Using MODE to explore the spatial and temporal characteristics of cloud cover forecasts from high‐resolution NWP models

Abstract

Cloud forecasts pose a particularly difficult challenge to verification through the three-dimensional structure of cloud. Two-dimensional total cloud forecasts from one convection-permitting and one near-convection-resolving Numerical Weather Prediction (NWP) model are compared to a cloud analysis using the Method for Object-based Diagnostic Evaluation (MODE) and MODE Time Domain (TD) which is currently being developed to also consider the time dimension. This study attempts to quantify the spatial and temporal evolution of specifically persistent (static) and systematic cloud scenarios where the cloud boundaries change only slowly. These scenarios can lead to significant cloud and temperature errors if the locations of these boundaries are not well forecast. MODE and MODE TD are used to identify and compare the number of cloud free objects, their evolution over time in terms of area or volume. The results show that including the time dimension provides a different perspective to the assessment, but that the choice of method-specific parameters is critical for enabling the matching of time objects. Crucially, model-relative behaviour can be isolated, and the specific characteristics of the cloud analyses are also quantifiable, and need to be accounted for when interpreting results.

Progress and challenges in forecast verification

“…The features-based methods have also continued to increase in popularity. The two most mature methods, namely the Contiguous Rain Area (CRA; Ebert and McBride, 2000) and the Method for Object-based Diagnostic Evaluation (MODE; Davis et al ., 2006) are most often used for precipitation evaluation (e.g., Gallus, 2010), although they are now being applied in other contexts for evaluating climate features (e.g., Moise and Delage, 2011;Rikus et al ., 2011), atmospheric rivers (Clark et al ., 2011) and cloud structures (Kucera et al ., 2011;Mittermaier and Bullock, 2013).…”

Macroscopic cloud properties in the WRF NWP model: An assessment using sky camera and ceilometer data

“…Gilleland et al [2009] presented a review of spatial forecast verification methods. More recently, Mittermaier and Bullock [2013] have proposed a specific methodology to cope with this problem when verifying cloud forecasts.…”

Simulating North American mesoscale convective systems with a convection-permitting climate model

“…MODE has been used successfully to detect biases in short-term weather forecasts (e.g., Davis et al 2006Davis et al , 2009Clark et al 2012;Mittermaier and Bullock 2013). It was reported that coarse-resolution forecast models produce too many large rain areas and underestimate rainfall intensities (e.g., Davis et al 2006;Wernli et al 2008).…”