Error Sources Affecting Variable Rate Application of Nitrogen Fertilizer

The accuracy and interaction among global positioning system (GPS) horizontal accuracy, differential GPS (DGPS) sampling frequencies and machine delay times of a hypothetical variable rate applicator for nitrogen (N) fertilizer application based on an application map in Florida citrus were studied. Parameters studied included: five GPS horizontal accuracy levels, two levels of DGPS horizontal accuracy, two DGPS sampling frequencies and two machine delay times. Two integrated models were developed which documented the effects of the parameters. Machine delay time was the most important factor and GPS horizontal accuracy was the second most important.     

Error Sources on Yield-Based Fertilizer Variable Rate Application Maps

Spatially-variable application of inputs to agricultural fields is dependent upon accuracy of the input variables. The effects of boundary determination, interpolation method, and GPS location errors were studied for determining a variable rate nitrogen application map based upon yield maps of a 3.6 ha Florida citrus block. A general linear model for the mean absolute error approximated the error effects.     

Where do you think you’re going? Accounting for ontogenetic and climate‐induced movement in spatially stratified integrated population assessment models

Understanding spatial population structure and biocomplexity is critical for determining a species’ resilience to environmental and anthropogenic perturbations. However, integrated population models (IPMs) used to develop management advice for harvested populations have been slow to incorporate spatial dynamics. Therefore, limited research has been devoted to understanding the reliability of movement parameter estimation in spatial population models, especially for spatially dynamic marine fish populations. We implemented a spatial simulation–estimation framework that emulated a generic marine fish metapopulation to explore the impact of ontogenetic movement and climate‐induced distributional shifts between two populations. The robustness of spatially stratified IPMs was explored across a range of movement parametrizations, including ignoring connectivity or estimating movement with various levels of complexity. Ignoring connectivity was detrimental to accurate estimation of population‐specific biomass, while implementing spatial IPMs with intermediate levels of complexity (e.g. estimating movement in two‐year and two‐age blocks) performed best when no a priori information about underlying movement was available. One‐way distributional shifts mimicking climate‐induced poleward migrations presented the greatest estimation difficulties, but the incorporation of auxiliary information on connectivity (e.g. tag‐recapture data) reduced bias. The continued development of spatially stratified modelling approaches should allow harvested resources to be better utilized without increased risk. Additionally, expanded collection and incorporation of unique spatially explicit data will enhance the robustness of IPMs in the future.     

Moving integrated weed management from low level to a truly integrated and highly specific weed management system using advanced technologies

Integrated weed management (IWM) is one of the most commonly referred to approaches for sustainable and effective weed control in agriculture, yet it is not widely practiced, likely because current IWM systems fail to meet performance expectations of growers. The effectiveness and value of IWM systems should increase with increasing application specificity and true integration made possible with contemporary advances in technology, information systems and decision support. IWM systems can be classified based on their degree of application specificity and level of integration of tactics. In the application specificity pathway, a tactic is applied at a range of scales, from subfield to plant specific. In the integration pathway, multiple weed control tactics are combined in a synergistic manner. We hypothesise that the full value of IWM can and will be realised only when current and emerging technological innovations, information systems and decision tools are synergistically combined for use in real time. The True IWM system we envision requires automation and robotic technologies, coupled with information and decision support systems that are available or emerging but not yet enabled, in a proven integrated platform. Examples of low‐level, traditional and precision IWM systems are discussed, and research needs for a True IWM system are presented. We conclude that the immediate call should be for a long‐term investment in R&D and education (both theoretical and empirical) to develop and implement True IWM systems, an effort best accomplished in a public–private partnership where all essential entities are fully engaged and adequately resourced, including growers from all countries who will utilise IWM.     

Design of a hyperspectral nitrogen sensing system for orange leaves

The orange (Citrus sinensis) is one of the most important agricultural crops in Florida. Heavy reliance on agricultural chemicals and low fertilizer use efficiencies in citrus production have raised environmental and economic concerns. In this study, a nitrogen sensor was developed to predict nitrogen concentrations in orange leaves. Four design criteria were chosen to maximize the sensing efficiency and reliability. They were: (1) coverage of the spectral N sensing range, (2) no moving parts, (3) single leaf detection, and (4) diffuse reflectance measurement. Based on chlorophyll and protein spectral absorption bands, the sensor's wavelength ranges were chosen to be 620–950 nm and 1400–2500 nm. A reflectance housing was designed to block environmental noise and to ensure single leaf measurement. A halogen light source, two detector arrays, two linear variable filters, and data acquisition cards with 16-bit analog-to-digital converters were used to collect data. The designed N sensor had a spectral resolution less than 30 nm. Test results showed that the nitrogen sensor had good linearity (r > 0.99) and stability. With averaged signal-to-noise ratio (SNR) of 299, the system was able to predict N content with a root mean square difference (RMSD) of l.69 g kg⁻¹ for the validation data set. Using the N sensor, unknown leaf samples could be classified into low, medium and high N levels with 70% accuracy.     

Macroscale drivers of Atlantic and Gulf Menhaden growth

The identification of anthropogenic and environmental drivers on length‐at‐age of fish stocks is important to understanding ecosystem dynamics and harvest intensity. We evaluated coastwide annual growth of n = 187,115 Atlantic Menhaden (Brevoortia tyrannus) and n = 299,185 Gulf Menhaden (B. patronus), using samples collected from the North, Mid‐, and South Atlantic from 1961 to 2016 and across the Gulf of Mexico from 1977 to 2016. Using hierarchical models of age 1 growth and age 2 growth, we evaluated a suite of candidate predictors including fishery landings, easterly (U) and northerly (V) wind velocity, river discharge, juvenile abundance, and the Atlantic Multi‐decadal Oscillation (AMO). We found age 2 growth rates were smaller than age 1 growth rates for both species and that Atlantic Menhaden growth rates were 3–4 times greater than Gulf Menhaden. Age 1 growth rate of Atlantic Menhaden was positively affected by landings lagged by one year, indicating a density‐dependent mechanism. In addition, AMO (negative effect), and wind U (positive effect) and wind V (negative effect) in the North Atlantic region were significant factors influencing coastwide age 1 Menhaden growth. Wind V (negative effect) and AMO (positive effect) influenced age 1 Gulf Menhaden growth. No environmental factors were found to have an effect on age 2 Atlantic Menhaden growth, and AMO was the only significant predictor (weak negative effect) of age 2 Gulf Menhaden growth. Fishing pressure was the primary influence on age 1 Atlantic Menhaden growth, whereas age 1 Gulf Menhaden growth was primarily influenced by environmental conditions.     

Ecosystem‐based forecasts of recruitment in two menhaden species

Gulf (Brevoortia patronus, Clupeidae) and Atlantic menhaden (Brevoortia tyrannus, Clupeidae) support large fisheries that have shown substantial variability over several decades, in part, due to dependence on annual recruitment. Nevertheless, traditional stock–recruitment relationships lack predictive power for these stocks. Current management of Atlantic menhaden explicitly treats recruitment as a random process. However, traditional methods for understanding recruitment variability carry the very specific hypothesis that the effect of adult biomass on subsequent recruitment occurs independently of other ecosystem factors such as food availability and predation. Here, we evaluate the predictability of menhaden recruitment using a model‐free approach that is not restricted by these strong assumptions. We find that menhaden recruitment is predictable, but only when allowing for interdependence of stock with other ecological factors. Moreover, while the analysis confirms the presence of environmental effects, the environment alone does not readily account for the complexity of menhaden recruitment dynamics. The findings set the stage for revisiting recruitment prediction in management and serve as an instructive example in the ongoing debate about how to best treat and understand recruitment variability across species and fisheries.