Performance of analytical footprint models in heterogeneous landscapes under varying atmospheric stability conditions

¹ Department of Civil Engineering, Indian Institute of Technology Hyderabad, India
² School of Science, Swinburne University of Technology, Melbourne, Australia

^* Corresponding author: shwetasingh7892@gmail.com

Abstract

Analytical footprint models that simulate the source area of scalar fluxes generally include a fundamental assumption that the fluxes originate from a horizontal, homogeneous surface. It is widely understood that this assumption is often violated in flux studies, especially for sites where there are significant variations in topography, leaf area, photosynthetic pathway and underlying soil properties. An accurate interpretation of the measured flux footprint under heterogeneous canopy condition can help alleviate the problem. We evaluated the performance of analytical models (Hsieh, K&M, and Schuepp) under stable and unstable atmosheric conditions for the homoeneous canopy (Cotton- C3, zm = 3m and Sugarcane- C4, zm = 4m) and heterogeneous canopy (mixed fetch) compared to FFP model in a complex sugarcane-cotton (C3-C4) cropping system. Performance of models were evaluated using a set of three eddy covariance (EC) towers (one each capturing homogenous C3 and C4 fluxes, and a third capturing heterogeneous, mixed (C3-C4) fluxes at zm = 8m). High-quality EC fluxes that fulfil stationarity and internal turbulence tests were analyzed on the basis of daytime, unstable condition datasets. K&M model (Corr >0.75 , RMSE <0.06 , SD <0.006) performed the best in comparison to FFP model flux footprint prediction under unstable atmospheric condition in heterogeneous canopy condition with respect to Hsieh (Corr <0.6, RMSE >0.01 , SD >0.005), and Schuepp analytical model (Corr =0.2, RMSE <0.01, SD>0.2 ). Unstable atmospheric condition is further classified into four categories (neutral, near neutral unstable, unstable, and very unstable). Relative performance of the analytical models was further analyzed with experimental flux tower generated flux footprint under neutral, near neutral unstable, unstable, and very unstable atmospheric condition. FFP model performs the best in heterogeneous canopy condition under varying neutral to very unstable atmospheric condition. We make clear recommendations for future analysis of fluxes in heterogeneous crop lands under varying atmospheric stability condition.