated by the WB-SAL model,based on the average reduction rate of the Nile River flow at the Nasser Lake entrance and such parameters as air temperature,precipitation,open water evaporation,evapotranspiration,and shallow groundwater salinity in the delta region(Table 1).This scenario did not assess adaptation measures,but only assumed that farmers would adapt to the reduction of cultivated area and livestock as well as the same 30 crop types currently produced.A linear relationship was assumed between water availability and cultivation area.Additionally,this scenario assumed that water shortage would be proportionally distributed across governorates.The change rate of salinity in irrigation water was considered in coastal governorates such as Beheira,Kafr El Sheik,Dakhalia,Damietta,Sharkia,and Port Said.

Table 2Basic information for current status and three future scenarios.

Scenario 2 evaluated the impacts of water supply increase from the HAD and the consequent water salinity reduction in the delta region.The increase in water supply from the HAD was estimated by the BlueM model based on the GCMprojected average increase percentage of the Nile River flow at the Nasser Lake entrance(14.3%)(Table 1).The results of Scenarios 1 and 2 were compared to estimate the impacts of the Nile River flow and water salinity in the delta region on cropping pattern,food security,and the socioeconomy in Egypt.

Scenario 3 evaluated the effectiveness of the cropping pattern adaptation measures as described in Section 2.2.5.The adaptation measures were utilized in the case of the reduction in Nile River flows.This scenario used the same reduction rate of the Nile River flow and growth rate of salinity as used in Scenario 1.The results of Scenarios 1 and 3 were compared to indicate the possible impacts of adaptation measures on cropping patterns,food security,and the socioeconomy in Egypt.

3.Results

3.1.Results of model calibration

With regard to the BlueM model,the average monthly water level data upstream of the HAD in the period from 2013 to 2017 were used for model calibration.Fig.3 shows that the BlueM model provided a satisfactory water level simulation with an absolute error index ranging from－0.23 m to 0.17 m.As for the WB-SAL model,the relative error for the salinity of water drainage to the Mediterranean Sea and coastal lakes was 1.45%,indicating a strong simulation performance(Fig.4).The ASME model obtained a relative error of 3.38% for the volume of reused drainage water(Fig.4),and the modeling performance was satisfactory.

3.2.Vulnerability of HAD water supply and salinity of water in delta region

Climate change in the Nile River Basin and in Egypt will expose the water balance,water salinity,food security,and the socioeconomy in Egypt to vulnerability in 2050.Climate change will alter the flow entering Nasser Lake.Based on the BlueM model,the mean annual water supply from the HAD under drying climate scenarios would be 64.4 km3and 48.8 km3with an average inflow change rate of 14.3% and－11.8%,respectively.Based on the WB-SAL model,the salinity of drainage water to the sea would increase from its current level of 1 859 mg/L to 2 073 mg/L under Scenario 1 with a growth rate of 11.5%,and the salinity of irrigation water would rise from 361.7 mg/L to 490.2 mg/L with a growth rate of 35.5%.By contrast,the drainage salinity and irrigation salinity would decrease to 1 561 mg/L and 305.4 mg/L,respectively,in Scenario 2.Notably,the salinity changes in the delta region under future scenarios could be attributed to the alteration of the five factors:water supply from the Nile River,rate of seawater intrusion to shallow groundwater,population,evaporation,and agricultural area.Given that all these factors,except for the Nile River flow,were assumed to be the same under future scenarios,the water salinity in the delta region would be inversely proportional to the Nile River flow.

Fig.3.Simulated mean monthly water level upstream of HAD in period from 2013 to 2017 using BlueM model.

Fig.4.TDS of drainage water flowing to sea simulated by WB-SAL model and volume of reused drainage water simulated by ASME model.

3.2.1.Vulnerability of cropping patterns

The simulated results from the BlueM and WB-SAL models were integrated with the ASME model to project future changes in cropping patterns.For the entire country,crop area for each crop type would significantly change in the three future scenarios(Fig.5).In comparison with the current situation,the crop areas for 13 crop types would decrease in Scenario 1 and increase in Scenario 2.These crop types are faba bean,groundnut,long berseem,lentils,Nili maize,summer onion,soybeans and sunflower,short berseem,sugar cane,cotton,Nili sorghum,and Nili tomato.Only the crop area for sugar beet would rise in Scenario 1 and drop in Scenario 2.In both scenarios,the crop areas for barley,summer maize,and rice would decrease.For the eight crop types of orchard,other legumes,summer tomato,winter tomato,Nili vegetables,summer vegetables,winter vegetables,and wheat,the crop areas were projected to increase in both scenarios.The total crop area would decrease from the current amount of 5 516 700 hm2to 4 146 660 hm2in Scenario 1 and 5 907 300 hm2in Scenario 2.

3.2.2.Vulnerability of food security

The projected cropping patterns in future scenarios will directly impact future food security(Table 3).Cropping intensity,which refers to the fraction of cultivated area that is harvested,is an important indicator of food security.A cropping intensity value exceeding 100%indicates that more than one cropping cycle is permitted in a year in the same region.As shown in Table 3,the cropping intensity would decrease from its current level of 161.07%－124.85% in Scenario 1.By contrast,it would increase to 169.05% in Scenario 2.In Scenario 1,the self-sufficiencies for various crops were projected to decrease,and the decrease magnitude for wheat,rice,cereal,and maize were 7.56%,83.8%,19.55%,and 16.80%,respectively.The self-sufficiencies in Scenario 2 were higher than those in Scenario 1,with those for wheat,rice,cereal,and maize in Scenario 2 being 19.71%,19.31%,26.31%,and 27.7% higher,respectively,than those in Scenario 1.

Fig.5.Cropping patterns in Egypt in current situation and three future scenarios.

Table 3Food security at current level and in three future scenarios.

3.2.3.Socioeconomic vulnerability

Evaluation of the socioeconomic situation in the agricultural sector depends on many criteria,including agricultural productivity at the farm-gate price,consumer-producer surplus,number of jobs,and water marginal value(Table 4).The farm-gate value of an agricultural product is the net price of this product when it leaves the farm after marketing costs are subtracted.Compared with the current status,the net agricultural productivity at the farm-gate price was projected to decrease in Scenario 1.It would be 2.49×109USD higher in Scenario 2 than in Scenario 1.Consumer-producer surplus is an important indicator for assessment of the economic progress in the agricultural sector,and it was calculated as the difference between the revenue of total crops and the combined costs for crop harvesting and irrigation.The consumerproducer surplus in Scenario 2 was projected to be 1.086×1010USD higher than that in Scenario 1.Water marginal value measures the willingness of water users to pay for additional water when water demand is not sustained.Water marginal value is related to water scarcity,and it was reformulated into a scarcity cost indicator.The reduction in Nile River flows in Scenario 1 led to an increase in water marginal value,indicating that water would become scarcer.The marginal value in Scenario 2 was projected be 52.17%lower than that in Scenario 1.Meanwhile,the number of jobs for crop production,animal husbandry,and agro-processing were projected to be higher in Scenario 2 than in Scenario 1.

3.3.Effectiveness of adaptation measures

The cropping pattern adaptation measures in Scenario 3 included fixing rice cultivation areas in coastal governorates and fixing the cultivation areas of orchards,lentils,maize,onions,and vegetables in all governorates.The adaptation measures have been suggested in the case of the reduction in Nile River flows and water salinity increase.Hence,the values of water supply from the HAD and water salinity in the delta region in Scenario 3 were defined as same as those in Scenario 1.The comparison of Scenarios 1 and 3 helps to clarify the effectiveness of adaptation measures,because all other factors were the same in these two scenarios.In Scenario 3,the cropping areas of faba beans,long berseem,other legumes,soybeans,sunflowers,sugar beets,short berseem,cotton,and Nili sorghum were predicted to increase.However,those for groundnuts,sugar cane,and sesame were projected to decrease.The total crop area would increase from 4 146 660 hm2in Scenario 1 to 4 373 040 hm2in Scenario 3(Fig.5).

With regard to food security,the adaptation measures in Scenario 3 increased the cropping intensity by 5.02% in comparison with those in Scenario 1(Table 3).These measures also augmented the self-sufficiencies of wheat,rice,cereal,and maize by 9.57%,22.23%,0.54%,and 1.20%,respectively(Table 3).Clearly,the cropping pattern adaptation measures had a significant impact on self-sufficiencies of rice and wheat but played a weak role in self-sufficiencies of cereal and maize.The self-sufficiencies of sugar and meat were projected to decrease in both Scenarios 1 and 3,and milk production would fully sustain the demand in these two scenarios.

Although the cropping pattern adaptation measures increased the total cropped area and improved food security in comparison with Scenario 1,they did not influence the socioeconomic situation.As shown in Table 4,the current net agricultural productivity at the farm-gate price,consumerproducer surplus,water marginal value,and number of jobs for crop production,animal husbandry,and agro-processing in both Scenarios 1 and 3 had similar values.

4.Discussion

This study evaluated the vulnerability of the water supply from the HAD,water salinity in the Nile Delta region,foodsecurity,and the socioeconomy in Egypt in response to climate change by the year 2050.The mean annual water supply from HAD release showed a large uncertainty,either with an average increase of 14.30% or a decrease of－11.80%.The salinity of irrigation water in the Nile Delta would increase by 35.5% in the case of the reduction in Nile River flows.

Table 4Socioeconomic situation currently and in three future scenarios.

It was concluded that crop area for each crop type throughout country would be vulnerable to climate change in 2050.A clear and direct relationship existed between Nile River flows and crop areas for crop types such as faba beans,groundnuts,long berseem,lentils,Nili maize,summer onions,soyabeans,sunflowers,short berseem,sugar cane,cotton,Nili sorghum,and Nili tomatoes.Meanwhile,only the sugar beet area was inversely proportional to the Nile River flow.The Nile River flow had an insignificant relationship with the crop areas for barley,summer maize,rice,orchards,other legumes,summer tomatoes,winter tomatoes,Nili vegetables,summer vegetables,winter vegetables,and wheat.In general,the projected total crop area was proportional to the Nile River flow.

This study found that the reduction in Nile River flows and the increased water salinity in the delta region tended to reduce the self-sufficiencies of wheat,rice,cereal,and maize,and vice versa.Therefore,food security in Egypt would be vulnerable to climate change.In addition,this study quantified the net agricultural productivity at the farm-gate price,consumer-producer surplus,water marginal value,and employment for crop production,animal husbandry,agroprocessing,and the agricultural sector.It was found that the reduced Nile River flow and the increased salinity in the delta region would deteriorate the socioeconomy of the agricultural sector.

This study proposed a set of cropping pattern adaptation measures,including fixing rice crop areas in coastal governorates and fixing the cropping areas of orchards,lentils,maize,onions,and vegetables in all governorates.These adaptation measures would be enacted only in the case of the reduction in the Nile River flow.The measures would result in an increase in the crop areas of faba beans,long berseem,other legumes,soybeans,sunflower,sugar beets,short berseem,cotton,and Nili sorghum but would lead to a decrease in the crop areas of groundnuts,sugar cane,and sesame.As a result,the total crop area would increase.The adaptation measures would be effective in improving the food security status in the country owing to the increase of cropping intensity and the self-sufficiencies of wheat,rice,cereal,and maize.This study found that when these measures were adopted,the net agricultural productivity at the farm gateprice,consumer-producer surplus,water marginal value,and employment for crop production,animal husbandry,and agroprocessing would not significantly change in comparison with the situation in which no measures were considered.Therefore,the improved food security status would come at the expense of socioeconomic status.

These results agree with the findings of Chebil et al.(2019),who adopted the agricultural supply model for Tunisia(ASMOT)using the data for 21 strategic crops.They found that the total crop area and agricultural employment in Tunisia were negatively affected due to the decreased irrigation water availability arising from climate change.The findings of this study are also in line with those of Adhikari(2018),who reported that 15 drought events took place in Nepal from 1972 to 2015 in climate change conditions.The adaptation measures included abandoning the crop types of rice,cereal crops,tubers,and sweet potatoes,which negatively affected agricultural production and the socioeconomic situation.

Comparison of the results in the three future scenarios clearly demonstrated that the Nile River flow would be largely impacted by climate change,which would subsequently affect the water salinity in the delta region,food security,and the socioeconomy.The Nile River flow reduction of 11.80% at the Dongola Station at the Nasser Lake entrance would reduce the average self-sufficiency of strategic crops by 31.08% and net agricultural productivity by 2.316×109USD,and lead to a loss in daily part-time income for 206 000 laborers.The reduction in the Nile River flow combined with cropping pattern adaptation measures would lower the average self-sufficiency of strategic crops by 21.29% and net agricultural productivity by 2.357×109USD,and result in a loss in daily part-time income for 203 000 laborers.However,the Nile River flow increase of 14.30% would augment the average selfsufficiency of strategic crops by 6.06% and net agricultural productivity by 1.73×108USD.It would lead to a gain in daily part-time income for 23 000 laborers.Therefore,it can be concluded that an increase in the Nile River flow would have more positive effects on food security and the socioeconomy than the cropping pattern adaptation measures.

5.Conclusions

In this study,the impacts of climate change on food security and the socioeconomy in Egypt were investigated,and the effectiveness of cropping pattern adaptation measures was evaluated by integrating three mathematical models.The main conclusions of this study can be summarized as follows:

(1)The vulnerability analysis in this study demonstrated that climate change would significantly affect the Nile River flow in Egypt.The Nile River flow would have an inverse relationship with the water salinity in the delta region.A reduction in the Nile River flow would reduce the average selfsufficiency of strategic crops,net agricultural productivity,and consumer-producer surplus.It would increase the number of laborers who lose their daily part-time income.In contrast,an increase in the Nile River flow would augment the average self-sufficiency of strategic crops,net agricultural productivity,and consumer-producer surplus and reduce the number of laborers with decreased daily part-time income.

(2)This study proposed to maintain a fixed crop area for rice in nine coastal governorates and to maintain fixed crop areas for orchards,lentils,maize,onion,tomatoes,and vegetables.These cropping pattern adaptation measures would increase the total crop area and improve food security without causing a deterioration of socioeconomic indicators.(3)Water availability would have more significant impacts on food security and socioeconomic status than cropping pattern adaptation measures.Accordingly,the country should increase water use efficiency and water supply to preserve agricultural productivity,food security,and socioeconomic health under different climate change scenarios.

Declaration of competing interest

The authors declare no conflicts of interest.