Finger millet is gaining importance as a food crop with the increasing emphasis on nutritional aspects and drought resilience. However, yield improvement has stagnated. Therefore, popular varieties have been examined for the purpose of whether direct selection for grain yield can be continued or an alternate trait-based approach using the germplasm is necessary. Direct selection for grain yield over the ruling variety, cv. GPU-28 (Germplasm Unit) has not been satisfactory. The path analysis has revealed a high direct effect of mean ear weight on grain yield followed by a moderate direct effect of photosynthetic rate and leaf area index. Furthermore, backward stepwise regression analysis revealed that among the independent traits, the mean ear weight made a significant contribution (60.8%) towards grain yield, followed by the photosynthetic rate (39.2%). The regression equation predicts the inclusion of mean ear weight by 1.0 g extra (as in GE-2672) to cv. GPU-28 will increase grain yield by 4.74%. The trait-specific genotypes are superior to the cv. GPU-28 were GPU-67 (photosynthetic rate) and GE-2672 (mean ear weight) and they could be used as donors for yield improvement. Future selection would aim for genotypes having 70–75 days for flowering with 4−5 productive tillers and mean ear weight of more than 8−9 g/ear. The possible approaches for enhancing grain yield are also discussed.

Predicting in-season crop yield is a unique tool for drawing important crop management decisions for precision farming. Field experiments were conducted at two locations in northwestern India under different agro-climatic zones to predict and validate spring maize yield using various in-season spectral indices. The spectral properties measured with leaf color chart (LCC), chlorophyll meter (SPAD meter), and GreenSeeker optical sensor were used to predict grain yield. A power function based on the Normalized Difference Vegetative Index (NDVI) measured with GreenSeeker optical sensor at V9 growth stage (9th leaf with fully exposed collar) presented higher values of coefficient of determination and explained 61% of the variability in spring maize grain yield, whereas NDVI measured at early and late growth stages were not reliable for the purpose. The spectral properties recorded with the SPAD meter and LCC rendered better grain yield estimates at VT growth stage (tasseling) and were respectively able to explain 75 and 76% variability in grain yield. The developed models were validated on an independent data set from another field experiment on spring maize. The normalized root mean square error (NRMSE) was <10% for LCC and SPAD at all the growth stages and at V9 growth stage for NDVI. The LCC, SPAD, and NDVI values adjusted with cumulative growing degree day were not helpful to improve NRMSE.

To develop a more complete understanding of the ecological factors that regulate crop productivity, we tested the relative predictive power of yield models driven by five predictor variables: wheat and wild oat density, nitrogen and herbicide rate, and growing-season precipitation. Existing data sets were collected and used in a meta-analysis of the ability of at least two predictor variables to explain variations in wheat yield. Yield responses were asymptotic with increasing crop and weed density; however, asymptotic trends were lacking as herbicide and fertilizer levels were increased. Based on the independent field data, the three best-fitting models (in order) from the candidate set of models were a multiple regression equation that included all five predictor variables (R2 = 0.71), a double-hyperbolic equation including three input predictor variables (R2 = 0.63), and a nonlinear model including all five predictor variables (R2 = 0.56). The double-hyperbolic, three-predictor model, which did not include herbicide and fertilizer influence on yield, performed slightly better than the five-variable nonlinear model including these predictors, illustrating the large amount of variation in wheat yield and the lack of concrete knowledge upon which farmers base their fertilizer and herbicide management decisions, especially when weed infestation causes competition for limited nitrogen and water. It was difficult to elucidate the ecological first principles in the noisy field data and to build effective models based on disjointed data sets, where none of the studies measured all five variables. To address this disparity, we conducted a five-variable full-factorial greenhouse experiment. Based on our five-variable greenhouse experiment, the best-fitting model was a new nonlinear equation including all five predictor variables and was shown to fit the greenhouse data better than four previously developed agronomic models with an R2 of 0.66. Development of this mathematical model, through model selection and parameterization with field and greenhouse data, represents the initial step in building a decision support system for site-specific and variable-rate management of herbicide, fertilizer, and crop seeding rate that considers varying levels of available water and weed infestation.