Observing and modeling land - atmosphere interactions on diurnal to seasonal timescales Craig R. Ferguson 1 ([email protected]) and Joshua K. Roundy 2 ([email protected]) 1 Atmospheric Sciences Research Center, University at Albany, State University of New York, Albany, NY, USA 2 Department of Civil, Environmental, and Architectural Engineering, University of Kansas, Lawrence, KS, USA Overview : Several prior land surface initialization studies using the NASA-Unified Weather Research and Forecasting model (NU- WRF) (e.g., Case et al. 2008; Santanello et al. 2016) and idealized multi-global model ensemble experiments (e.g., Guo et al. 2006; Koster et al. 2010; Koster et al. 2004; Koster et al. 2006; Seneviratne et al. 2013; van den Hurk et al. 2012) have demonstrated the knock-on impact of realistic surface states on short-term to sub-seasonal surface air temperature and precipitation forecasts, respectively. The role of land-atmosphere (L-A) coupling is clearly isolated in these aforementioned modeled cases. However, an open question is: What is the realism of the modeled coupling? Due to observational constraints, models have been poorly, if ever evaluated for coupling realism (Roundy et al. 2014). What is further troubling is that even in so-called coupling ‘hot spots’, the inter-model spread is substantial (i.e., Koster et al. 2004; their Fig. 1 insets). As the climate modeling community moves from atmosphere-ocean global climate models (AOGCMs) to Earth System Models (ESMs) with dynamic vegetation and fully- coupled carbon, energy, and water cycles, realistic L-A coupling becomes of central importance. Objectives and Approaches : We are actively working with counterparts on the Global Land-Atmosphere System Study (GLASS) and Global Hydroclimatology (GHP) panels of the Global Energy and Water Exchanges Project (GEWEX) to meet the following (2) objectives: (1) Evaluate the realism of coupling in models and diagnose systematic biases. Approach : (a) Develop a multi-metric L-A coupling verification and diagnostic toolkit tailored for application with, for example, DOE Atmospheric Radiation Measurement Southern Great Plains (ARM- SGP), New York State Mesonet, or Coupled Model Intercomparison Project Phase 6 (CMIP-6) data. (b) Facilitate greater application (assimilation) of satellite remote sensing data (e.g., Soil Moisture Active Passive, Moderate Resolution Imaging Spectroradiometer) in model evaluation (simulations). (c) Perform process-oriented model sensitivity studies, especially in the context of circulation anomalies and land use/land cover change. (2) Improve our ability to measure L-A coupling across major climate zones at multiple scales. Approach : (a) Fill current L-A coupling related observing system gaps with ground station supplements (e.g., remote sensing atmospheric profilers at FLUXNET sites). (b) Conduct targeted seasonal (e.g., Ferguson et al. 2014) and multi-year field campaigns (e.g., GEWEX North American Regional Hydroclimate Project). (c) Secure enhanced spacebourne monitoring capabilities (e.g., Hyperspectral Environmental Sounder for: water vapor and temperature profiles in the lower troposphere, height of the PBL, and PBL entrainment, as a next Decadal Mission) In the figures that follow we present a few highlights from our most recent activities. Fig. 2(a) New York State Mesonet (http ://www.nysmesonet.org), to be completed by early 2017. Enhanced sites will have the full standard meteorological sensor suite, which includes Stevens HydraProbes at 5-, 25- and 50-cm depths, as well as: 4-component radiation, sonic anemometer, gas analyzer, ground heat flux, sun photometer (Q.Min; U-Albany), microwave radiometer (Radiometrics MP-3000A), and lidar (Leosphere Windcube 100s). (b) Skew-T diagram for Albany, NY, USA @ 00Z (1930 LT) 27 May 2016; deep PBL enabled by strong sensible heat flux and weak subsidence. Potential Temperature (K) Findell and Eltahir (2003) Frye and Mote (2010) Fig. 1 (a) DOE sponsored an Atmospheric Radiation Measurement (ARM) Enhanced Soundings for Local Coupling Studies Field Campaign (ESLCS; Ferguson et al. 2014), which took place at the ARM-SGP Central Facility from June 15 to August 31, 2015. During ESLCS, routine 4 times daily radiosonde measurements at the ARM-SGP CF were augmented on 12 days (June 18 and 29; July 11, 14, 19, and 26; August 15, 16, 21, 25, 26, and 27) with daytime 1-hourly radiosondes and 10-minute ‘trailer’ radiosondes every 3 hours. (b) 16 August 2015 PBL height estimates from a theta-v (potential temperature) gradient approach (black line) as well as other approaches included in the ARM PBLh Value-Added Product (Sivaraman et al. 2013). (c) 16 August 2015 potential temperature profiles for all ESLCS launches. (a) (b) (c) (a) (b) (a) (b) Fig. 3 (a) Frequency distribution of daily differences between NLDAS-2 afternoon (1400–1900 CST) and nonafternoon (0600–1300 CST Day 0 and from 2000 CST Day 0 to 0500 CST Day 1) P computed for the period of 1979–2014 (gray line; right y axis). (b) Diurnal cycle of NLDAS-2 P (mm/hr) over the SGP domain as a function of coupling event persistence and type (i.e., wet or dry coupling). The warm-season climatological (i.e., all days; sample size n=5424) diurnal P cycle is provided for reference (black line). Whiskers denote 95% confidence intervals. The underlying sample size (days) and the fractional composition of those days with AP precipitation is notated above. Taken from Song et al. (2016; their Figs. 4 and 5). Wet-coupling Dry-coupling Climatological frequency Fig. 4 (left) NARR-derived regime composites for the 24 h period preceding CTP-HI event classification for the SGP domain (0600 CST Day -1 to 0500 CST Day 0). Hashing represents difference from climatology significant at 95% CI. Taken from Song et al. (2016; their Fig. 2). Fig. 5 (right) Afternoon Peak (AP) precipitation probabilities across the distribution of identified explanatory variables. Taken from Song et al. (2016; their Fig. 7).