Also available as PDF (figures and tables appear in the PDF).
Growing competition for freshwater, increasing evapotranspiration, irregular rainfall and recurrent drought are trans- forming irrigation scheduling from a routine farm operation into a central component of climate-resilient crop produc- tion. Smart irrigation scheduling integrates soil, plant, weather and crop information to determine when irrigation should be applied, how much water should be supplied and where water is most needed. This review synthesizes the physiological, agronomic and technological basis through which smart irrigation scheduling improves water productivity and sustains crop yield under water-limited conditions. Conventional calendar-based irrigation frequently fails to account for spatial and temporal variability in soil moisture, atmospheric demand and crop developmental sen- sitivity, resulting in over-irrigation, deep percolation, nutrient loss or avoidable water stress. In contrast, soil-moisture sensors, evapotranspiration-based scheduling, canopy-temperature measurements, remote sensing, Internet of Things platforms, decision-support systems, artificial intelligence and machine-learning models allow irrigation to respond dynamically to actual crop and environmental conditions. Smart scheduling improves productivity principally by maintaining root-zone water within physiologically acceptable limits, reducing non-beneficial water losses, protecting flowering and reproductive development, improving nutrient uptake and coordinating controlled water deficits with crop tolerance. Its effectiveness, however, depends strongly on crop type, genotype, phenological stage, root-zone characteristics, irrigation method, sensor calibration, weather variability and economic feasibility. The plant micro- biome, stress-responsive genetic traits, hormonal regulation and soil biological properties add further complexity because they influence root water acquisition and crop tolerance to intermittent soil drying. Evidence from smart horticulture and precision-agriculture studies increasingly supports integration of IoT sensors, artificial intelligence, crop models and climate forecasts into adaptive irrigation systems. Future irrigation management should move from fixed water application toward predictive, closed-loop and plant-responsive scheduling capable of distinguishing beneficial water conservation from yield-threatening stress. Such integration offers a practical pathway for increasing crop water productivity, stabilizing yield and conserving scarce freshwater under increasingly water-limited production environments.
Keywords: smart irrigation; irrigation scheduling; water productivity; soil-moisture sensor; evapotranspiration; Internet of Things; artificial intelligence
Growingcompetitionforfreshwater,increasingevapotranspiration,irregularrainfallandrecurrentdroughtaretrans- formingirrigationschedulingfromaroutinefarmoperationintoacentralcomponentofclimate-resilientcropproduc- tion.Smartirrigationschedulingintegratessoil,plant,weatherandcropinformationtodeterminewhenirrigation shouldbeapplied,howmuchwatershouldbesuppliedandwherewaterismostneeded.Thisreviewsynthesizes thephysiological,agronomicandtechnologicalbasisthroughwhichsmartirrigationschedulingimproveswater productivityandsustainscropyieldunderwater-limitedconditions.Conventionalcalendar-basedirrigationfrequently failstoaccountforspatialandtemporalvariabilityinsoilmoisture,atmosphericdemandandcropdevelopmentalsen- sitivity,resultinginover-irrigation,deeppercolation,nutrientlossoravoidablewaterstress.Incontrast,soil-moisture sensors,evapotranspiration-basedscheduling,canopy-temperaturemeasurements,remotesensing,InternetofThings platforms,decision-supportsystems,artificialintelligenceandmachine-learningmodelsallowirrigationtorespond dynamicallytoactualcropandenvironmentalconditions. Smartschedulingimprovesproductivityprincipallyby maintainingroot-zonewaterwithinphysiologicallyacceptablelimits,reducingnon-beneficialwaterlosses,protecting floweringandreproductivedevelopment,improvingnutrientuptakeandcoordinatingcontrolledwaterdeficitswith croptolerance.Itseffectiveness,however,dependsstronglyoncroptype,genotype,phenologicalstage,root-zone characteristics,irrigationmethod,sensorcalibration,weathervariabilityandeconomicfeasibility.Theplantmicro- biome,stress-responsivegenetictraits,hormonalregulationandsoilbiologicalpropertiesaddfurthercomplexity becausetheyinfluencerootwateracquisitionandcroptolerancetointermittentsoildrying.Evidencefromsmart horticultureandprecision-agriculturestudiesincreasinglysupportsintegrationofIoTsensors,artificialintelligence, cropmodelsandclimateforecastsintoadaptiveirrigationsystems.Futureirrigationmanagementshouldmovefrom fixedwaterapplicationtowardpredictive,closed-loopandplant-responsiveschedulingcapableofdistinguishing beneficialwaterconservationfromyield-threateningstress.Suchintegrationoffersapracticalpathwayforincreasing cropwaterproductivity,stabilizingyieldandconservingscarcefreshwaterunderincreasinglywater-limitedproduction environments.
Highlights evapotranspiration,cropdevelopmentalstage,plant stress,weatherandexpectedrainfall[1,5–13]. • Smartirrigationschedulingapplieswateraccord- ingtocropdemandratherthanpredetermined FAO evapotranspiration methodology established calendarintervals[1–6].
a widely used framework in which reference evapo- • Soil-watersensors,plantindicators,weatherin- transpiration(ET )isadjustedbycropcoefficientsto formationandevapotranspirationestimatespro- estimate crop evapotranspiration (ETc) [1]. Modern videcomplementaryinformationforirrigation schedulingexpandsthisframeworkbycombiningET decisions[1,5–8].
estimates with soil sensors, canopy measurements, • Sensor-guided irrigation can substantially remote sensing, weather forecasts and automated reduceirrigation-waterusewhilemaintainingor controllers.
Systematicevaluationofsoil-watersensorsindicates • Artificialintelligence, IoTandDSSenablereal- thatsensor-guidedschedulingfrequentlyuseslessir- time,predictiveandsite-specificirrigationman- rigation water while maintaining or even improving agement[9–13,34,50].
cropyieldcomparedwithconventionalscheduling[7]. • Smartschedulingisparticularlyvaluableunder Emerging intelligent irrigation approaches similarly drought,heatandirregularrainfallbecauseirri- combineIoT,artificialintelligenceandpredictivemod- gationthresholdscanbemodifiedaccordingto ellingtoaccountforfieldhydrological, climaticand atmosphericdemand[14–18,33].
• Biologicalstresstolerance,rootarchitecture,hor- The significance of smart scheduling increases monalregulationandrhizospherefunctioning under water-limited conditions. When available influencecropresponsetowater-savingirriga- irrigationcannotfullysatisfycropevapotranspiration, tion[35,36,39,46–48].
management must distribute limited water among growth stages in a way that maximizes economic 1.Introduction yield or water productivity. Under such conditions, Agriculture is the dominant consumptive user of indiscriminatereductioninirrigationislesseffective freshwater in many regions, and increasing compe- thanstrategicallytimedwaterrestriction[3,4,19–23]. tition among agricultural, domestic, industrial and Theattachedliteraturefurthersupportsthismultidi- environmental sectors has intensified pressure to mensionalinterpretation.Climate-smarthorticultural produce more crop yield per unit of water supplied systemsemphasizeefficientirrigation,soil-watercon- [1–4]. Climate change further complicates this servation,protectedcultivationanddigitaltechnolo- challengethroughincreasingtemperatures,irregular giesascomplementarystrategiesforresource-efficient precipitation,higheratmosphericevaporativedemand cropproduction[44].Precision-horticulturesystems andmorefrequentdroughtandheatepisodes[14–18]. usesensorsandreal-timedatatoidentifysoil-moisture Thesechangesareparticularlyimportantinirrigated variabilityandtailorinputsaccordingtocropdemand agriculturebecausethetimingofwatersupplycanbe [50].
asinfluentialastotalseasonalirrigation. Smartirrigationshouldthereforebeunderstoodnotas Traditionalirrigationschedulingcommonlyrelieson aparticularsensororcontrollerbutasanintegratedde- fixedcalendars,farmerexperienceorpredetermined cisionframeworkconnectingsoil,plant,atmosphere irrigation intervals. Such approaches can work un- and management. This review examines the mech- derstableproductionenvironmentsbutmaybecome anisms by which such scheduling affects water pro- inefficientwhensoilproperties, weatherconditions ductivityandcropyield,themajortechnologicalap- andcropdevelopmentvaryovertime.Irrigationsup- proachesavailable,theinteractionwithwaterdeficit pliedtooearlycanincreaseevaporation,runoff,deep andclimatevariability, andfutureopportunitiesfor drainageandnutrientleaching,whereasdelayedirri- AI-enabledirrigation.
gationcanreducestomatalconductance, photosyn- thesis,cellexpansion,reproductivedevelopmentand finalyield[2–6].
2. Conceptual Basis of Smart Irrigation Scheduling Smart irrigation scheduling attempts to overcome theselimitationsbyansweringthreeinterconnected Irrigationschedulingistheprocessofdeterminingirri- questions:whentoirrigate,howmuchtoirrigateand gationtimingandapplicationdepthtomaintaincrop whereirrigationisrequired.Decisionsarederivedfrom waterstatuswithinadesirablerange.Smartscheduling measurementsorpredictionsofsoil-wateravailability, differsfromconventionalschedulingbycontinuously orperiodicallyupdatingirrigationdecisionsusingmea- mustthereforebalancewaterconservationwitheco- sured,modelledorforecastinformation[5–13]. nomicallyacceptableproduction.
Aneffectivesmart-irrigationsystemnormallycontains fourfunctionalcomponents.Thefirstis sensing,which 3. Information Sources for Smart Irrigation measures soil, plant or atmospheric variables. The Decisions secondis datatransmissionandstorage,commonly 3.1Soil-Moisture-BasedScheduling achievedthroughIoTnetworksandcloudplatforms. Soil-watermonitoringprovidesoneofthemostdirect Thethirdis decisionanalysis,usingthresholdrules, approachestodeterminingwhethersufficientwaterre- cropmodels,DSSormachinelearning.Thefourthis ac- mainswithinthecroprootzone.Technologiesinclude tuation,wherebyirrigationvalves,pumpsorvariable- tensiometers,capacitanceprobes,frequency-domain rateirrigationsystemsexecutetherecommendedwa- reflectometry,time-domainreflectometry,resistance- terapplication[9–13,34,50].
basedsensorsandmatric-potentialsensors[5–8]. Thesimplestsmartcontrollermayswitchirrigationon Sensorplacementiscrucialbecausesoil-waterdistri- whenvolumetricsoil-watercontentfallsbelowapre- butionvariesverticallyandhorizontally.Shallowsen- definedthreshold.Moreadvancedsystemsintegrate sorscanindicatedepletionintheactivesurfaceroot soilmoisturewithETc,rainfallforecasts,phenological zone, whereas deeper sensors can identify whether stageandexpectedatmosphericdemand.Closed-loop irrigationispenetratingbeloweffectiverootingdepth. systemssubsequentlyevaluatetheeffectofirrigation Asystematicreviewofsoil-watersensorsreportedthat andadjustfutureapplicationsaccordingly. sensor-guided scheduling generally reduced water Thisfeedbackstructuredistinguishessmartirrigation applicationrelativetoconventionalschedulingwhile from fixed irrigation programming. Rather than as- achieving similar or higher yield [7]. Such evidence sumingcropwaterdemand,itcontinuallyestimates supportssoilmoistureasausefulfoundationforsmart whethertherootzoneisapproachingaphysiologically irrigation,althoughcalibrationandinstallationerrors importantthreshold.
2.1Irrigationamountversusirrigationtiming 3.2Evapotranspiration-BasedScheduling Seasonalirrigationvolumealonedoesnotdetermine ET-basedschedulingestimatesthevolumeofwater cropresponse.Twofieldsreceivingidenticalamounts lostthroughsoilevaporationandplanttranspiration. ofwatercanproducesubstantiallydifferentyieldsif Referenceevapotranspirationcalculatedfromweather thetimingofirrigationdiffers[3,4,19–23]. variablesismultipliedbycropcoefficientstoestimate cropwaterdemand[1].
Water supplied during periods of low crop demand ETschedulinghas severalstrengths. Itdirectlycon- maybelostthroughdrainage,whereasthesamequan- nectswaterapplicationtoatmosphericdemand,can tityappliedaroundfloweringorreproductivedevelop- operateatlargespatialscalesandiscompatiblewith mentmaypreventsubstantialyieldloss.Consequently, weatherstationsandforecasts. However, cropcoef- smartschedulingseekstomaximizethe marginalcrop ficientsvarywithvariety, canopydevelopment, soil benefitperunitofirrigationwater.
evaporation,plantingdensityandmanagementcondi- tions.
2.2Waterproductivityastheoperationaltarget SmartsystemscanupdateKcvaluesusingremotesens- Waterproductivitycanbeexpressedascropyieldper ingorcanopydevelopment,improvinguponstatictab- unitofwaterconsumedorsupplied.Dependingonthe ulatedcoefficients.
purpose,denominatorsmayincludeevapotranspira- tion, irrigationwater, ortotalwaterinput. Improve- mentsinwaterproductivitythereforeariseeitherby 3.3Plant-BasedIndicators producingmoreyieldwithsimilarwaterconsumption Plants integrate the combined effects of soil-water orbymaintainingacceptableyieldwithlowerwater availability and atmospheric demand. Plant-based use[3,4,19–23].
scheduling therefore provides valuable information thatsoilsensorsalonemaynotcapture[5,6]. Underwater-limitedconditions,maximizingabsolute Usefulindicatorsinclude: yield and maximizing water productivity are not al- waysidenticalobjectives. Extremelyconservativeir- • leaforstemwaterpotential; rigationmaygeneratehighwaterproductivitywhile • stomatalconductance; significantly reducing total yield. Smart scheduling • sapflow; • canopytemperature; agronomic intelligence in addition to technological • cropwaterstressindex; connectivity.
Canopytemperatureisespeciallyattractivebecause decliningtranspirationgenerallyincreasesleaftemper- ature.Thermalcamerasmountedongroundplatforms, UAVsorsatellitescanthereforeidentifyareasexperi- encingwaterstress.
3.4WeatherForecasting Short-rangeforecastsofrainfall,temperature,solarra- diation,humidityandwindcanpreventunnecessaryir- rigationimmediatelybeforerainfallorallowadditional
ProductivityandCropYield This predictive capability is increasingly important becauseclimatevariabilitymakeshistoricalirrigation 5. Artificial Intelligence, Machine Learning schedulesunreliable[14–18,33,34,44].
andDecision-SupportSystems Artificial intelligence expands irrigation scheduling beyondsimplethresholdcontrol. Machine-learning 4.InternetofThingsandAutomatedIrrigation models can analyse nonlinear relationships among TheInternetofThingshastransformedirrigationau- soil moisture, temperature, humidity, rainfall, crop tomationbylinkingdistributedfieldsensorswithcon- stage, canopy response, irrigation history and yield trollers,mobiledevicesandcloud-basedplatforms[9– [9–13,34].
Decision-support systems integrate such data with IoT-enabledsystemscontinuouslymeasureroot-zone cropsimulationmodelsandmanagementrules.The soilmoisture,temperature,relativehumidityandother attached DSS paper describes integration of soil environmentalconditions. Measurementsaretrans- properties, climatic conditions, crop requirements mittedtosoftwareplatforms,wherepredefinedthresh- and forecasts to optimize irrigation scheduling and oldsorpredictivealgorithmsdeterminewhetherirri- improveresource-useefficiency[34].
Machinelearningisespeciallyusefulwhereirrigation Theattachedprecision-horticulturestudyspecifically responsedependsoninteractingfactors.Forexample, describesIoTirrigationsystemsthatmonitorsoil-water asoil-moisturevalueof20%mayrepresentadequate statuscontinuouslyanddeliverwateronlywhenre- supply on one soil but severe depletion on another. quired,reducingunnecessarywaterapplicationand Thecropresponsemaydifferfurtheraccordingtoroot allowingremotecontrol[50].
Similarly,theDSSarticledescribessensornetworks AImodelscanpotentiallylearntheseinteractionsand collectingsoilmoisture, temperature, humidityand predictirrigationrequirementdirectly. crop-growthinformationandtransmittingthesedata The attached machine-learning and multispectral- fordynamicirrigationrecommendations[34]. imaging review, although focused on plant disease Automation has several advantages under water- detection,demonstratesthecapacityofSVM,Random limited conditions. Short irrigation pulses can be Forest,CNNandtransformerarchitecturestointerpret delivered precisely when depletion thresholds are complex spectral and physiological crop signals reached, avoiding both severe stress and over- [37]. Multispectral sensing can detect changes in irrigation.Automationisalsoparticularlysuitablefor chlorophyll,tissuewaterrelationsandcellularcondi- dripandmicro-irrigationbecauseapplicationdepth tion before visible symptoms develop. These same canbeaccuratelycontrolled.
analyticalapproacheshaveconsiderablepotentialfor irrigationstressclassification.
However,automationcannotcompensateforincorrect thresholds.Apoorlycalibratedsensorconnectedtoan RemotesensingandAIshouldthereforebeviewedas automaticvalvecanrepeatedlymakeanincorrectirri- enablingtechnologiesforplant-responsiveirrigation gation decision. Smart irrigation therefore requires ratherthanmerelyasmonitoringtools.
Scheduling Primary informa- Majoradvantage Majorlimitation Smart-irrigationappli- approach tion cation Soilmoisture Volumetric water Directassessmentof Sensor placement Threshold-triggeredir- content/root-zone storedwater andcalibration rigation depletion Soil matric po- Soil-watertension Closely linked with Soil-specific inter- Irrigationonsetthresh- tential wateravailability pretation old ET-based ET0 × Kc Quantifies at- Dependsonaccurate Dailyirrigationreplace- mospheric crop Kcandweatherdata ment demand Plant water po- Leaf/stemwatersta- Direct physiological Labour-intensive Stress-threshold tential tus indicator measurements scheduling Canopytempera- Thermalresponse Rapid and spatially Stronglyaffectedby Thermal stress detec- ture scalable weather tion Remotesensing Thermal/multispectralField-scale spatial Resolution/cloud Variable-rateirrigation indices coverage constraints Weatherforecast Rainfall, VPD and Predictive rather Forecastuncertainty Delayoradvanceirriga- temperature thanreactive tion IoT+DSS Multi-sensor infor- Real-timeintegrated Connectivityandin- Automatedsmartirriga- mation decisions frastructure tion AI/ML Historical+real-time Detects complex Training data and Predictive irrigation multidimensional nonlinear relation- transferability scheduling data ships 6.WhySmartSchedulingImprovesWaterPro- erantcropstageswhileprotectingsensitiveperiods [3,4,19–23].
ductivity Smartirrigationincreaseswaterproductivitythrough Meta-analyticalevidenceshowsthatdeficitirrigation severalcomplementarymechanisms.
frequently raises water productivity but produces context-dependentyieldpenalties.Wheat,vegetables 6.1Reductionofnon-beneficialwaterlosses and other crops exhibit large variation according to Applyingwateraccordingtoroot-zonedemandreduces irrigation level, timing, soil and climatic conditions runoff,deeppercolationandunnecessarysoilevapo- [19–23].
ration. Dripandmicro-irrigationfurtherconcentrate Recentmeta-analysisfurtherindicatesthatmoderate wateraroundtheactiverootsystem[3,4,44]. deficitirrigationcanmaintainyieldmuchmoresuc- Climate-smarthorticulturalresearchemphasizesthat cessfullythanseveredeficitirrigation,reinforcingthe dripandmicro-irrigationreduceevaporation,runoff importanceofthresholdselectionratherthanindis- anddeep-percolationlosseswhileimprovingwater- criminatewaterreduction[23].
7.SmartIrrigationandCropYieldFormation 6.2Avoidanceofover-irrigation Maintainingyieldunderwaterlimitationrequirescoor- Over-irrigationcanreduceoxygenavailabilityinthe dinationbetweenirrigationandcropphenology. rootzone,increasenutrientleachingandpromoteun- necessaryvegetativegrowth.Smartschedulinglimits waterapplicationafterthetargetsoil-waterstatusis 7.1Vegetativegrowth restored.
Moderatesoil-waterdeficitsduringearlyvegetativede- velopmentmayreduceexcessiveleafareaandtran- 6.3Preventionofdamagingwaterstress spirationwithoutsubstantiallyaffectingfinalyieldin Yield loss often accelerates once crop water status somecrops.Severestress,however,limitscelldivision, crosses a critical physiological threshold. Smart cellexpansionandcanopyestablishment[14–18]. sensors enable irrigation before this threshold is exceeded.
7.2Floweringandreproductivedevelopment Floweringisoftenoneofthemostdrought-sensitive 6.4Strategicuseofcontrolleddeficit cropstages.Waterdeficitcanreducepollenviability, Wherefullirrigationisimpossible,smartscheduling fertilization,flowerretentionandseedorfruitset[16– allowswaterdeficittobeplacedduringrelativelytol- 18,33,47].
Theattachedreviewonhormonalregulationofrepro- acrosscrops.
ductivedevelopmenthighlightsthecoordinatedrole ofauxin,gibberellins,cytokinins,ABA,ethylene,brassi- 9.1Cereals nosteroids,jasmonatesandsalicylicacidinflowering, Wheatandmaizearemajortargetsforirrigationopti- fertilization,fruitsetandseeddevelopment[47]. mizationbecauselargeareasareproducedunderlim- Smartirrigationcanthereforeprotectreproductivepro- itedwater.
cessesbyprioritizingwaterduringthesecriticalstages. Inwheat,moderatedeficitirrigationcanincreaseWUE, buttheoutcomedependsonirrigationlevel,timing, 7.3Grainandfruitfilling precipitationandsoilproperties[20].
Waterstressduringfillingcanshortenassimilateac- Maizeisparticularlysensitivearoundtasselling,silk- cumulation, acceleratesenescenceandreducefinal ingandearlykernelestablishment.Smartscheduling grainorfruitsize. Inhorticulturalcrops,wateravail- shouldprioritizethesereproductiveperiodswhensea- abilitycanalsoaffectsolublesolids,texture,nutrient sonalwaterislimited.
Consequently, scheduling should incorporate both 9.2Vegetablecrops yieldquantityandqualitytargets.
Vegetable yield and quality respond rapidly to root-zone water status. A global meta-analysis 8. Water-Limited Conditions, Heat and Cli- demonstrates that water-productivity benefits of maticVariability deficitirrigationdependstronglyoncropspecies,soil Afixedsoil-moisturethresholddoesnotalwayspro- textureandproductionenvironment[21].
ducethesameplantresponse.Theseverityofagiven Tomato is particularly useful for smart irrigation re- soil-waterdeficitchangeswithatmosphericevapora- searchbecausefruityield,sizeandqualityrespondto tivedemand.
bothirrigationamountandenvironmentalconditions. HightemperatureandhighVPDincreasetranspiration Theattachedtomatoexperimentdemonstratesstrong demand,causingleafwaterpotentialtodeclinerapidly. genotype × sowing-windowinteractionsaffectingcrop Duringcombinedheatanddrought,plantsexperience yield under changing weather and disease environ- reductionsinphotosynthesis,waterrelationsandre- ments[31].Thebroaderimplicationisthatirrigation productiveperformancethatmaybegreaterthanun- schedulingcannotbeseparatedcompletelyfromcrop dereitherstressalone[33].
genotypeandseasonalenvironment.Weathercondi- Theattachedheat–droughtreviewreportsthatsimulta- tionssuchastemperature,rainfallandhumidityinflu- neousstressdisruptsphotosynthesis,nutrientuptake encedcropperformanceinthatstudy.
and reproductive development, enhances oxidative stressandmodifieshormonalbalance[33]. 9.3Fruitcrops Smart scheduling should therefore adopt climate- Fruitcropsrequiremulti-seasonmanagementandof- responsivethresholds.Duringaheatwave,irrigation tenhavedeepperennialrootsystems.Smartirrigation may need to be triggered at a higher soil-moisture cansupportregulateddeficitstrategies,butwaterre- levelthanduringmildweather.Conversely,irrigation strictionmustconsiderflowering,fruitset,rapidfruit canbedelayedwhenrainfallishighlyprobable. enlargementandreturnbloom.
Thisapproachchangesirrigationmanagementfrom Geneticandmoleculardifferencesamongfruitcrops reactivereplacementofdepletedsoilwater to antic- furtherinfluencetolerancetodrought,heatandsalinity ipatorymanagementofcropstressrisk.
Theclimate-resilienthorticultureliteraturesimilarly emphasizesdrought,heatandcombinedstressasma- 9.4Protectedhorticulture jor threats to crop performance and identifies WUE, Greenhousesandprotectedstructuresareparticularly rootarchitectureandmaintenanceofreproductivede- suitedtosmartirrigationbecausebothcropenviron- velopmentasimportantadaptivetraits[43]. mentandfertigationcanbetightlyregulated[44,50]. Theclimate-smarthorticulturearticlenotesthatirri- 9.Crop-SpecificResponses gationandfertigationcanbepreciselycontrolledun- Smartirrigationprinciplesarebroadlyapplicable,but derprotectedsystems,increasingoverallresourceeffi- threshold values cannot be universally transferred ciency[44].
Smartschedulingfunc- Immediateeffect Plant/cropconsequence Expected production tion benefit Irrigationbeforecritical Maintainsroot-zonewa- Preventsseverestomatalclo- Higher biomass and depletion tersupply sureandphotosyntheticde- yield cline Reducedunnecessaryir- Limits deep drainage Improves proportion of ap- Higher irrigation water rigation andrunoff pliedwaterusedbycrop productivity Stage-specificirrigation Protects flowering and Maintains reproductive suc- Greater harvestable grain/fruitset cess yield ET-basedadjustment Matches water supply Reduces under- and over- Stablecropwaterstatus with atmospheric de- irrigation mand Weatherforecastintegra- Anticipatesrainandheat Avoidsunnecessaryirrigation Water saving and yield tion andsevereheatstress protection Sensor+dripirrigation Delivers precise root- Improveswaterandnutrient Increased WUE and zonewater acquisition nutrient-useefficiency Variable-rateirrigation Respondstospatialfield Reduceswetanddryzones Improved uniformity variability andyield AIprediction Anticipates crop de- Enablesproactiveratherthan Greateryieldstability mand reactiveirrigation Controlleddeficitirriga- Conserveswaterduring Allocatessavedwatertocriti- Improved seasonal wa- tion tolerantstages calstages terproductivity 10.RootBiology,MicrobiomesandSmartIrri- functionalgenesandecologicalresilience[38].Metage- nomicsmayultimatelyhelpidentifymicrobiomestates gation associatedwithimprovedcropperformanceunderre- Theresponseofacroptoirrigationdependsnotonly peateddryingandrewetting.
onphysicalsoilmoisturebutalsoonrootarchitecture andbiologicalactivityintherhizosphere. 11. GeneticandMolecularTraitsInfluencing Beneficialsoilmicroorganismscanmodifyrootgrowth, nutrientacquisition,hormonesignalling,osmoticad- IrrigationResponse justmentandantioxidantdefence[35,36,39]. Smart irrigation cannot completely eliminate water stress.Cropgeneticcapacitythereforeremainsimpor- The attached functional-profiling review identifies tant.
PGPR,mycorrhizalfungiandendophytesascontrib- utorstodroughtandotherstresstolerancethrough Traitsrelevanttosmartschedulinginclude: phytohormonemodulation,osmolyteaccumulation • deepandproliferativerooting; andantioxidantactivation[35].
• rapidstomatalregulation; Rhizospheremicrobiomerestructuringcansimilarly • osmoticadjustment; influencerootarchitecture,nutrientcyclingandstress • antioxidantcapacity; resilience[36].
• reproductivestability; Endophytesprovideanadditionalinternalmechanism • rapidrecoveryafterrewatering; bymodifyingplanthormonesignalling,redoxregula- • tolerancetocombineddroughtandheat.
The attached climate-resilient horticultural review This has important implications for smart irrigation. highlights breeding for WUE, root architecture and Twocropsexperiencingthesamesoilmoisturemay reproductivestabilityunderdrought[43]. Molecular differineffectivewateruptakebecauseofrootarchi- andgenomicapproachesincludingDREB,WRKYand tecture,mycorrhizalassociationormicrobialeffects NAC transcription factors are proposed as tools for onrootphysiology.
Futuresmart-irrigationmodelscouldthereforeincor- Thegeneticandmolecularreviewoffruitcropssimi- porateindicatorsofroot-zonebiologicalfunctioning larlyidentifiesWRKY,NAC,MYB,bZIPandDREBfactors rather than treating soil simply as a physical water andhormone-mediatedsignallingasimportantregu- reservoir.
Theattachedmetagenomicanalysisalsoemphasizes WRKY-mediatedstress-responsenetworksarealsorel- thatmanagementpracticesaltermicrobialdiversity, evantbecauseirrigationmismanagementcanaggra- vatesalinityinirrigatedsoils.Theattachedsalt-stress Pollinationtechnologiesprovideanotherexample.The articlespecificallylinkssalinizationtoirrigationmis- sonicandvibrationalpollinationreviewdemonstrates managementanddescribesWRKYinvolvementinion thatfruitsetandyielddependstronglyoneffective homeostasis,antioxidantdefenceandstresssignalling reproductiveprocessesandthatautomationandAI [48].
canbeintegratedintoprotectedhorticulturalsystems [45].Smartirrigationandsmartpollinationarethere- The future objective should therefore be genotype- forecomplementaryratherthancompetingformsof specific irrigation scheduling, where thresholds are automationincontrolled-environmenthorticulture. adjustedaccordingtocultivarwater-usestrategyand stresstolerance.
Growing-media properties also influence water management. Peat-based substrates possess high water-holding capacity but interact with irrigation 12. SmartIrrigationwithinaWiderClimate- and fertilization and have important carbon and SmartProductionSystem greenhouse-gasimplications[49].Substrate-specific Irrigationcannotbeoptimizedindependentlyofother irrigation thresholds will therefore be essential in crop-managementpractices.
Nutrient supply influences root growth and water Theseexamplesbroadensmartirrigationintosmart uptake. Integratednutrientmanagementinbanana production-systemmanagement,wherewaterinter- demonstratedlargeimprovementsinproductivityand acts with substrate, nutrient, pollination, genotype, farmer-levelperformanceunderfieldconditions[32]. climateandcrop-qualityprocesses.
Although the study did not specifically investigate smart irrigation, it illustrates the importance of 14. Translating Smart Irrigation from Re- synchronizing resource-management technologies searchtoFarmers’Fields ratherthanoptimizingasingleinputinisolation. Technical effectiveness does not ensure adoption. Likewise, the commercial marigold technology- Smart irrigation systems can involve soil probes, assessmentstudydemonstratesthatfieldproductivity controllers, communication networks, weather dependsnotonlyontechnicaleffectivenessbutalso stations and subscription-based software, creating onfarmeradoption,extensionsupport,timelyinput barriersforresource-constrainedfarms. availabilityandlocalimplementation[40]. The climate-smart horticulture literature identifies Precisionirrigationshouldthereforeincreasinglybe highinvestmentrequirementsfordripirrigationand integratedwithprecisionnutrientmanagement,crop precisiontoolsasimportantconstraintstoadoption protectionandclimate-adaptivemanagement. [44].
Biofortificationresearchalsohighlightstheroleofagro- Farmer adoption is more likely when technologies nomicnutrientmanagement,microbialinterventions demonstrate: andprecision-agriculturetoolsinimprovingnutrient • measurableirrigation-watersavings; uptakeandnutritionalquality[42]. Thisprovidesan • stableorincreasedcropyield; importantreminderthatirrigationinfluencesnotonly • lowoperatingcomplexity; yieldbutalsonutrienttransportandproducequality. • cleareconomicreturns; • reliabletechnicalsupport; 13. Additional Horticultural Interfaces Rele- • compatibilitywithexistingpumpsandirrigation vanttoSmartWaterManagement systems.
Someoftheattachedarticlesaddressproductionpro- Thebananafieldstudydemonstrateshowdemonstra- cessesthatarenotdirectlyirrigation-focused. They tionsandfarmer–scientistinteractioncanaccelerate neverthelessillustratewhyirrigationmustultimately adoptionofimprovedresource-managementtechnol- beincorporatedintowhole-systemhorticulturalman- ogy[32]. Themarigoldstudysimilarlydemonstrates agementratherthantreatedindependently. thatextensiongapscanremainsubstantialevenwhen theunderlyingtechnologyisagronomicallyfeasible Micropropagation and in vitro technologies enable [40].
rapid production of uniform planting material [41]. Uniform crop establishment can improve Smartirrigationdevelopmentshouldthereforeinclude the predictability of crop water demand, although participatorytrials,farmertrainingandeconomicanal- the uploaded article itself does not test irrigation ysisfromthebeginningratherthanaftertechnicalde- scheduling.
15.ResearchGapsandFuturePriorities valueifinstallationcost,connectivityrequirementsor operationalcomplexitypreventpracticaluse. First,irrigationstudiesrequirestandardizedreporting. Soiltexture,rootingdepth,sensorplacement,calibra- tion,irrigationthreshold,ETmethodology,cropstage, 16.Conclusion weather,rainfall,irrigationamountandfinalwaterpro- Smart irrigation scheduling represents a transition ductivityshouldallbedocumented.
from fixed irrigation calendars toward dynamic Second,researchshouldmovefromsinglesensorsto- managementofcropwaterdemand.Itsfundamental wardsensorfusion.Soilmoisture,canopytemperature, advantageisnotautomationalonebuttheabilityto VPDandweatherforecaststogetherprovideamorere- matchirrigationtimingandquantitywithactualsoil, liableassessmentofirrigationneedthananyvariable cropandatmosphericconditions.
Soil-moisture sensors provide information on root- Third,AImodelsrequiregreaterfieldvalidation.Algo- zone water supply, ET-based methods quantify rithmstrainedinonecrop,soilorclimatemayperform atmospheric demand, plant measurements reveal poorlywhentransferredtoanotherenvironment. physiological stress, and weather forecasts provide Fourth,smartirrigationshouldbecomemorepredic- predictiveinformation. IoTplatformsconnectthese tive.Futuresystemsshouldanticipatesoil-waterdeple- datastreams,whileDSS,artificialintelligenceandma- tion,rainfallandheatstressratherthanwaitinguntila chinelearningconvertthemintoirrigationdecisions thresholdhasalreadybeenexceeded.
Fifth,crop-stagesensitivityshouldbeincorporatedex- Under water-limited conditions, this integration plicitlyintoirrigationalgorithms. Asinglefixedsoil- becomesparticularlyvaluablebecauselimitedwater moisturethresholdthroughouttheseasonignoresma- must be allocated to the periods when it generates jordifferencesbetweenvegetativeandreproductive thegreatestcropbenefit.Moderateandstrategically waterrequirements.
timedwaterdeficitsmayincreasewaterproductivity, Sixth,genotype-specificschedulingrequiresinvestiga- whereas severe or poorly timed stress can sharply tion.Climate-resilientbreedingandhigh-throughput reduce photosynthesis, reproductive development phenotyping are generating cultivars with different andfinalyield[19–23,33].
rootsystems,stomatalbehaviourandstresstolerance [43,46].Theirirrigationthresholdsmaythereforediffer. Cropresponsealsodependsongenotype,rootarchi- tecture,hormonalregulation,rhizospherebiologyand Seventh,microbiome–irrigationinteractionsdeserve associated microorganisms [35,36,38,39,43,46–48]. greaterattention. Soildryingandrewettingaltermi- Smart scheduling should therefore evolve toward crobialcommunities,whilebeneficialmicroorganisms biologicallyinformedthresholdsratherthantreating influencerootwateruptake,osmolytesandplantstress cropsasuniformwater-demandsystems.
Eighth, irrigation research should include salinity. Future systems are likely to combine continuous Repeated irrigation in poorly drained systems can soil monitoring, thermal and multispectral sensing, increasesaltaccumulation,andwater-savingregimes weatherforecasts,cropsimulation,AIpredictionand mayfurtheraltersaltdistribution[48].
automatedvariable-rateirrigation.Suchclosed-loop Ninth,remotesensingshouldincreasinglydistinguish irrigationwillbecapableofdetectingemergingwater water stress from disease, nutrient deficiency and stress, predictingfuturecropdemandandapplying heat stress. Machine-learning and multispectral waterpreciselybeforestressbecomesyieldlimiting. approachesprovideafoundationforsuchmultimodal Thebroaderobjectiveshouldnotbetomaximizewa- classification[37].
tersavingatanycost.Instead,smartirrigationshould Tenth,environmentalevaluationshouldmovebeyond maximize productive use of scarce water while pro- watersavings. Energyconsumption,pumpingcosts, tectingeconomicallyimportantcropyieldandqual- greenhouse-gasimplicationsandsubstratesustainabil- ity.Integrationofagronomy,cropphysiology,sensors, ityshouldbeincorporatedintoirrigation-systemas- artificialintelligence,climateinformationandfarmer- sessment[49].
centredimplementationoffersapracticalpathwayto- Finally, farmer adoption and economic feasibility wardresilientcropproductionunderintensifyingwater must become central performance indicators. A scarcity.
technicallysophisticatedirrigationcontrollerhaslittle [13] LiakosKG,BusatoP,MoshouD,PearsonS,Bochtis