The variable response of dryland corn yield to soil water content at planting

Farmers in the central Great Plains want to diversify crop rotations from the traditional monoculture system of winter wheat-fallow. Corn (Zea mays L.) could work well as a rotation crop, but inputs are expensive and farmers would like to know the chances of producing a certain yield before investing in seed, fertilizer, herbicides, etc. Information on the yield response of corn to available soil water at planting could help guide the crop choice decision regarding corn. This study was conducted to determine if a predictive relationship exists between dryland corn yield and available soil water at planting time and, if such a relationship exists, to use it to assess the risk in obtaining profitable yields. Yield and soil water data from 10 years of a dryland crop rotation study at Akron, CO were analyzed by linear regression to determine predictive relationships. The yield-soil water content production function was highly variable, with values ranging from 0.0 to 67.3 kg ha⁻¹ per mm of available soil water in the 0 to 1.8 m soil profile at planting. The differences in yield response to soil water were related to the amount and timing of precipitation that fell during the corn growing season. Because dryland corn yield is highly dependent on precipitation during reproductive and grain-filling stages, soil water content at corn planting cannot be used alone to reliably determine whether corn should be planted in a flexible rotational system. The predictive relationships developed in this study indicate that under typical amounts of available soil water at corn planting, profitable corn production under dryland conditions is a risky and speculative activity in the central Great Plains of the United States.     

Comparison of corn yield response to plant water stress caused by salinity and by drought

The effect of water stress on corn yield was studied in a salinity experiment and in a drought experiment. The plant water status was determined by measuring the pre-dawn leaf water potential regularly during the whole growing season and expressed by the water stress day index (WSDI). The yield response of corn did not differ under salinity and drought conditions. The WSDI is a useful indicator for determining crop-response to salinity and drought.     

An analysis of the effects of water stress on leaf growth and yield of three barley cultivars

The main objective of the study was to evaluate the sensitivity of Iraqi local barley cultivar (Black) to soil water deficit as compared to other barley cultivars, namely: CM-72 and Arivat. The local cultivar proved to be susceptible during germination and emergence, but it developed resistance during the vegetative growth and yield formation stages. Growth analysis of individual leaves clearly showed that Black barley produced leaves of long growth duration which could affect the growth processes especially cell division. Thus, the reduction in the number of cells composing those leaves was small in the cultivar Black. Leaf growth rates as soil matric potential decreased, were reduced considerably, although, no clear trends were observed between cultivars. On the other hand, the smaller cell volume of local cultivar may have a role in its ability to tolerate water deficit. Some evidence that the local cultivar is post-flowering resistant is discussed.     

Effect of soil water osmotic potential on growth and water relationships in barley during soil water depletion

Summary Barley plants (Hordeum distichum, L., cv. Zita) grown in a sandy soil in pots were adjusted during a pretreatment period of 5 days to three levels of soil water osmotic potential by percolating 61 of a nutrient solution with additional 0, 22.3 and 44.6 mM KCl. A drying cycle was then started and the plants were harvested when the soil water matric potential had decreased to –1.4 MPa, respectively 6, 7 and 8 days later.No significant differences in dry matter yields, transpiration coefficients and wilting percentages were found between treatments.During the drying cycle leaf water potential ( _l ) decreased concomitantly with decrease in soil water potential ( _s ) with almost constant and similar differences ( _l – _s ) for all treatments despite differences in levels of potentials. The concomitant decrease in leaf osmotic potential () was due partly to dehydration (58%) and partly to increase in leaf solute content (42%) independent of treatment. The part of total osmotic solutes due to K decreased relatively during the drying cycle.Close relationships were found between and _l as functions of relative water content (RWC). Identical curves for the two levels of salt treatment agree with similar concentrations of K, Cl, and ash found for salt treated plants indicating that maximum uptake of macro nutrients may have been reached.During the main part of the drying cycle the turgor potential as function of RWC was higher and decreased less steeply with decreasing RWC in the salt treated than in the non-salt treated plants.In the beginning of the drying cycle additions of KCI lowered the transpiration rates of the salt treated plants resulting in a slower desiccation of the soil and hence an increased growth period. A delay in uptake from a limited soil water supply may be advantageous during intermittent periods of drought.     

Simulation of yield decline as a result of water stress with a robust soil water balance model

The relative yield decline that is expected under specific levels of water stress at different moments in the growing period is estimated by integrating the FAO K_y approach [Doorenbos, J., Kassam, A.H., 1979. Yield response to water. FAO Irrigation and Drainage Paper No. 33. Rome, Italy] in the soil water balance model BUDGET. The water stored in the root zone is determined in the soil water balance model on a daily basis by keeping track of incoming and outgoing water fluxes at its boundary. Given the simulated soil water content in the root zone, the corresponding crop water stress is determined. Subsequently, the yield decline is estimated with the K_y approach. In the K_y approach the relation between water stress in a particular growth stage and the corresponding expected yield is described by a linear function. To account for the effect of water stresses in the various growth stages, the multiplicative, seasonal and minimal approach are integrated in the model. To evaluate the model, the simulated yields for two crops under various levels of water stress in two different environments were compared with observed yields: winter wheat under three different water application levels in the North of Tunisia, and maize in three different farmers’ fields in different years in the South West of Burkina Faso. Simulated crop yields agreed well with observed yields for both locations using the multiplicative approach. The correlation value (R²) between observed and simulated yields ranged from 0.87 to 0.94 with very high modeling efficiencies. The root mean square error values are relatively small and ranged between 7 and 9%. The minimal and seasonal approaches performed significantly less accurately in both of the study areas. Estimation of yields on basis of relative transpiration performed significantly better than estimations on basis of relative evapotranspiration in Burkina Faso. A sensitivity analysis showed that the model is robust and that good estimates can be obtained in both regions even by using indicative values for the required crop and soil parameters. The minimal input requirement, the robustness of the model and its ability to describe the effect on seasonal yield of water stress occurring at particular moments in the growing period, make the model very useful for the design of deficit irrigation strategies. BUDGET is public domain software and hence freely available. An installation disk and manual can be downloaded from the web.     

Crop coefficient, yield response to water stress and water productivity of teff (Eragrostis tef (Zucc.)

In the semi-arid region of Tigray, Northen Ethiopia a two season experiment was conducted to measure evapotranspiration, estimate yield response to water stress and derive the crop coefficient of teff using the single crop coefficient approach with simple, locally made lysimeters and field plots. During the experiment we also estimated the water productivity of teff taking into account long-term rainfall probability scenarios and different levels of farmers’ skills. During the experimental seasons (2008 and 2009), the average potential evapotranspiration of teff ranged from 260 to 317 mm. The total seasonal water requirement of teff was found to lower in contrast to the assumptions of regional agronomists that teff water requirement is comparable to that of wheat and barley (375 mm). The average single crop coefficient values (k_c) for the initial, mid and late season stages of teff were 0.8-1, 0.95-1.1 and 0.4-0.5, respectively. The seasonal yield response to water stress was 1.04, which indicates that teff exhibits a moderately sensitive and linear response to water stress. The results suggest that teff is likely to give significantly higher grain yield when a nearly optimal water supply is provided. The study showed that, in locations where standard equipment is not affordably available, indicative (rough) crop evapotranspiration values can be obtained by using field plots and employing locally made lysimeters. The difference in economic water productivity (EWP) and the crop water productivity (CWP) for teff were assessed under very wet, wet, normal, dry and very dry scenarios. In addition two groups of farmers were evaluated, a moderately (I) and a highly skilled (II) group. The results showed that higher EWP and CWP were obtained under very wet scenario than very dry scenario. There was also a 22% increase in EWP and CWP under group II compared to group I farmers. The increase was due to a 22% reduction in unwanted water losses achieved through use of improved technology and better irrigation skills. Both EWP and CWP can be used to evaluate the pond irrigation water productivity (IWP) for a given climate, crop and soil type, and skill and technology level of the farmer. For special crops like teff extra criteria may be needed in order to properly evaluate the pond irrigation water productivity. During the experimental seasons, a high IWP for teff was attained when about 90% of the optimal water need of the crop was met. IWP can be used as an indicator as how much supplementary irrigation has to be applied in relation to the rainfall and other sources of water supply in order to assure greatest yield from a total area. However, the supplemental irrigation requirement of the crops may vary with season due to seasonal rainfall variability.     

Response of cotton water stress indicators to soil salinity

Summary A field study was conducted on cotton (Gossypium hirsutum L. c.v. Acala SJ-2) to investigate the effects of soil salinity on the responses of stress indices derived from canopy temperature, leaf diffusion resistance and leaf water potential. The four salinity treatments used in this study were obtained by mixtures of aqueduct and well water to provide mean soil water electrical conductivities of 17, 27, 32 and 38 dS/m in the upper 0.6 m of soil profile. The study was conducted on a sandy loam saline-alkali soil in the lower San Joaquin Valley of California on 30 July 1981, when the soil profile was adequately irrigated to remove any interference of soil matric potential on the stress measurements. Measurements of canopy temperature, leaf water potential and leaf diffusion resistance were made hourly throughout the day.Crop water stress index (CWSI) estimates derived from canopy temperature measurements in the least saline treatment had values similar to those found for cotton grown under minimum salinity profiles. Throughout the course of the day the treatments affected CWSI values with the maximum differences occurring in mid-afternoon. Salinity induced differences were also evident in the leaf diffusion resistance and leaf water potential measurements. Vapor pressure deficit was found to indicate the evaporative demand at which cotton could maintain potential water use for the various soil salinity levels studied. At vapor pressure deficits greater than 5 kPa, cotton would appear stressed at in situ soil water electrical conductivities exceeding 15 dS/m. The CWSI was as sensitive to osmotic stress as other, more traditional plant measures, provided a broader spatial resolution and appeared to be a practical tool for assessing osmotic stress occurring within irrigated cotton fields.     

Grapefruit response to variable salinity in irrigation water and soil

Summary Results are reported from a long-term field experiment designed to determine the effect of irrigation water salinity on the yield and water uptake of mature grapefruit trees. Treatments were started in 1970 and consisted of chloride concentrations in the irrigation water of 7.1, 11.4 and 17.1 meq/1 added as NaCl+CaCl₂ at a 1 : 1 weight ratio.For the last four years of the experiment, 1973 to 1976, yield was linearly related to the mean chloride concentration in the soil saturation extract weighted according to the distribution of water uptake with depth and time (Fig. 2, Table 1). There was a 1.45% (1.68 Mg/ha) yield reduction for each 1 meq/1 increase in chloride concentration above a threshold value of 4.5 meq/1. This corresponded to a 13.5% (14.7 Mg/ha) decrease per 1 mmho/cm increase in the electrical conductivity of the soil saturation extract above a threshold value of 1.2 mmho/cm.Total water uptake was reduced as salt concentration in the soil increased (Fig. 3, Table 2). In the high salinity treatment, root concentration in, and water uptake from, the lower portion of the root zone were decreased. The maximum electrical conductivity (ECe) measured at the bottom of the root zone was 7.90 mmho/cm similar to the values of EC, obtained by linear extrapolation to zero yield and also to zero water uptake.Salt accumulation in the soil depended on the quantity and salt concentration of the irrigation water, rainfall, and on the amount of leaching. SAR and the Na⁺ concentration of the soil remained low throughout the experiment (Table 3). No leaf symptoms of either Cl^– or Na⁺ injury were observed. The results indicate an osmotic — rather than a specific ion effect — of salinity on grapefruit yield.Contribution from the Agricultural Research Organization, The Volcani Center, Bet Dagan, Israel. 1977 Series No. 197-E     

Rice growth and yield respond to changes in water depth and salinity stress

Depth of standing water in rice paddy fields is an important agronomic parameter in the management of irrigation-related salinity problems. It was hypothesized that reductions in the yield of rice under salinity stress can be ameliorated by adjusting the water depth. This study was designed to determine the interactive effects of salinity and water depth on seedling establishment and grain yield in rice. Plants were grown in a greenhouse and irrigated with nutrient solutions amended with NaCl and CaCl₂ (5:1 molar concentrations). Treatments were three salt levels with electrical conductivities at 0.9, 3.3 and 6.0 dS m⁻¹ and six water depths at 4, 7, 10, 13, 16 and 20 cm. The effects of both salinity and water depth were significant on plant growth and yield. However, there was no interaction between the effects of salinity and water depth. Reductions in seedling establishment and grain yield with increases of salinity and water depth resulted from a simple combination of the two different stresses on plants. Highly significant negative correlations were identified between water depth and seedling establishment and also between water depth and grain yield when data were combined across salt levels. Generally, plants performed better with respect to seedling establishment and grain yield in shallow water (i.e. <10 cm) than in deep water (i.e. >10 cm). Under salt stress, the effect of water depth was significant for panicle number, but not significant for panicle weight. The loss of grain yield under salt stress with the increases of water depth was mainly due to reduction in fertile tiller number. We suggest that water depth be lowered during the initiation and growth of productive tillers. However, the practice by lowering water depth must be incorporated with appropriate field management such as the increase of irrigation frequency, precision leveling, and effective weed control methods.     

The effect of drought on barley yield and water use in two contrasting years

Summary Results are presented for an experiment in 1979 in which mobile shelters were used to exclude rainfall; responses are compared with those in a hotter and drier year, 1976. The continuous drought treatment achieved a larger maximum deficit in 1979 than in 1976. The relationship between rate of water use as a fraction of the potential evaporation rate and soil water deficit was similar in the two years. Yield reductions due to drought were smaller in 1979, as expected from the smaller evaporation rates and therefore smaller potential soil water deficits. Dry matter production was related to water use, but the relationship differed between years. The difference can be related to the saturation vapour pressure deficit. Components of yield were affected differently in the two years; drought after anthesis decreased yield in 1979 by decreasing grain numbers not grain size.     

Effect of irrigation and nitrogen on yield,yield components and water use efficiency of barley in Saudi Arabia

Water use efficiency and yield of barley were determined in a field experiment using different irrigation waters with and without nitrogen fertilizer on a sandy to loamy sand soil during 1994–1995 and 1995–1996. Depending upon different fertilizer treatments, the overall mean crop yield ranges for two crop seasons were: greenmatter from 19.48–55.0 Mg ha⁻¹ (well water) and 21.92–66.5 Mg ha⁻¹ (aquaculture effluent); drymatter from 6.86–20.69 Mg ha⁻¹ (well water) and 7.87–20.90 Mg ha⁻¹ (aquaculture effluent); biomass from 4.12–21.31 Mg ha⁻¹ (well water) and 8.10–19.94 Mg ha⁻¹ (aquaculture effluent) and grain yield from 2.12–5.50 Mg ha⁻¹ (well water) and 3.25–7.25 Mg ha⁻¹ (aquaculture effluent). The WUE for grain yield was 3.37–8.74 kg ha⁻¹ mm⁻¹ (well water) and 5.17–11.53 kg ha⁻¹ mm⁻¹ (aquaculture effluent). The WUE for total biomass ranged between 6.55–33.88 kg⁻¹ ha⁻¹ mm⁻¹ (well water) and 12.88–31.70 kg ha⁻¹ mm⁻¹ (aquaculture effluent). The WUE for drymatter was 10.91–32.90 kg ha⁻¹ mm⁻¹ (well water) and 12.51–33.22 kg ha⁻¹ mm⁻¹ (aquaculture effluent). It was found that grain yield and WUE obtained in T-4 and T-5 irrigated with well water and receiving 75 and 100% nitrogen requirements were comparable with T-4 and T-5 irrigated with aquaculture effluent and receiving 0 and 25% nitrogen requirements. In conclusion, application of 100 to 150 kg N ha⁻¹ for well water and up to 50 kg N ha⁻¹ for aquaculture effluent irrigation containing 40 Mg N l⁻¹ would be sufficient to obtain optimum grain yield and higher WUE of barley in Saudi Arabia.     

Foliar and yield response of Santa rosa plum to saline water spray

Summary Canopies of 22-year-old Santa Rosa plum trees irrigated with mini-sprinklers below the canopy with nonsaline (0.3 dS/m) water were sprayed weekly during one irrigation season with water having six levels of salinity (0.3, 1.1, 2.1, 3.3, 4.5, and 6.8 dS/m) to evaluate the extent of leaf injury, foliar absorption of Cl and Na, and yield response. Recognizable leaf injury was caused by spray water containing 29 mol/m³ of chloride and 15 mol/m³ of sodium. Severe leaf damage occurred when the leaf chloride and sodium concentrations exceeded 300 and 125 mmol/kg (dry weight), respectively. These concentrations were higher than those causing foliar damage on other trees in the same orchard which had been irrigated below the canopy with water having the same salinity as that sprayed on the canopy. No residual foliar injury was observed during the irrigation season following the year when the spray treatments were applied. Fruit yield measured six weeks after treatments were initiated was unaffected. In the following 2 years, yield was reduced by the highest salinity levels, even though the salt spray treatments were not continued and no foliar injury was visible.     

Different irrigation methods and water stress effects on potato yield and yield components

This research was conducted during the spring seasons of 2000 and 2002 in Hatay province located in the East Mediterranean Region of Turkey. The research investigated the effects of two drip irrigation methods and four different water stress levels on potato yield and yield components. The surface drip (SD) and subsurface drip (SSD) irrigation methods were used. The levels were full irrigation (I₁₀₀), 66% of full irrigation (I₆₆), 33% of full irrigation (I₃₃) and un-irrigated (I₀) treatments. Five and three irrigation were applied in 2000 and 2002 early potato growing seasons, respectively. Total irrigation amount changed from 102 to 302 mm and from 88 to 268 mm in 2000 and 2002, respectively. Seasonal evapotranspiration changed between 226 and 473 mm and 166 and 391 mm in 2000 and 2002, respectively. SD and SSD irrigation methods did not result in a significant difference on yield. However, SD method has more advantages than SSD method, which has difficulties in replacement and higher system cost. Irrigation levels resulted in significant difference in both years on yield and its components. Water stress significantly affected the yield and yield parameters of early potato production. Water deficiency more than 33% of the irrigation requirement could not be suggested.Water use efficiency (WUE) of SD irrigation methods had generally higher values than SSD irrigation methods. Treatment I₃₃ gave maximum irrigation water use efficiency (IWUE) for both years. SSD irrigation method did not provide significant advantage on yield and WUE, compared to SD irrigation in early potato production under experimental conditions. Therefore, the SD irrigation method would be recommended in early potato production under Mediterranean conditions.     

Camelina water use and seed yield response to irrigation scheduling in an arid environment

Camelina sativa (L.) Crantz is a promising, biodiesel-producing oilseed that could potentially be implemented as a low-input alternative crop for production in the arid southwestern USA. However, little is known about camelina’s water use, irrigation management, and agronomic characteristics in this arid environment. Camelina experiments were conducted for 2 years (January to May in 2008 and 2010) in Maricopa, Arizona, to evaluate the effectiveness of previously developed heat unit and remote sensing basal crop coefficient (K _cb ) methods for predicting camelina crop evapotranspiration (ET) and irrigation scheduling. Besides K _cb methods, additional treatment factors included two different irrigation scheduling soil water depletion (SWD) levels (45 and 65 %) and two levels of seasonal N applications within a randomized complete block design with 4 blocks. Soil water content measurements taken in all treatment plots and applied in soil water balance calculations were used to evaluate the predicted ET. The heat-unit K _cb method was updated and validated during the second experiment to predict ET to within 12–13 % of the ET calculated by the soil water balance. The remote sensing K _cb method predicted ET within 7–10 % of the soil water balance. Seasonal ET from the soil water balance was significantly greater for the remote sensing than heat-unit K _cb method and significantly greater for the 45 than 65 % SWD level. However, final seed yield means, which varied from 1,500 to 1,640 kg ha^?1 for treatments, were not significantly different between treatments or years. Seed oil contents averaged 45 % in both years. Seed yield was found to be linearly related to seasonal ET with maximum yield occurring at about 470–490 mm of seasonal ET. Differences in camelina seed yields due to seasonal N applications (69–144 kg N ha^?1 over the 2 years) were not significant. Further investigations are needed to characterize camelina yield response over a wider range of irrigation and N inputs.     

Grain yield,yield components,drought sensitivity and water use efficiency of spring wheat subjected to water stress at various growth stages

Summary The influence of water stress at various growth stages on yield and yield structure of spring wheat (Triticum aestivum, L., cv. Sappo) was investigated using lysimeters in the field, automatically protected from rain by a mobile glass roof. Each drought treatment consisted of a single period without irrigation. Irrigation was resumed when all available soil water (100 mm between field capacity and permanent wilting to a depth of 100 cm) had been used. The drought periods were defined as beginning when relative evapotranspiration decreased below one and ending at reirrigation. The first drought occurred during tillering and jointing and the final one during grain formation.     

时段划分对棉花作物水模型的影响与改进

为了改善时段划分对作物水模型模拟精度的影响,依据山西水利职业技术学院试验基地2006和2008年棉花田间试验资料,将棉花全生育期等间隔地划分为不同时段,用非线性优化方法求得了不同时段数条件下的作物水模型参数,分析研究了模型参数与时段数的关系,据此在作物水模型的水分敏感指数累积函数中引入了时段数,并与现有的作物水模型进行了比较。结果表明,引入时段数的作物水模型模拟产量的相对误差随时段数的增加而减小,当时段数大于11时,相对误差平均值和最大值分别减小到7%和15%以下,与现有的作物水模型比较,模拟精度有所提高,但参数个数未增加。该模型更多地反映了水分胁迫时间对作物产量影响的信息。     

水肥对土壤盐分影响及增产效应

为了内蒙古河套灌区盐渍化土壤的肥料高效利用,采用田间试验的方法,将不同种类肥料和灌溉定额进行组合,研究其对土壤盐分的动态影响及增产效应.结果表明:小麦收获后,除尿素处理外,有机肥、控释肥和缓释肥处理在常规及节水灌溉条件下耕层和剖面土壤电导率(EC)均值较试验初都有不同程度的降低.有机肥处理在常规灌水条件下(灌水定额为1 005 m³/hm²),控盐效果略显优势;缓释、控释肥在节水灌溉条件下(灌水定额为750 m³/hm²),控盐效果更明显,剖面土壤EC均值较试验前分别下降16.4%,14.3%;尿素处理在常规灌水条件下,耕层及剖面土壤EC均值较播前分别增加3.6%,2.7%,积盐程度略高于节水处理.4种肥料处理的小麦产量较对照处理增产效果显著;缓释肥处理在常规及节水灌溉条件下均表现出显著的增产优势.综合考虑节水、增产、土壤脱盐等效应,获得优化灌水施肥模式为:缓释肥配二铵基施,生育期内不进行追肥,缓释肥为800.4 kg/hm²,二铵为350.6 kg/hm²,灌水定额为750 m³/hm²,产量为8 374.5 kg/hm², 较当地农民习惯灌水施肥处理可增产2.14%、节水25%,作物耕层EC值和剖面土壤EC均值较播前分别下降18.6%,16.4%.     

Yield response of sugar beets to water stress under Western European conditions

The average yield of sugar beet has almost doubled within the last 30 years. With the raise in average yields and the increase in sensitivity to water stress of sugar beets, the yield response factor (Ky) derived by Doorenbos and Kassam (1979) needs an update. In this article, the soil water balance model BUDGET (Raes et al., 2006) was calibrated and validated to obtain correct estimations of the evapotranspiration deficit (1 − ETa/ETc, where ETa = actual crop evapotranspiration and ETc = maximum crop evapotranspiration under standard conditions) of sugar beets in two locations in France. Datasets of observed soil water contents of several years and different irrigation treatments were used. The simulated evapotranspiration deficits and observed yields were used to derive a seasonal Ky. The obtained linear and polynomial yield response relation between observed yield decline and evapotranspiration deficit showed a high goodness-of-fit. The coefficient of determination (R²) = 0.83, the Nash-Sutcliffe efficiency (EF) = 0.79, the relative root mean squared error (RRMSE) = 0.26 for linear; the coefficient of determination (R²) = 0.85, the Nash-Sutcliffe efficiency (EF) = 0.79, the relative root mean squared error (RRMSE) = 0.25 for polynomial). The results suggested a more pronounced response of sugar beet to water stress in Europe as compared to the values previously reported by Doorenbos and Kassam (1979). The comparison between the observed and simulated yields (with the updated Ky) for another site in France confirmed the findings.     

Comparison of adaptive techniques to predict crop yield response under varying soil and land management conditions

Decision-making processes in agriculture often require reliable crop response models to assess the impact of specific land management. While process-based models are often preferred over empirical ones in current modelling communities, empirical crop growth models can play an important role in identifying the hidden structure of crop growth processes relating to a wide range of land management options. This study investigates the potential of predicting crop yield responses under varying soil and land management conditions by applying three different adaptive techniques: general linear models (GLMs), artificial neural networks (ANNs), and regression trees (RTs). The crop yield data used in this research consist of 720 maize yield indices from 11 different land management trials in southern Uganda. GLM showed the poorest results in terms of modelling accuracy, prediction accuracy, and model uncertainty, which might suggest its inability to model the non-linear causal relationships present in complex soil–land and crop-management interactions. The other two non-parametric adaptive models show significantly higher prediction accuracy than GLM. RT is the most robust technique for predicting crop yield at the study site. ANN is also a promising tool for predicting crop yield and offers insight into the causal relationships through the use of sensitivity analyses, but the complex parameterization and optimum model structure require further attention. The three adaptive techniques compared in this research showed different advantages and disadvantages. When these methods are used together, valuable information can be provided on crop responses, and more reliable crop growth models may result.