For winter wheat in the US Central Great Plains, phenotypic plasticity of yield is agronomically adaptive, that is, genotypes with higher plasticity have higher yield in high yielding environments with no tradeoff in stressful, low yielding environments. Using data from variety trials conducted between 2000 and 2022 and cultivars released between 1991 and 2022, we explored time trends in phenotypic plasticity and heritability of yield. Given that yield plasticity is agronomically adaptive in the Great Plains, we hypothesize that (i) newer cultivars will have higher yield plasticity; (ii) heritability of yield is declining in the time series; and (iii) genomic regions associated with yield depend on the environment and do not fully overlap with those associated with phenotypic plasticity of yield. Breeding for yield and agronomic adaptation increased phenotypic plasticity of yield at 0.5% year-1; broad sense heritability of yield decreased from 0.23 in 1993 to 0.15 in 2017. Genome-wide association analysis shows genomic regions associated with yield varied between high yielding and stressful environments and were partially independent of those associated with yield plasticity. Newer cultivars have a higher frequency of alleles associated with yield and its plasticity. We discuss implications for breeding and agronomy aimed at improving wheat phenotypes.

Context: Maize (Zea mays L.) yield and resource use efficiency are often constrained by the availability of water and nitrogen (N). Plant density is a key management factor that can modulate the timing and magnitude of resource uptake, potentially mitigating these limitations. However, the extent to which plant density interacts with water and N supply to shape crop performance across variable environments remains unclear with few studies exploring ultra-low plant densities. Objective and methods: This study evaluated the combined effects of plant density, N fertilization, and irrigation on biomass accumulation, N uptake, transpiration, grain yield, and associated resource use efficiencies. Three field experiments were conducted in the humid temperate region of Uruguay across two growing seasons. Weather, as well as N and water availability was recorded through the season. The experimental design consisted in factorial combinations of two water regimes (rainfed vs. irrigated), two N supply levels (0 vs. 170-218 kg N ha(-1)), and four plant densities (3-12 plants m(-2)). Results: These experiments encompassed a wide yield range (3.8-17.1 Mg ha(-1)), providing contrasting environments to assess treatment effects. Under favorable conditions, higher plant densities (>6 plants m(-2)) increased biomass, N uptake, and yield with respect to lower plant densities (<6 plants m(-2)) (10.05-13.53 Mg ha(-1), p < 0.05). In contrast, under water-or N-limited conditions, lower plant densities (<6 plants m(-2)) sustained comparable or higher yields (5.77 Mg ha(-1)) than higher plant densities (5.47-5.99 Mg ha(-1)) presumably by delaying resource depletion and maintaining crop N status. Peaks in transpiration and N uptake occurred earlier than the peak in crop growth, indicating a temporal decoupling that influenced efficiency. Overall, the optimal plant density for maximizing N, water, and radiation use efficiencies varied with environmental conditions. Conclusions: These findings highlight the importance of adjusting plant density according to resource availability. Higher plant densities permit maximizing use efficiencies and yield under non-limiting conditions, while low plant densities enable to achieve both yield stability and efficient resource use in restrictive maize systems.

This paper describes the data set that was used to test the accuracy of twenty-nine crop models in simulating the effect of changing sowing dates and sowing densities on wheat productivity for a high-yielding environment in New Zealand. The data includes one winter wheat cultivar (Wakanui) grown during six consecutive years, from 2012-2013 to 2017-2018, at two farms located in Leeston and Wakanui in Canterbury, New Zealand. The simulations were carried out in the framework of the Agricultural Model Intercomparison and Improvement Project for wheat (AgMIP-Wheat). Data include local daily weather data, soil profile characteristics and initial conditions, crop measurements at maturity (grain, stem, chaff and leaf dry weight, ear number and grain number, grain unit dry weight), and at stem elongation and anthesis (total above ground dry biomass, leaf number per stem and leaf area index). Several in-season measurements of the normalized difference vegetation index (NDVI) and the fraction of intercepted photosynthetically active radiation (FIPAR) are also available. The crop model simulations include both daily in-season and end-of-season results from twenty-nine wheat models.

Digital agriculture (DA) can contribute solutions to meet an increase in healthy, nutritious, and affordable food demands in an efficient and sustainable way. South America (SA) is one of the main grain and protein producers in the world but the status of DA in the region is unknown. A systematic review and case studies from Brazil, Argentina, Uruguay, and Chile were conducted to address the following objectives: (1) quantify adoption of existing DA technologies, (2) identify limitations for DA adoption; and (3) summarise existing metrics to benchmark DA benefits. Level of DA adoption was led by Brazil and Argentina followed by Uruguay and at a slower rate, Chile. GPS guidance systems, mapping tools, mobile apps and remote sensing were the most adopted DA technologies in SA. The most reported limitations to adoption were technology cost, lack of training, limited number of companies providing services, and unclear benefits from DA. Across the case studies, there was no clear definition of DA. To mitigate some of these limitations, our findings suggest the need for a DA educational curriculum that can fulfill the demand for job skills such as data processing, analysis and interpretation. Regional efforts are needed to standardise these metrics. This will allow stakeholders to design targeted initiatives to promote DA towards sustainability of food production in the region.

Normalized difference vegetation index (NDVI) is an essential remote measurement for agricultural studies because of its strong correlation with crop growth and yield. Accurate and comprehensive NDVI forecasts thus provide effective future projections of crop yield for precise agricultural planning and budgeting. Previous recurrent neural network (RNN) based forecasting methodologies have only performed single-pixel or large-area-average NDVI predictions. We present an alternative RNN-based deep-learning architecture, the convolutional long short-term memory (ConvLSTM), to supply much more comprehensive and detailed NDVI forecasts. In this paper, a single ConvLSTM is capable of 10,000-pixel field-level NDVI predictions, providing a more practical methodology for agricultural producers than single-pixel studies. We compare our model to the parametric crop growth model (PCGM), another multipixel field-level NDVI forecasting technique. We test each model over the same set of soybean crop field pixels with the root mean square error (RMSE) metric. The training configuration of each model is defined by the number of seasons of historical data used for weight optimization. When the best training configuration of the model found is used, the ConvLSTM obtains an RMSE of 0.0782, outperforming the PCGM's RMSE of 0.0989 (an improvement of 0.0207 in precision represents a large gain in the accuracy of production volume prediction when projected into large production areas). Finally, by comparing the ConvLSTM predictions with the ground truth data over the entire target region rather than just the soybean crop pixels, we discover that the ConvLSTM can also predict NDVI values over the nonsoybean crop as effectively.

Grain production must increase by 60% in the next four decades to keep up with the expected population growth and food demand. A significant part of this increase must come from the improvement of staple crop grain yield potential. Crop growth simulation models combined with field experiments and crop physiology are powerful tools to quantify the impact of traits and trait combinations on grain yield potential which helps to guide breeding towards the most effective traits and trait combinations for future wheat crosses. The dataset reported here was created to analyze the value of physiological traits identified by the International Wheat Yield Partnership (IWYP) to improve wheat potential in high-yielding environments. This dataset consists of 11 growing seasons at three high-yielding locations in Buenos Aires (Argentina), Ciudad Obregon (Mexico), and Valdivia (Chile) with the spring wheat cultivar Bacanora and a high-yielding genotype selected from a doubled haploid (DH) population developed from the cross between the Bacanora and Weebil cultivars from the International Maize and Wheat Improvement Center (CIMMYT). This dataset was used in the Agricultural Model Intercomparison and Improvement Project (AgMIP) Wheat Phase 4 to evaluate crop model performance when simulating high-yielding physiological traits and to determine the potential production of wheat using an ensemble of 29 wheat crop models. The field trials were managed for non-stress conditions with full irrigation, fertilizer application, and without biotic stress. Data include local daily weather, soil characteristics and initial soil conditions, cultivar information, and crop measurements (anthesis and maturity dates, total above-ground biomass, final grain yield, yield components, and photosynthetically active radiation interception). Simulations include both daily in-season and end-of-season results for 25 crop variables simulated by 29 wheat crop models.

Methods for accurately estimating within-field yield are essential to improve site-specific crop management and resource use efficiencies, which would be a major step toward sustainable intensification of agricultural systems. We set out to assess the accuracy of within-field soybean yields predicted by two data assimilation methods and to assess these methods' assimilation efficiency (AE). Yields were estimated by assimilating remotely sensed leaf area index (LAI) data from Sentinel-2 into a soybean crop growth model on a pixel basis. The LAI data was integrated into the model by Ensemble Kalman Filtering (EnKF) or by recalibrating with the Subplex algorithm (recalibration-based). An open-loop setting which only integrates information on the soil layers was used as a baseline scenario for quantifying the AE. We assessed both data assimilation techniques on eight fields (3067 pixels) in the Corn Belt region (Nebraska, Kansas and Kentucky) in the United States. The data set encompassed substantial variation in crop growth conditions: three growing seasons (2018, 2019 and 2020), rainfed and irrigated fields, and early and late planting dates. Ground truth yield acquired from combine monitors was used to validate the yield estimations. Agreement between predicted and observed yield at pixel level was two times higher for both data assimilation methods compared to the open-loop. The root mean square error (RMSE) was 476 kg.ha(-1) (RRMSE of 10 %) in the recalibration-based method and 573 kg.ha(-1) (RRMSE of 12 %) in the EnKF-based method. For both data assimilation methods, assimilating the LAI improved predictions for 68 % of the pixels. For a further 12 % of pixels, there was no accuracy improvement. For the remaining 20 %, AE was positive for one of the two assimilation methods. The high proportion of pixels with positive AE indicates the potential for overcoming the limitations in applying crop models at high spatial resolution by integrating a crop growth indicator. Assimilating an in-season indicator of crop growth (LAI) into a soybean model made it possible to adjust the simulation pathway, thereby greatly improving the accuracy of the yield estimations at the pixel level. This study elucidates the practical applications of data assimilation strategies for fine-scale within-field crop yield mapping.

Wheat is the most widely grown food crop, with 761 Mt produced globally in 2020. To meet the expected grain demand by mid-century, wheat breeding strategies must continue to improve upon yield-advancing physiological traits, regardless of climate change impacts. Here, the best performing doubled haploid (DH) crosses with an increased canopy photosynthesis from wheat field experiments in the literature were extrapolated to the global scale with a multi-model ensemble of process-based wheat crop models to estimate global wheat production. The DH field experiments were also used to determine a quantitative relationship between wheat production and solar radiation to estimate genetic yield potential. The multi-model ensemble projected a global annual wheat production of 1050 ± 145 Mt due to the improved canopy photosynthesis, a 37% increase, without expanding cropping area. Achieving this genetic yield potential would meet the lower estimate of the projected grain demand in 2050, albeit with considerable challenges.

Current nitrogen (N) fertilization schedule for spring wheat was developed under a dominant crop-pasture rotation. After the year 2002, this cropping system was converted to continuous annual cropping systems under no-till, reducing soil N supply capacity progressively. Additionally, highest grain yield of new varieties increased N demand. The required additional N fertilizer can be adjusted by monitoring nutritional status of the crop. Our objectives were: i) to determine optimal N status at different phenological stages; ii) to quantify the wheat yield gap explained by N supply deficit, and iii) to assess the critical nitrogen nutrition index (NNI) value as a predictor of response to N fertilizer applied at GS 3.3. We adjusted the nitrogen dilution curve (Nc=4.17DM-0.31), deriving a critical NNI at GS 3.3 (NNI=1.24). Depending on soil N supply capacity and NNI at GS 3.3, wheat yield gap attributed to N supply deficit varied from 0 to 2.74 Mg ha-1, averaging 0.76 Mg ha-1. The critical NNI proposed at GS 3.3 was effective to diagnose the N crop demand to reach the attainable yield under different scenarios.

Crop multi-model ensembles (MME) have proven to be effective in increasing the accuracy of simulations in modelling experiments. However, the ability of MME to capture crop responses to changes in sowing dates and densities has not yet been investigated. These management interventions are some of the main levers for adapting cropping systems to climate change. Here, we explore the performance of a MME of 29 wheat crop models to predict the effect of changing sowing dates and rates on yield and yield components, on two sites located in a high-yielding environment in New Zealand. The experiment was conducted for 6 years and provided 50 combinations of sowing date, sowing density and growing season. We show that the MME simulates seasonal growth of wheat well under standard sowing conditions, but fails under early sowing and high sowing rates. The comparison between observed and simulated in-season fraction of intercepted photosynthetically active radiation (FIPAR) for early sown wheat shows that the MME does not capture the decrease of crop above ground biomass during winter months due to senescence. Models need to better account for tiller competition for light, nutrients, and water during vegetative growth, and early tiller senescence and tiller mortality, which are exacerbated by early sowing, high sowing densities, and warmer winter temperatures.

CONTEXT Global climate change is resulting in more frequent and more damaging extreme events affecting the performance of production systems. It is imperative to develop good season-specific crop management recommendations to help farmers to improve their adaptive capacity to a changing climate one season at a time. OBJECTIVE: We aimed to evaluate the skill of the International Research Institute for Climate and Society (IRI) seasonal precipitation forecasts and the interaction between the forecasted seasonal precipitation scenarios and management practices for rainfed soybean cropping systems using a crop simulation model. METHODS: We used a crop simulation model (CROPGRO-Soybean) coupled with weather data to assess the potential use of the IRI seasonal precipitation forecasts as a tool to optimize season-specific management strategies for rainfed soybean in Uruguay. We used a total of 620-668 IRI seasonal precipitation forecasts released from 2003 to 2016 for each of the five weather stations located in the main soybean producing area. The analysis was performed for two soybean cropping systems (i.e., sown as a single crop or as double-cropped soybean), for which we considered combinations of sowing dates and maturity groups (11 sowing dates x 3 maturity groups combinations for each soybean cropping system). RESULTS AND CONCLUSIONS: The IRI seasonal precipitation forecasts were able to successfully forecast belownormal precipitation scenarios in 77% of the total predictions developed for this scenario considering all weather stations during the study period (2003-2016), while it was less accurate in forecasting above-normal precipitation scenarios (60% of success). We found that earlier sowing dates were a better strategy for years when an above-normal precipitation forecast was released for the December-January-February period (4.7 Mg ha(-1) average seed yield). In contrast, delayed sowing dates were more appropriate for below-normal precipitation forecasts (3.7 Mg ha(-1) average seed yield). Applying season-specific management practices farmers could potentially increase their soybean yields by up to 0.6 and 1.6 Mg ha(-1), in years with below- or above-normal forecasted precipitations, respectively. The benefit of season-specific management will depend on the interaction among all management practices, the effective capacity of farmers to implement it, and the risk profile the farmer adopts and it is exposed to. SIGNIFICANCE: Here we built a novel approach to assess the impact of considering seasonal precipitation forecasts for optimizing crop production. This assessment provided insights on how farmers can use seasonal precipitation forecasts to optimize rainfed soybean yield for a specific cropping season.

Accurate within-field yield estimation is an essential step to conduct yield gap analysis and steer crop management towards more efficient use of resources. This study aims to develop and validate a process-based soybean model and to predict within-field yield variability by coupling leaf area index (LAI) retrieval from Sentinel-2 into the crop model. First, a soybean model is presented, which was successfully validated with field observations of total aboveground biomass, LAI and yield from seven contrasting field campaigns with strongly varying conditions. Within-field yield predictions were achieved by combining the model and the observations of LAI through an assimilation strategy. Four model parameters were chosen to optimize against the LAI curve: soil depth, field capacity, initial LAI and nitrogen translocated from leaves to seed. Six fields were used to evaluate the methodology (21175 pixels). The accuracy assessment was conducted on a pixel-by-pixel basis using high density of information from the yield monitor. The overall accuracy quantified by the relative root mean square error (rRMSE) ranged from 28 to 51% (overall rRMSE 35.8%) across the studied fields. The Lee statistics index ranged from 0.61 to 0.71, confirming a high level of similarity between observed and simulated yield maps. Therefore, the methodology was capable of representing the observed spatial patterns of yield. Furthermore, the high consistency of the optimized WHC reflects the value of the assimilation data strategy to spatialize this relevant characteristic. Some challenges were identified for further study to reduce the sources of uncertainty and improve accuracy: i) the inability of the model to reallocate biomass by simulating plant response to source limitation, ii) the generalization of empirical algorithms to retrieve LAI, and iii) the exploration of an updating method as an assimilation strategy to overcome discrepancy between simulated and retrieved LAI.

Evapotranspiration (ET) is a key process in hydrological systems and, consequently, in agroecosystems. It can be measured or derived with a large variety of models at scales ranging from leaf to catchment. MODIS16A2 is a satellite ET product with 500 meters / 8-day spatio-temporal resolution worldwide. It is based on the Penman-Monteith equation and considers the effect of vegetation dynamics, albedo and land cover. This technical paper compares the ET estimated from MODIS16A2 against the ET estimated at different scales from three reference methods: (1) the INIA-GRAS Water Balance on a country-scale, (2) the SWAT model of the Santa Lucia basin on the catchment scale, and (3) the Eddy Covariance Flux located in Colonia on a farmer scale. The analysis shows similarities between MODIS16A2 and the reference methods depending on seasonality, geographic location and scale of ET estimation. The assumptions about vegetation cover, vegetation dynamics, meteorological forcing and soil characteristics of the reference methods compared with MODIS16A2 ones could explain some deviations in the ET estimations. The results of this work contribute with a first approximation towards the quantification of the uncertainty of MODIS16A2 in Uruguay.

Accurate estimation of aerial net primary production (ANPP) using remotely acquired data is one of the main challenges in both environmental monitoring and precision agriculture. Reflectance-based techniques have been widely used for decades, but detection of fluorescence emission by chlorophyll has emerged as a promising alternative in recent years. Although passive sun-induced fluorescence (SIF) monitoring has shown interesting results, the information it provides is limited to few wavelengths (Fraunhofer and telluric lines). On the other hand, active measurements of steady-state fluorescence and its spectral distribution cover the full-emission spectrum but have not been fully explored due to obvious experimental limitations. In this work we develop a novel active fluorescence measurement procedure, based on lamps and sensors mounted on a field tractor. This technique allowed the detection of the full spectrum of fluorescence emission of a plant crop for the first time in the literature. The main objective of this work was to analyze how the information based on reflectance and fluorescence, recorded by the new proposed methodology, tracks the differences caused by different irrigation treatments in the ANPP of three soybean varieties. We observed that reflectance-based vegetation indices showed limited sensitivity to these cumulative differences, as only EVI2, NDWI and SRWI were able to distinguish between rainfed and irrigation treatments in some few cases. Passive, irradiance-normalised SIF showed this same trend, but active fluorescence peak ratio (FRed/FFar-red) revealed statistically significant differences for the three cultivars studied. In addition, the latter showed a significant correlation with ANPP for two soybean varieties after correction for light re-absorption and scattering (p < 0.05, R2 > 0.5), which was observed for only EVI and foliar water status VIs among passive indicators. Active fluorescence measurements at leaf level by PAM fluorometry did not show differences between treatments in the upper part of the canopy but revealed a biomass-dependent decrease in PSII yield along the vertical axis. Our study demonstrated that fluorescence emission spectrum holds highly valuable information that might allow monitoring ANPP changes upon irrigation from remote sensing applications, and therefore should be carefully studied. Lastly, it highlights the potential of SIF retrieval at both O2-A and O2-B lines.

Normalized Difference Vegetation Index (NDVI) is an important remote measurement in agriculture because it has a high correlation with crop growth and yield result. In this paper, we present a methodology to predict the NDVI by training a crop growth model with historical data. Although we use a very simple soybean growth model, the methodology could be extended to other crops and more complex models. The training process is an optimization problem, that is solved using the spectral projected gradient method. The quality of the prediction is measured by computing the Root-Mean-Square Error (RMSE) between predicted and true values, obtaining an error lower than 9%, which improves the results obtained by simple forecast techniques used as baseline estimators.

Early prediction of crop yields has been a challenge frequently resolved through the combination of remote sensing data and crop models. The aim of this study was to evaluate two different methods based on remote sensing data for predicting winter wheat (Triticum aestivum L.) yield at field scale. We compared the accuracy of: (i) a simple regression method between different vegetation indices at anthesis and grain yield, and (ii) a crop model method based on optimization of two parameters (specific leaf nitrogen and initial aboveground-biomass) using time series of vegetation indices. Vegetation indices were derived from Landsat-7 Enhanced Thematic Mapper Plus (ETM +) and Landsat-8 Operational Land Imager (OLI) images acquired for two growing seasons (2013, 2014) across 22 fields in south western Uruguay with an average size of 128 ha. At all sites, leaf area index (LAI) was measured during a field campaign, and grain yield was measured with yield monitors on harvesters. The simple regression method (SRM) achieved higher accuracy than the model-based method (CMM) for the estimation of yield at field scale (RMSE = 966 kg ha(-1) and RMSE = 1532 kg ha(-1), respectively). When deviations between observed and estimated yields were evaluated at pixel (30 x 30 m) level, the model-based method was better at detecting existing spatial variability in grain yield and at identifying areas of different yield potential. Even though both methods have limited utility to estimate yield at field scale with very high accuracy due to large RMSE, the methodologies are suitable to predict harvest volumes at large agricultural areas or at country level, and to construct synthetic yield maps reflecting within field variability. Higher temporal resolution of images would improve accuracy in estimating yield and spatial variability at field scale.

Genotype and environment jointly determine yield and bread-making quality of bread wheat, being possible modulators for these characters. Eleven cultivars were grown at three different nitrogen regimes combining quantity and timing of nitrogen availability for two years; yield, grain protein concentration and Alveograph parameters were determined. Amount and size distribution of proteins were measured using Size Exclusion-High Performance Liquid Chromatography. The results showed year to be the most important determinant of percentage of unextractable polymeric protein in total polymeric protein (%UPP) and yield, while nitrogen regime determined grain protein concentration, the total amount of SDS-extractable and -unextractable proteins, Alveograph W and L. High nitrogen regime generated three types of responses among the cultivars; i) an increase of yield and grain protein concentration, ii) an increase in grain protein concentration but a reduction in %UPP, and iii) an increase in %UPP. For group ii, a low nitrogen regime at early stages but high at later stages was preferential, while for other groups, a high nitrogen regime during the whole season was the most beneficial. This study showed that it is possible to modulate yield and quality in bread wheat by the nitrogen regime, although an understanding of genotype-environmental relationships is necessary.

Estimating total crop water use and the associated variability are critical for planning, for the design of supplementary irrigation strategies and for subsequent management. Tough there have been great advances in estimating actual evapotranspiration, achieving accurate estimates at the field level is still a challenge. The characterization of site specific characteristics influencing evapotranspiration like soil properties impose a great challenge to the use of modeling, and the accurate characterization of the variability within a field of crop status may bias estimates. Using remote sensing to estimate crop status and energy balance at a fine scale overcome some of these issues. Nevertheless there is a need to have accurate and precise measurements of evapotranspiration that can serve as validation sites. The eddy covariance method provides the accuracy and footprint necessary to be used as a reference. Two towers were installed between 2010 and 2015 at two contrasting locations each year in the south-west of Uruguay at agricultural fields with wheat-soybean crop rotation. The sites were maintained at the same location during wheat and soybean. Sites were managed with supplementary irrigation at full demand or without irrigation. Each tower had instruments to measure in parallel the energy balance (radiometers, flux plates and soil temperature probes), and evapotranspiration directly from eddy covariance (sonic anemometer, IRGA). All locations met fetch requirements, were representative of agricultural fields and were situated in an area dominated by agricultural land. Season long totals showed large variability depending on crop status and seasonal precipitation regime that determined crop growth and leaf area development. Totals for irrigated fields in the north (Salto, 650-800mm) were higher than in the south (Colonia-Soriano, 600-750), and higher than standard requirements for the region. This study provides reference values for a significant number of growing conditions and years for irrigated or rain-fed soybean crops.

Tough there have been great advances in estimating actual evapotranspiration, achieving accurate estimates at the field level is still a challenge. The characterization of site specific characteristics influencing evapotranspiration like soil properties impose a great challenge to the use of modeling, and the accurate characterization of the variability within a field of crop status may bias estimates. Using remote sensing to estimate crop status and energy balance at a fine scale overcome some of these issues. Nevertheless there is a need to have accurate and precise measurements of evapotranspiration that can serve as validation sites. The eddy covariance method provides the accuracy and footprint necessary to be used as a reference. Two towers were installed between 2010 and 2015 at two contrasting locations each year in the south-west of Uruguay at agricultural fields with wheat-soybean crop rotation. The sites were maintained at the same location during wheat and soybean, and were moved when other crops were planted at the site. Each tower had instruments to measure in parallel the energy balance (radiometers, flux plates and soil temperature probes), and evapotranspiration directly from eddy covariance (sonic anemometer, IRGA). All locations met fetch requirements, were representative of agricultural fields and were situated in an area dominated by agricultural land. Season long totals showed large variability depending on crop status and seasonal precipitation regime that determined crop growth and leaf area development. Even after full canopy cover was reached (maximum Kc) there was large variation in evapotranspiration reflecting periods of severe stress in some years. This study provides reference values for a significant number of growing conditions and years and highlights the need for considering the variability among years and crops when making estimates of demand for supplemental irrigation. It also provides reference values for methods based on remote sensing of evapotranspiration.