No abstract

Quantifying global carbon and water balances requires accurate estimation of gross primary production (GPP) and evapotranspiration (ET), respectively, across space and time. Models that are based on the theory of light use efficiency (LUE) and water use efficiency (WUE) have emerged as efficient methods for predicting GPP and ET, respectively. Currently, LUE and WUE estimates are obtained from biome-specific look-up tables and coarse resolution remote sensing data with large uncertainties. The major objective of this study was to parameterize eddy covariance tower-based ecosystem LUE (ELUE EC), defined as the ratio of tower-based GPP (GPP EC) to photosynthetically active radiation (PAR), and ecosystem WUE (EWUE EC), defined as the ratio of GPP EC to tower-based ET (ET EC), using the Moderate Resolution Imaging Spectroradiometer (MODIS)-derived enhanced vegetation index (EVI) for predicting maize (Zea mays L.) GPP and ET, respectively. Three adjacent AmeriFlux maize sites with different rotations (continuous maize vs. annual rotation of maize and soybean, Glycine max L.) and water management practices (rainfed vs. irrigated) located near Mead, NE, USA were selected. The EVI tracked the seasonal variations of ELUE EC (R 2 = 0.83) and EWUE EC (R 2 = 0.74) across sites, indicating that EVI can be explicitly used as a measure of ELUE EC and EWUE EC. The predicted GPP (GPP ELUE) using the parameterized ELUE model correlated well with GPP EC (slope =1.0, R 2 = 0.83, and RMSE = 2.85 g C m-2 d-1) and was significantly improved when compared to widely used models that estimate GPP by integrating EVI and climate variables (Greenness and Radiation, Temperature and Greenness, and Vegetation Index) and the standard MOD17 GPP product. Similarly, the predicted ET (ET EWUE) using the parameterized EWUE correlated well with ET EC (slope = 1.02, R 2 = 0.62, and RMSE = 0.83 mm ET-1) and was significantly improved when compared to the standard MOD16 ET product. Preliminary data 3 demonstrate that ELUE and EWUE can be parameterized using EVI, offering new methods for predicting GPP and ET.

This paper investigates the operation and loading of a 5 kW HAWT using the aeroelastic code FAST. Wind data from built environment site at Port Kennedy (PK) and from a flat terrain site in Östergarnsholm (OG), are analysed and compared with IEC 61400-2. The longitudinal turbulence intensity (TI u ) in the PK wind field was 22%; which was higher than the estimated value in IEC 61400-2 Normal Turbulence Model. The TI in the flat terrain (OG) was below 18% for all mean wind speeds. The selected wind conditions from the two locations were used as input in FAST simulation to investigate the performance and loading of the turbine. The elevated turbulence in PK wind fields increased the output rotor power which was more than that predicted by the standard. Similarly, PK wind field also showed higher blade root flapwise bending moment resulting into twice as much damage load on the turbine blades due to large short-term fluctuations in both wind speed and direction. This value for OG was below the standard's prediction. We observe that the current IEC standard seems inadequate for urban siting of SWTs and requires modification for more reliable deployment in turbulent sites.

Abstract. This paper investigates the applicability of the assumed wind fields in International Electrotechnical Commission (IEC) standard 61400 Part 2, the design standard for small wind turbines, for a turbine operating in the built environment, and the effects these wind fields have on the predicted performance of a 5 kW Aerogenesis turbine using detailed aeroelastic models developed in Fatigue Aerodynamics Structures and Turbulence (FAST). Detailed wind measurements were acquired at two built environment sites: from the rooftop of a Bunnings Ltd. warehouse at Port Kennedy (PK) (Perth, Australia) and from the small wind turbine site at the University of Newcastle at Callaghan (Newcastle, Australia). For both sites, IEC 61400-2 underestimates the turbulence intensity for the majority of the measured wind speeds. A detailed aeroelastic model was built in FAST using the assumed wind field from IEC 61400-2 and the measured wind fields from PK and Callaghan as an input to predict key turbine performance parameters. The results of this analysis show a modest increase in the predicted mean power for the higher turbulence regimes of PK and Callaghan as well as higher variation in output power. Predicted mean rotor thrust and blade flapwise loading showed a minor increase due to higher turbulence, with mean predicted torque almost identical but with increased variations due to higher turbulence. Damage equivalent loading for the blade flapwise moment was predicted to be 58% and 11% higher for a turbine operating at Callaghan and PK respectively, when compared with IEC 61400-2 wind field. Time series plots for blade flapwise moments and power spectral density plots in the frequency domain show consistently higher blade flapwise bending moments for the Callaghan site with both the sites showing a once-per-revolution response.

No abstract