PreprintPDF Available

# Connecting Land-Atmosphere Interactions to Surface Heterogeneity in CHEESEHEAD 2019

Authors:
Preprints and early-stage research may not have been peer reviewed yet.
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
CONNECTING LAND-ATM OS PH ER E INTERACTIONS TO
1
2
3
4
BRIAN J. BUTTERWORTH, ANKUR R. DESAI, STEFAN METZGER, PHILIP A. TOWNSEND, MARK D. SCHWARTZ,
5
GRANT W. PETTY, MATTHIAS MAUDER, HANNES VOGELMANN, CHRISTIAN G. ANDRESEN, TRAVIS J.
6
AUGUSTINE, TIMOTHY H. BERTRAM, WILLIAM O.J. BROWN, MICHAEL BUBAN, PATRICIA CLEARLY, DAVID J.
7
DURDEN, CHRISTOPHER R. FLORIAN, ELICEO RUIZ GUZMAN, TREVOR J. IGLINSKI, ERIC L. KRUGER, KATHLEEN
8
LANTZ, TEMPLE R. LEE, TILDEN P. MEYERS, JAMES K. MINEAU, ERIK R. OLSON, STEVEN P. ONCLEY, SREENATH
9
PALERI, ROSALYN A. PERTZBORN, CLAIRE PETTERSEN, DAVID M. PLUMMER, LAURA RIIHIMAKI, JOSEPH SEDLAR,
10
ELIZABETH N. SMITH, JOHANNES SPEIDEL, PAUL C. STOY, MATTHIAS SÜHRING, JONATHAN E. THOM, DAVID D.
11
TURNER, MICHAEL P. VERMEUEL, TIMOTHY J. WAGNER, ZHIEN WANG, LUISE WANNER, LOREN D. WHITE,
12
JAMES M. WILCZAK, DANIEL B. WRIGHT, TING ZHENG
13
14
15
AFFILIATIONS: BUTTERWORTH, DESAI, METZGER, PETTY, MINEAU, AND PALERI Department of
16
17
TOWNSEND, KRUGER, AND ZHENG Department of Forest and Wildlife Ecology, University of
18
19
Observatory Network Program, Battelle, Boulder, Colorado; SCHWARTZ AND IGLINSKI – Department
20
of Geography, University of Wisconsin-Milwaukee, Milwaukee, Wisconsin; MAUDER, VOGELMANN,
21
SPEIDEL, AND WANNER Institute of Meteorology and Climate Research - Atmospheric
22
Environmental Research, Karlsruhe Institute of Technology, Garmisch-Partenkirchen, Germany;
23
24
AUGUSTINE Class ACT Charter School, Chequamegon School District, Park Falls, WI; BERTRAM AND
25
26
BROWN, ONCLEY National Center for Atmospheric Research, Earth Observing Laboratory,
27
Boulder Colorado; BUBAN AND LEE Cooperative Institute for Mesoscale Meteorological Studies
28
and NOAA Air Resources Laboratory Atmospheric Turbulence and Diffusion Division, Oak Ridge,
29
Tennessee; CLEARLY Department of Chemistry and Biochemistry, University of Wisconsin-Eau
30
Claire, Eau Claire, Wisconsin; LANTZ, RIIHIMAKI, AND SEDLAR – Cooperative Institute for Research in
31
32
33
Turbulence and Diffusion Division, Air Resources Laboratory, National Oceanic and Atmospheric
34
Administration (NOAA), Oak Ridge, Tennessee; OLSON, PETTERSEN, THOM, AND WAGNER – Space
35
36
PERTZBORN – Center for Climate Research, Nelson Institute for Environmental Studies, UW-
37
38
Wyoming, Laramie, Wyoming; GUZMAN Department of Forest Production, University of
39
40
Oceanic and Atmospheric Administration (NOAA), Norman, Oklahoma; STOY – Department of
41
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
42
– Institute of Meteorology and Climatology, Leibniz University of Hannover, Germany; TURNER
43
Global Systems Laboratory, National Oceanic and Atmospheric Administration (NOAA), Boulder,
44
Colorado; WANG Laboratory for Atmospheric and Space Physics and Department of
45
46
Department of Chemistry, Physics, and Atmospheric Science, Jackson State University, Jackson,
47
Mississippi; WILCZAK Physical Sciences Laboratory, National Oceanic and Atmospheric
48
49
50
51
52
CORRESPONDING AUTHOR: Brian J. Butterworth, bbutterworth@wisc.edu
53
54
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
ABSTRACT
55
56
The Chequamegon Heterogeneous Ecosystem Energy-balance Study Enabled by a High-density
57
Extensive Array of Detectors 2019 (CHEESEHEAD19) is an ongoing National Science
58
Foundation project based on an intensive field campaign that occurred from June-October 2019.
59
The purpose of the study is to examine how the atmospheric boundary layer responds to spatial
60
heterogeneity in surface energy fluxes. One of the main objectives is to test whether lack of
61
energy balance closure measured by eddy covariance (EC) towers is related to mesoscale
62
atmospheric processes. Finally, the project evaluates data-driven methods for scaling surface
63
energy fluxes, with the aim to improve model-data comparison and integration.
64
65
To address these questions, an extensive suite of ground, tower, profiling, and airborne
66
instrumentation was deployed over a 10×10 km domain of a heterogeneous forest ecosystem in
67
the Chequamegon-Nicolet National Forest in northern Wisconsin USA, centered on the existing
68
Park Falls 447-m tower that anchors an Ameriflux/NOAA supersite (US-PFa / WLEF). The
69
project deployed one of the world’s highest-density networks of above-canopy EC measurements
70
of surface energy fluxes. This tower EC network was coupled with spatial measurements of EC
71
fluxes from aircraft, maps of leaf and canopy properties derived from airborne spectroscopy,
72
ground-based measurements of plant productivity, phenology, and physiology, and atmospheric
73
profiles of wind, water vapor, and temperature using radar, sodar, lidar, microwave radiometers,
74
infrared interferometers, and radiosondes. These observations are being used with large eddy
75
simulation and scaling experiments to better understand sub-mesoscale processes and improve
76
formulations of sub-grid scale processes in numerical weather and climate models.
77
78
CAPSULE SUMMARY
79
80
A regional-scale observational experiment designed to address how the atmospheric boundary
81
layer responds to spatial heterogeneity in surface energy fluxes.
82
83
INTRODUCTION
84
85
Land-atmosphere exchanges of energy, water, and carbon influence weather and climate. The
86
biological processes that mediate these exchanges with the atmosphere occur at multiple spatial
87
and temporal scales, necessitating a variety of cross-scale observational platforms. Accurate
88
accounting of land-atmosphere interactions is critical for improving the predictive performance
89
of numerical weather and climate models. Unfortunately, there is a persistent mismatch between
90
the scales of observations and models. This scale mismatch is problematic because natural
91
environments exhibit substantial heterogeneity in their surface characteristics, which means that
92
observations are not always accurate reflections of the entire model grid cell. Furthermore, the
93
atmosphere is strongly influenced by nonlinear two-way interactions with radiation, land cover,
94
and soil, so that the spatial and temporal scaling of surface fluxes is fundamental to assessing the
95
parameterizations used in atmospheric models to represent land-atmospheric interactions.
96
97
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
The notion that land surface heterogeneity influences the surface energy balance, and the
98
resulting atmospheric responses, emerged from early model simulations showing the importance
99
of soil moisture, vegetation, albedo, roughness, and heating on the atmosphere (Garratt 1993;
100
Mahrt 2000; Betts et al. 1996; Charney 1975; Avissar 1995; Pielke et al. 1998). Theories on how
101
land surface variations drive atmospheric boundary layer (ABL) growth vary (e.g., Desai et al.
102
2006; Reen et al. 2014; Platis et al. 2017; Gantner et al. 2017), with no consensus on whether
103
responses scale linearly or non-linearly and whether they differ for dry versus moist dynamics
104
(Raupach and Finnigan 1995). Modeling studies on this topic have been developed from limited
105
sets of observations of prior field experiments and from specialized modeling domains using
106
simplified boundary conditions (e.g., Kang et al., 2007; Hill et al., 2008, 2011; Zhu et al., 2016).
107
From these previous studies, scaling laws have been derived based on numerical simulations
108
(van Heerwaarden et al. 2014; Rihani et al. 2015), but a systematic regional-scale observational
109
experiment that quantifies the multi-scale nature of sub-grid scaling and patterning has never
110
been fully realized (Steinfeld et al. 2007).
111
112
An issue related to how heterogeneity influences transport processes in the ABL is the energy
113
balance closure problem. This refers to an observed tendency in eddy covariance (EC) flux
114
measurements, where the sum of incoming available energy (net radiation [RN] minus ground
115
heat flux [G]) exceeds surface turbulent sensible and latent heat fluxes (HS and HL) over sub-
116
hourly time scales (Foken et al. 2011). Systematic studies have ruled out instrument errors as the
117
primary cause (Twine et al. 2000; Frank et al. 2013; Liu et al. 2011). Incomplete observation of
118
sub-measurement height storage flux accounts for only some of this lack of closure (Leuning et
119
al. 2012; Xu et al. 2018). Advection terms are not expected to have a systematic direction that
120
would always lead to lack of closure (e.g., Aubinet et al. 2010; Barr et al. 2013; Nakai et al.
121
2014; Zitouna-Chebbi et al. 2012), while topography contributes mostly in extreme cases
122
(Mcgloin et al. 2018).
123
124
EC sites with more variable land cover tend to have larger closure imbalances (Stoy et al. 2013;
125
Xu et al. 2017b). One proposed hypothesis for lack of closure in the energy budget is that surface
126
127
(e.g., Charuchittipan et al. 2014; Gao et al. 2016; Foken et al. 2011; Mauder et al. 2007b). An
128
intensive suite of energy flux measurements between surface and atmosphere at the mesoscale
129
(on the order of tens of kilometers) can help address this key uncertainty in land-atmosphere
130
exchange (Xu et al. 2020).
131
132
EXPERIMENTAL GOALS
133
134
CHEESEHEAD19 was designed to provide a new level of observation density and
135
instrumentation reliability to test hypotheses on spatial heterogeneity and atmospheric feedbacks.
136
The two main research objectives for the CHEESEHEAD19 experiment were to 1) investigate
137
causes of energy balance non-closure over heterogeneous ecosystems and 2) to address the
138
problem of scaling surface energy fluxes.
139
140
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
There is currently no definitive answer as to what is responsible for energy balance non-closure.
141
The project was designed specifically to test the hypothesis that heterogeneity is responsible for
142
generating organized (sub-)mesoscale structures that are not resolved by traditional EC methods.
143
144
Various theories suggest that “spatial” EC, where multiple towers are combined to estimate the
145
mesoscale contribution to the total flux, could be used to analyze this contribution and “close”
146
the energy balance (Steinfeld et al. 2007; Mauder et al. 2008b). To calculate spatial fluxes,
147
CHEESEHEAD19 deployed an EC tower network and airborne EC measurements. These
148
measurements provide spatial patterns of surface energy fluxes across various vegetation and
149
surface types in the heterogeneous landscape. Alongside this EC flux network, multiple
150
platforms were deployed to characterize the atmospheric environment by profiling relevant
151
atmospheric characteristics across a range of scales. This allows us to determine the existence
152
and to characterize the nature of organized mesoscale structures. We can investigate the degree
153
to which mesoscale eddies are responsible for energy balance non-closure in EC measurements,
154
and whether land surface energy partitioning and atmospheric responses differ from the sum of
155
their individual components.
156
157
To systematically address surface energy balance variability in the heterogeneous forested
158
landscape, a pre-campaign large eddy simulation (LES) analysis of the study domain was
159
conducted. It was found that, while 12 flux towers would be sufficient to adequately sample land
160
cover variation, >15 flux towers are required to sample mesoscale eddy structures and close the
161
energy budget (a similar result to Steinfeld et al., 2007). Therefore, the CHEESEHEAD19 field
162
campaign deployed 20 flux towers, a marked increase over many previous experiments.
163
164
CHEESEHEAD19 asks how we can optimally observe and simulate the terms of the surface
165
energy balance and the corresponding atmospheric responses to heterogeneous surface forcings.
166
The objective is to evaluate methods for scaling surface energy fluxes, with the aim of improving
167
model-data comparisons. To this end, we conduct LES and machine-learning scaling
168
experiments to simulate sub-mesoscale responses. These will be compared to measured
169
quantities to test existing theoretical concepts and to improve our understanding of how scale-
170
dependent transport processes in the lower atmosphere respond to surface heterogeneity.
171
The dataset collected during this study will help test multiple scaling methodologies across
172
heterogeneous land cover. Specifically, it aims to test the environmental response function -
173
virtual control volume (ERF-VCV) approach (Metzger 2018; Xu et al. 2018), which combines
174
the strengths of both data-driven and mechanistic strategies.
175
176
Several additional research objectives are addressed by using the unique data resources of
177
CHEESEHEAD19. These include a separately funded study to use CO2 fluxes of Integrated
178
Surface Flux System (ISFS) towers and hyperspectral imagery of canopy functional traits to
179
determine the principal drivers of variation in NPP and carbon use efficiency across a broad
180
array of forest ecosystems. Additionally, concurrent measurements of ozone (O3) mixing ratios at
181
30- and 122-m on the tall tower were made using a chemical ionization time-of-flight mass
182
spectrometer (CI-ToFMS; TOFWERK AG and Aerodyne Research Inc.) and a photometric
183
analyzer (Model 49i; Thermo Fisher) to obtain vertical O3 profiles above the forest canopy
184
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
(Bertram et al. 2011; Novak et al. 2020). These measurements were accompanied by flights of a
185
sUAS-mounted lightweight O3 monitor (POM; 2B) that obtained vertical concentration
186
gradients. These measurements are being used to determine the relative contributions of stomatal
187
uptake and other nonstomatal loss pathways to O3 deposition within a mixed forest canopy.
188
189
190
THE EXPERIMENT
191
192
Overview
193
194
CHEESEHEAD19 investigators deployed an extensive suite of ground, tower, profiling, and
195
airborne instrumentation over a 10 × 10 km domain in a forested and aquatic landscape in
196
northern Wisconsin USA (Fig. 1; Table 1), centered on the existing Park Falls 447-m tower
197
Ameriflux/NOAA supersite (US-PFa / WLEF). The main components of the CHEESEHEAD19
198
field campaign were:
199
a) ground-based fluxes and meteorology
200
b) airborne fluxes and meteorology
201
c) atmospheric profiling
202
d) surface environment characterization
203
204
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
205
Fig. 1. Map and schematic diagram of CHEESEHEAD19 domain. Map shows the
206
location of all measurements made during the field campaign. Insets show
207
Lakeland and Prentice airports where SURFRAD (in addition to the one in ISS
208
field), radar wind profilers with RASS, and CLAMPS systems were installed.
209
Schematic diagram shows instrument location and a conceptual model of airborne
210
data collection. (Wisconsin Department of Natural Resources 2019)
211
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
212
The EC tower network consisted of 17 towers from the NSF Lower Atmosphere Observing
213
Facility (LAOF) ISFS, two additional towers, and the central tall Ameriflux tower. Ground-
214
based measurement of vegetation occurred at 41 plots in the domain, plus an additional 10 plots
215
for measuring phenology. Airborne spectroscopy imaging was used to map leaf chemistry and
216
canopy properties.
217
218
The suite of atmospheric profiling instruments included the LAOF Integrated Sounding System
219
(ISS; Fig. 2c) and the UW SPARC system (Fig. 2a). Additional instrument systems were brought
220
by collaborators, including the combined ATMONSYS lidar for measuring aerosol, T, and H2O
221
profiles and two Doppler wind lidars brought by Karlsruhe Institute of Technology (KIT), two
222
Collaborative Lower Atmospheric Profiling Systems (CLAMPS – NOAA NSSL), two 915 MHz
223
radar wind profilers with radio acoustic sounding systems (RASS) with MWRs (NOAA PSL),
224
and the Surface Radiation Budget Network (SURFRAD – NOAA GML) systems for measuring
225
incoming and outgoing radiation and cloud properties. While many of these instruments were
226
located within the 10 ×10 km CHEESEHEAD19 domain, some instruments were located at the
227
Prentice and Lakeland airports, located approximately 45 km south and east of the WLEF tower
228
respectively, to provide information on the spatial variability of boundary layer structure and
229
230
231
Three seven-day intensive observation periods (IOP) occurred on July 7 – 13, August 18 – 24,
232
and September 22 – 28. During these IOPs the University of Wyoming King Air (UWKA) flew
233
transects over an extended 30 × 30 km domain to measure EC fluxes, ABL depth, and
234
atmospheric profiles of water vapor and temperature. These observations will be used to test flux
235
tower scaling, observe atmospheric mesoscale patterning, and evaluate large eddy simulations
236
(LES). Also, during the IOPs, a team from NOAA ARL ATDD brought multiple sUASs for
237
measuring profiles of meteorological variables (T, H2O, U, P – see appendix for a list of
238
variables used in this paper) and land surface temperature. Additional information on the spatial
239
variations of surface meteorology was obtained using mobile observing systems operated in
240
pedestrian, boat, and car modes.
241
242
The four-month deployment spanned the summer to fall transition, capturing the shift in surface
243
energy balance from a more uniform evapotranspiration (latent heat flux) dominated landscape to
244
a patchy sensible heat flux dominated landscape. These energy balance shifts arise from seasonal
245
changes in plant phenological phases, ecosystem water use for photosynthesis, and available net
246
radiation. These shifts also provide a “natural experiment” with which to test hypotheses on how
247
heterogeneity influences energy balance closure and spatial scaling.
248
249
The study domain was partly chosen due to the history of atmospheric science research in the
250
region. Since 1995, University and NOAA investigators have sampled greenhouse gas profiles,
251
meteorology, and EC flux measurements (energy, carbon, momentum) at 30 m, 122 m, and 396
252
m above ground level (AGL; Fig. 2b) on the WLEF tall tower (Bakwin et al. 1998; Davis et al.
253
2003). The site also includes an FTIR solar-pointing spectrometer (TCCON) for total greenhouse
254
column observations operated by CalTech and NASA JPL. Two additional EC towers (US-WCr,
255
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
30 m in mature forest, and US-Los, 10 m in shrub fen wetland) have been operating for 20 years,
256
approximately 20 km from the tall tower (Cook et al. 2004; Desai et al. 2005; Sulman et al.
257
2009).
258
259
CHEESEHEAD19 builds upon previous tower mesonet experiments, including BOREAS
260
(Sellers et al. 1995), CASES99 (Poulos et al. 2002), SGP97 (Desai et al. 2006), IHOP (Kang et
261
al. 2007), LITFASS-2003 (Beyrich et al. 2006), EBEX (Oncley et al. 2007), BEAREX
262
(Anderson et al. 2012), HiWATER-MUSOEXE (Wang et al. 2015), SCALE-X (Wolf et al.
263
2017), that were aimed at understanding scaling of non-linear land-atmosphere interaction.
264
265
266
Fig. 2. (a) HSRL beam next to WLEF tall tower, (b) EC instruments at 396 m AGL
267
on the WLEF tall tower, and (c) the ISS field with modular wind profiler, sodar-
268
RASS, ceilometer, SURFRAD, EC and meteorological towers with UWKA flying
269
overhead and WLEF tall tower in the distance.
270
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Instrumentation & Measurements
271
272
Table 1. List of instruments and data collected during CHEESEHEAD19. For
explanation of the variable abbreviations please see the appendix.
Data source
Data provider
Location(s)
Period
Ground-based Measurements
Ameriflux/NOAA tall tower (US-
PFa/WLEF)
UW AOS
WLEF
Continuous
ChEAS Ameriflux towers: US-
WCr / US-Los / US-Syv / US-
Alq
UW AOS
Ameriflux sites (4)
Continuous
ISFS Eddy covariance towers
NCAR EOL ISFS
10x10 km (17 sites)
June-Oct
MSU Eddy covariance towers
Montana State U
& UW BSE
NW5 (ISS) and
SE1
June-Oct
Surface meteorology
NCAR EOL ISS
ISS field
July-Oct
& TWST
NOAA GML
ISS field1
Prentice Airport2
Lakeland Airport2
Downwelling SW/LW1,2, direct SW1,2,
diffuse SW1,2, upwelling SW/LW1,
PAR1, sky images1, cloud optical
depth1, cloud fraction1,2, cloud base
height2, mixed layer depth2,
meteorology1
July-Oct
(TWST:
Sep-Oct)
Vehicle/ Pedestrian/ Boat
transects
Jackson State U
Trails / Hay Lake
IOP 1, 2, 3
Chemical ionization mass spec
& ozone photometric analyzer
UW Chem
WLEF
IOP 1
Tall tower greenhouse gases
NOAA GML
WLEF
Continuous
& Biweekly
Tree temperature
Chequamegon
HS
5 sites, 10 trees
Oct
Atmospheric Profiling
449 MHz modular wind profiler
NCAR EOL ISS
ISS field
July-Oct
Sodar / RASS
NCAR EOL ISS
ISS field
July-Oct
Ceilometer
NCAR EOL ISS
ISS field
July-Oct
NCAR EOL ISS
ISS field
July-Oct
NCAR EOL ISS
ISS field
IOP 1, 2, 3
AERI
UW SSEC
SPARC
WLEF
July-Oct
HALO Lidar (1) vertical stare
UW SSEC
SPARC
WLEF
July-Oct
HSRL
UW SSEC
SPARC
WLEF
July-Oct
UW SSEC
WLEF
July-Oct
Precipitation Imaging Package
UW SSEC
WLEF
July-Oct
ATMONSYS:
Backscatter, Raman, and
Differential Absorption Lidar
KIT IMK-IFU
WLEF
July-Sep
HALO Lidars (2,3) RHI scans
KIT IMK-IFU
WLEF
July-Sep
915 MHz radar wind profiler w/
NOAA PSL
Prentice Airport,
Lakeland Airport
July-Oct
MWR
NOAA PSL
ISS field1
Prentice Airport2
Lakeland Airport3
July-Oct3
July-Sep2
Sep-Oct1
CLAMPS (MWR, AERI,
Doppler wind lidar)
NOAA NSSL
Prentice Airport,
Lakeland Airport
Sep-Oct
Airborne Measurements
Airborne eddy covariance
UWKA
30x30km, 24 flights
IOP 1, 2, 3
UWKA
30x30km, 24 flights
IOP 1, 2, 3
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Compact Raman Lidar (CRL)
UWKA
30x30km, 24 flights
IOP 1, 2, 3
Wyoming Cloud Lidar (WCL)
UWKA
30x30km, 24 flights
IOP 1, 2, 3
Meteodrone SSE sUAS
NOAA ARL ATDD
WLEF and SW2
IOP 1, 2, 3
Ozone sUAS
UWEC
WLEF
IOP 1
Surface Environment
HySpex
UW FWE
10x10 km, 4 flights
June-Aug
DJI S-1000 (sUAS)
NOAA ARL ATDD
WLEF and SW2
IOP 1, 2
sUAS leaf-on canopy LiDAR
UW Geog
11 tower sites
June
QL2 leaf-off LiDAR
USFS
30x30 km
Fall 2018
Vegetation/phenology sampling
UWM Geog
10x10 km (10 plots)
Sep-Oct
Vegetation Sampling
UW FWE
10x10 km (41 plots)
June-Oct
Soil bulk density and heat
capacity
NCAR EOL
17 tower sites
July-Oct
Soil samples
UW AOS
16 tower sites
Oct
Soil samples
Butternut Schools
7 sites
July
ECOSTRESS, GEDI, OCO3
NASA JPL
30x30 km
Oct 8
273
Ground-based Measurements
274
275
Towers sampled three-dimensional wind velocity, temperature, moisture, and CO2 at 20 Hz to
276
measure land-atmosphere fluxes (τ, HS, HL, FCO2). Each tower also measured net radiation, soil
277
heat flux at 5 cm depth (and soil temperature profile, heat capacity, and moisture to determine
278
soil heat storage), and a 3-level air temperature and humidity profile to estimate canopy heat
279
storage. A majority of the sites were forested and had flux instruments mounted 33 m AGL (Fig.
280
3; Table S1). Instruments for wetland, grass, and lake sites were mounted between 1 – 3 m AGL
281
to maintain consistent vegetation within the flux footprint. Tower placement within the 10 × 10
282
km study domain followed a stratified random grid pattern, taking into account practical
283
considerations including distance to road, suitable gap in trees for a tower, USFS-owned land,
284
etc. Individual towers were an average of 1.4 km from their nearest neighboring tower and an
285
average of 3.5 km from the tall tower. This meant that under certain conditions (e.g., high wind
286
speeds, stable stratification) several of the towers shared overlapping flux footprints; a favorable
287
condition for applying some of the data-driven scaling methods used in the project. Additionally,
288
the semi-random placement meant that the towers were not chosen by distributing the towers in
289
the centers of the most homogeneous areas of the various land cover types. Thus, within the
290
individual footprint of each tower there was often spatial variability in vegetation height and type
291
(deciduous vs. evergreen). While this can complicate analyses of flux measurements, it generates
292
more representative data from these types of mixed forests. Furthermore, we expect it will
293
enhance the ability of the data-driven methods for estimating domain-wide fluxes.
294
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
295
Fig. 3. (a) EC tower SW1 – an example of the 33 m AGL telescoping towers
296
deployed by NCAR ISFS and (b) EC instruments mounted at the top of tower
297
SW2.
298
299
A suite of high-quality radiation sensors was deployed in the ISS field as a complement to the
300
net radiometers installed on each flux tower. The full suite included high-quality upwelling and
301
302
budget, as well as ancillary measurements of meteorological parameters, photosynthetically
303
active radiation (PAR), and clouds as described in Table S2. Radiation measurements are
304
manually screened and then processed through an automated data quality procedure (Long and
305
Shi 2008). Clear sky radiation fluxes are estimated using the Radiative Flux Analysis method
306
(Long and Ackerman 2000; Long and Turner 2008), from which derivation of cloud radiative
307
effects as well as other data products such as fractional sky cover (Long et al. 2006; Dürr and
308
Philipona 2004) and cloud optical depth (Barnard and Long 2004; Niple et al. 2016) are
309
calculated. Measurements of cloud properties will allow us to quantify their impacts on the
310
radiative and turbulent heat fluxes to better understand the two-way coupling between cloud-
311
radiative interactions and boundary layer evolution, and to investigate the effect on EC non-
312
closure.
313
314
A smaller suite of radiation, cloud, and surface meteorological measurements were deployed at
315
the Prentice and Lakeland Airports, approximately 45 km south and east from the ISS field,
316
respectively (Fig. 1), to characterize the larger spatial scale inhomogeneities. These
317
measurements include downwelling shortwave and longwave irradiance as well as diffuse and
318
direct components of shortwave irradiance (Table S2); sufficient information to derive cloud
319
radiative effects and fractional sky cover using the Radiative Flux Analysis method described
320
above. Ceilometers deployed at the two airport sites provided additional cloud and boundary
321
layer information.
322
323
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Airborne Measurements
324
325
During each IOP the UWKA flew over the study area to measure spatial EC fluxes of heat, water
326
vapor, and CO2. The purpose of the observations was to test flux tower scaling and observe
327
atmospheric mesoscale patterning. The UWKA also measured cross-sectional profiles of water
328
vapor and temperature below the flight level using a downward pointing Compact Raman Lidar
329
(CRL, Wu et al. 2016) and ABL depth with the upward looking Wyoming Cloud Lidar (WCL,
330
Wang et al. 2009).
331
332
Flights over the domain occurred on four days during each of the three IOPs (Table S3). On each
333
day there were two three-hour flights, one in the morning (1400 – 1700 UTC) and one in the
334
afternoon (1900 – 2200 UTC). Flights consisted of ten 30-km down-and-back transects across
335
the domain. The first leg of each transect was flown at 400 m AGL, while the return leg was
336
flown at 100 m AGL. Flight transects alternated between straight and diagonal passes.
337
338
Three different flight patterns were determined prior to the experiment (oriented SE→NW,
339
SW→NE, and W→E). Flying them either in forward or reverse order resulted in six distinct
340
flight sequences that maximize data coverage under different wind conditions (see sidebar
341
Continuity through Environmental Response Functions). The main objectives were to maximize
342
1) the number of independent atmospheric eddies and 2) surface flux footprint observed by the
343
aircraft EC measurements, while 3) ensuring crew safety. This was achieved by designing a
344
parsimonious set of only three flight patterns that allowed the UWKA to fly perpendicular to the
345
prevailing winds within a range of ±45° on any given day (Metzger et al., in preparation). The
346
30-km flight legs extended an average of 10 km beyond the domain to compute a robust
347
mesoscale eddy flux (Mauder et al. 2007a, 2008a) by capturing enough eddies and mesoscale
348
variation to properly compute statistics for fluxes using the wavelet decomposition method.
349
350
The low-altitude legs were primarily used to measure EC fluxes. The altitude 100 m AGL was
351
chosen to ensure flux measurements were made in the surface layer, as well as to minimize flux
352
footprint errors over the 10 × 10 km sampling domain. It was also the lowest altitude deemed
353
safe to fly, as canopy height extended up to 35 m. The low-altitude legs were also used to
354
identify ABL depth with the upward pointing 355 nm WCL. The primary purpose of the high-
355
altitude legs (400 m AGL) was to map temperature and moisture profiles of the atmosphere with
356
the CRL. These data were collected to estimate mesoscale development and calculate flux
357
divergence and storage terms.
358
359
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
360
Fig. 4. The location (superimposed) of all 480 flight legs completed during the
361
CHEESEHEAD19 field campaign. The yellow square represents the study domain
362
and the red dots indicate the flux tower locations.
363
364
Atmospheric Profiling
365
366
Characterizing the mesoscale environment over the study domain was accomplished with a range
367
of platforms and instruments to measure profiles of wind, water vapor, temperature, aerosols,
368
and gases at different temporal and spatial scales (Fig. 1; Table 1).
369
370
The NCAR ISS was located in a field 1.6 km west of the tall tower (45.946°N, 90.294°W). It
371
deployed a radar wind profiler, sodar-RASS, ceilometer, all-sky camera, and a surface
372
meteorology station to measure ABL depth, winds, water vapor, and temperature. The 449 MHz
373
modular wind profiler measured 30-minute wind profiles with 150 m vertical resolution up to
374
several km AGL, while the sodar-RASS was capable of higher resolution (20 m; 10-minute), but
375
only penetrated to ~400 m AGL. Meteorological profiles were also measured with 172
376
radiosonde launches (daily 18Z soundings and 3 – 4 additional soundings on IOP days). These
377
instruments characterized the ABL from nocturnal boundary layer (sunrise sounding), through
378
ABL development (mid-morning and afternoon), to peak ABL (late afternoon sounding). In
379
mid-September, one of the MWRs located at the Prentice Airport was relocated to this location,
380
due to the failure of the AERI at the tall tower site in early September.
381
382
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Several profiling systems were deployed at the base of the tall tower. SPARC (Wagner et al.
383
2019) was located 50 m north of the WLEF tower and was equipped with an Atmospheric
384
385
2004]), a High-Spectral Resolution Lidar (HSRL [Eloranta 2005]), and a ceilometer. Profiles of
386
387
(Turner and Löhnert 2014; Turner and Blumberg 2019). The HSRL sampled ABL aerosol
388
backscatter and depolarization ratio at 532 nm and 1064 nm. The ceilometer provided an
389
390
391
The ATMONSYS system was placed beside the SPARC system, measuring atmospheric water
392
vapor, temperature, and aerosol. The primary light source of the ATMONSYS lidar is a 100 Hz
393
diode pumped Nd:Yag laser with the harmonic generation of 532 nm and 355 nm. The 532 nm
394
light (P 27 W) is used for optical pumping a Ti:Sapphire laser, generating 817 nm (P 2 W)
395
for water vapor profiling with the high resolving DIAL (Differential Absorption Lidar) method
396
as well as for profiling aerosol backscatter. The 355 nm light is used for temperature profiling
397
from rotational Raman backscatter. The system setup as installed during CHEESEHEAD19
398
(Vogelmann et al. 2020) allows for spatial sampling of 7.5 m and integration times of 20 s for
399
aerosols and water vapor measurements and 300 s for temperature profiling.
400
401
In the field to the east of the trailers were three Doppler wind lidars. One lidar (LVS) measured
402
in vertical stare mode throughout the measurement campaign. The other two lidars (LA, LB)
403
were placed 90 meters away from the LVS and made range–height indicator (RHI) scans (66º –
404
87º elevation angle) pointing towards the LVS. This setup constitutes a virtual tower that
405
provides vertical wind speed measurements and calculates average horizontal wind speed at
406
multiple height levels above the LVS (Calhoun et al. 2006; Klein et al. 2015; Wulfmeyer et al.
407
2018). Additionally, the collocation of lidars for measuring 3D winds, temperature, and water
408
vapor facilitates calculation of flux profiles of τ, HS, and HL, as well as flux divergence
409
(Wulfmeyer et al. 2016).
410
411
Two precipitation instruments (a Precipitation Imaging Package [PIP] and a Micro Rain Radar
412
Pro [MRRPro; Metek GmbH]) were installed at WLEF. The PIP is a video disdrometer system
413
that records information about hydrometers and produces end user products such as particle size
414
distributions, fall speeds, and rain rate at one-minute resolution (Newman et al., 2009; Pettersen
415
et al., 2020a; Pettersen et al., 2020b). The MRRPro is a 24-GHz, frequency modulated
416
continuous wave, vertically profiling Doppler radar (Klugmann et al. 1996) that is used for
417
observations of both rain (i.e., Peters et al. 2002) and snow (Kneifel et al. 2011).
418
419
Additional thermodynamic profiling systems were operated at the Prentice and Lakeland airports
420
throughout the experiment to characterize the boundary layer variability and evolution around
421
the CHEESEHEAD19 domain. The primary motivation of these two profiling sites was to
422
characterize the mesoscale transport and role of advection on the ABL mass balance of the
423
CHEESEHEAD19 domain. At each location, a 915 MHz wind profiler with radio acoustic
424
sounding system was deployed together with a multi-channel MWR. These instruments
425
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
provided profiles of horizontal wind and temperature, and low vertical resolution profiles of
426
water vapor.
427
428
Prior to IOP3, two mobile CLAMPS facilities (Wagner et al. 2019) were deployed at Prentice
429
and Lakeland. The systems contained a Doppler lidar wind profiler, an AERI, and a microwave
430
radiometer (MWR). The information content in the AERI observations is higher than in the
431
MWR, and thus the retrieved water vapor and temperature profiles have better vertical resolution
432
and accuracy (Löhnert et al 2009; Blumberg et al. 2015). The Doppler lidars complemented the
433
radar wind profilers by providing higher temporal and vertical resolution measurements than the
434
radars, but the radars were able to profile winds several km higher than the lidars.
435
436
Two small unoccupied aircraft systems (sUAS) were flown to characterize surface and near-
437
surface conditions (Fig. S1). During IOP1 (IOP2), a DJI S-1000 (e.g., Lee et al. 2019) was flown
438
adjacent to the SW2 tower (WLEF tall tower) to quantify the variability in surface sensible heat
439
flux (e.g., Lee et al. 2017). During all three IOPs, the Meteomatics Meteodrone SSE sUAS was
440
used to sample the evolution of near-surface profiles of temperature, moisture, and wind up to
441
213 m AGL, which was the maximum altitude to which we could operate our sUAS per our
442
cooperative agreement with the FAA. Additionally, the Meteodrone SSE was used to sample the
443
horizontal variability in temperature, moisture, and wind fields over a ~ 100 × 100 m box
444
surrounding the SW2 and WLEF towers. Over all IOPs, 26 (103) flights were conducted with the
445
DJI S-1000 (Meteodrone SSE).
446
447
Surface Environment
448
449
Data on the ecological environment were collected to provide the boundary conditions of canopy
450
type, activity, and stress, needed for estimating scaling properties. This was done with a variety
451
of methods, including airborne imaging spectroscopy, ground-based phenological
452
characterization, and tree growth measurements.
453
454
Foliar functional traits such as leaf mass per area (LMA) and nitrogen concentration strongly
455
influence photosynthetic capacity and plant growth (i.e., net primary production, NPP)
456
(Niinemets 2001; Kattge et al. 2009), and can be mapped using imaging spectroscopy (aka
457
hyperspectral remote sensing, Singh et al. 2015). To map foliar functional traits across the
458
domain a full-range imaging spectroscopy system comprising two co-aligned imagers (VNIR-
459
1800 and SWIR-384; HySpex, Skedsmokorset, Norway) was operated from a Cessna 210 at
460
1400 m AGL on four days (6/26, 7/11, 8/4, 8/30), producing images with 1 m spatial resolution.
461
The HySpex collects 474 bands with a spectral resolution of 3.26 nm in the VNIR (400-1000
462
nm) and 5.45 nm in the SWIR (1000-2500 nm).
463
464
Extensive ground-based vegetation samples were collected to support the hyperspectral image
465
analyses. These included 41 plots in the domain for measuring tree species (400+ trees), root
466
growth, tree height, diameter at breast height (DBH), net primary production (NPP), biometry,
467
leaf area index (LAI). This also included 122 top-of-canopy foliar samples to estimate leaf level
468
function traits following the protocol from Serbin et al. (2014).
469
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
470
In combination with an existing extensive database of foliar traits and image spectra (Wang et al.
471
in press), we will use the 122 foliar samples to develop and validate 1 m resolution maps for all
472
four dates of numerous foliar functional traits hypothesized to influence NPP, including LMA,
473
nitrogen concentration, chlorophyll and other pigments, phosphorus, non-structural
474
carbohydrates, fiber and lignin, and phenolics). From this, we will test the relationship between
475
functional traits and GPP (as derived from towers) and peak-season integrated NPP (early-July to
476
early-September, derived from the 41 plots). We will generate 1 m maps of NPP and GPP, and
477
identify the foliar factors that most influence each.
478
479
Additional plots were used to measure vegetation phenology as it changed through the season,
480
building upon several years of previous phenological observations collected in the domain.
481
Autumn tree leaf color and fall phenology levels were visually observed and recorded at least
482
twice weekly over six weeks during the senescence period (Sep 1 to Oct 25) for a group of 214
483
individual trees (at ten sites distributed over the 10 ×10 km area) that were representative of the
484
major species.
485
486
Forest canopy structure was characterized using an sUAS-based lidar system (Routescene;
487
Edinburgh, Scotland) acquiring high density point clouds (500 pts m−2) within footprints from 11
488
CHEESEHEAD19 flux tower sites including aspen, pine, poplar, larch, cedar, and hardwood
489
forests. Areas surveyed ranged between 0.25 – 1 km2 per site. Additional canopy information for
490
the entire domain came from leaf-off LiDAR from USFS sampling (1 m2 resolution) conducted
491
for the three counties that comprise the study area between 2014 and 2017.
492
493
Land surface temperature (LST) is a key environmental driver of the surface energy balance
494
(e.g., Metzger et al., 2013; Xu et al. 2017a). Spatially explicit LST can be acquired from satellite
495
remote sensing (Fig. 5). However, there are tradeoffs in space and time resolutions such that no
496
single sensor provides sufficient resolution for use as a land surface driver to map heat fluxes
497
across space at sub-kilometer and hourly time steps required for the hypotheses here. Also,
498
remote sensing methods may not be able to distinguish between true surface temperature and
499
upper canopy temperature. Here, we are investigating multi-sensor fusion using a combination of
500
in situ thermal drone and infrared camera imagery, ECOSTRESS, Landsat, VIIRS and/or GOES
501
(Wu et al. 2013).
502
503
504
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
505
Fig. 5. Land surface temperature on June 15, 2019 from (a) ERA5 reanalysis and
506
(b) derived from Landsat 8, where sub-grid spatial resolution is present, but
507
temporal resolution is low (Gerace et al. 2020; Landsat 8 data courtesy of the U.S.
508
Geological Survey; ERA5 data generated using Copernicus Climate Change Service
509
Information 2020).
510
511
512
Data Analysis & Modeling
513
514
Two analytical methods have been proposed to test the hypotheses of this study. The first is the
515
application of ERF-VCV – a data driven approach that can be used to account for the dispersive
516
fluxes missed by single-tower EC measurements, and to upscale fluxes across the
517
CHEESEHEAD19 domain (Metzger, 2018, Xu et al., 2018, Xu et al., 2020). ERF-VCV uses a
518
machine learning algorithm to find relationships between measured fluxes and their
519
meteorological and surface drivers within the flux footprints (see sidebar).
520
521
We will perform LES for the IOP days using the Parallelized LES Model PALM (Raasch and
522
Schröter 2001; Maronga et al. 2015; 2020). In the LES we will emulate airborne- and tower-
523
mounted flux observations to compare them against the ‘real-world’ observations with the ability
524
to also evaluate flux footprints using Lagrangian particle modelling, radiation footprints, storage
525
fluxes at various locations and points in time. To accurately simulate the physical processes as
526
observed during the IOPs of the field experiment as realistically as possible, we will assume
527
realistic topography for the experiment site, and apply a Land Surface Model (LSM) with a
528
coupled soil and radiation model, as well as a Plant Canopy Model (PCM). The use of the LSM
529
and PCM runs instead of prescribed surface fluxes will allow us to study land-atmosphere
530
feedbacks such as self-reinforcement of mesoscale circulations over the heterogeneous study
531
domain. The LSM will be set up for each IOP test case, with land use classes, soil, and
532
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
vegetation data as observed during the field experiment. Further, in order to account for
533
synoptic-scale processes during the IOPs (e.g., advection of air masses with different
534
characteristics) we will nest the LES domain into a larger-scale model.
535
536
One proposed goal is to derive a parametric heterogeneity correction of dispersive fluxes
537
by setting up virtual towers within the LES, applying it to CHEESEHEAD19 tower flux field
538
data, and evaluating it with ERF-VCV flux grids. Therefore, tower-level turbulence
539
characteristics will be simulated as observed during the field campaign to investigate the energy
540
balance non-closure problem. Additionally, by emulating 'real-world' measurements we intend to
541
help interpret the observations – such as giving hints where secondary circulations occur or how
542
far heterogeneity signals extend downwind.
543
544
545
546
SIDEBAR: CONTINUITY THROUGH ENVIRONMENTAL RESPONSE FUNCTIONS
547
548
CHEESEHEAD19 disentangles how land surface heterogeneity relates to atmospheric transport
549
in mesoscale eddies, which contributes to the discrepancy between EC flux observations and
550
model predictions. We strive to create a new class of observational flux data product that
551
reconciles resulting biases on orders of 10% (Chen et al. 2011; Foken et al. 2011) and reveals
552
actual surface emissions. For non-uniform exchange surfaces such as in CHEESEHEAD19, this
553
requires us to evaluate the conservation of mass and energy continuously in time and space
554
throughout the study domain (e.g., Finnigan 2008). However, even intensive field
555
instrumentation campaigns such as CHEESEHEAD19 cannot produce observations everywhere,
556
all the time. Here, Environmental Response Functions (ERF; Metzger et al. 2013; Metzger 2018)
557
can help attain the necessary information continuum from individual observation plots to model
558
grid scale. To achieve this, ERFs complement information across disciplines and observation
559
types by using a machine learning algorithm to find relationships between measured fluxes and
560
their meteorological and surface drivers within the flux footprints (Fig. 6A). This provides a
561
powerful approach not only for post-field data synthesis, but already in the experiment planning
562
stage e.g. in combination with Large Eddy Simulations (Fig. 6B). Maximizing scientific return
563
on experimental investment (Fig. 6C; Metzger et al., in preparation) is one example of how ERFs
564
can help close the circle among obtaining “knowledge from data” and “data from knowledge”
565
(Reichstein et al. 2019).
566
567
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
568
Fig. 6. Panel A: Environmental Response Functions (ERFs) augment sparse
569
response observations (e.g., tower and aircraft EC) with abundant driver
570
observations (e.g., meteorological stations and satellites). High-rate time-frequency
571
decomposition and source area modeling facilitate data joins among these
572
response and driver observations at minute- and meter-scale. Machine learning
573
then extracts a driver-response process model from the resulting space- and time-
574
aligned dataset. Ultimately, this driver-response process model complements the
575
properties of response and driver observations in the response data product. In the
576
present example these are meter-scale sensible heat flux maps, which can be used
577
to more reliably evaluate the conservation of energy across the non-uniform
578
579
580
PanelB: During the experiment planning stage we used Large Eddy Simulations
581
(LES) to create synthetic atmospheres over the CHEESEHEAD19 domain for
582
different synoptic conditions. We simultaneously sampled the synthetic
583
atmospheres as observed by different virtual experiment designs. Each experiment
584
design resulted in a separate set of virtual observations which we independently
585
processed through the ERFs in PanelA.
586
587
PanelC: We benchmarked the different experiment designs against their ability to
588
reproduce the LES reference in the form of flux grids that ERF reconstructed from
589
the virtual observations alone. Identifying the optimal experiment design not only
590
allowed us to double the scientific return on experimental investment, but also to
591
simplify flight plans and increase crew safety. For additional detail see the full study
592
by Metzger et al. (in preparation).
593
594
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
PRELIMINARY RESULTS
595
596
Over the course of the four-month study period the region exhibited light winds (diurnal means
597
from 1 – 4 m s−1) from all directions, with the most prevalent direction being southwesterly (Fig.
598
7). Air and soil temperatures decreased over the period, while soil moisture increased (Fig. 7b,c).
599
Daily mean net radiation decreased over the course of the study, which showed a direct
600
relationship with ABL height (measured as the height of the inversion on the diurnal radiosonde
601
launches [Fig. 7d]). One of the most relevant seasonal changes with respect to energy balance
602
was the change in the daytime Bowen Ratio (HS / HL) which averaged 0.5 in the summer and 1.0
603
in the fall, with the latter period having more variability than the former. Diurnal cycles of
604
sensible and latent heat flux show that latent heat flux is much larger in the summer when the
605
canopy is fully evapotranspiring compared to the fall, when senescence of broadleaf trees
606
reduces HL, allowing HS to comprise a larger share of the total heat flux over the region (Fig. 7f
607
– i).
608
609
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
610
Fig. 7. Daily mean (a) wind speed and direction, (b) temperature and relative humidity, (c) soil
611
moisture, (d) net radiation and ABL height, and (e) Bowen ratio averaged across all ISFS towers.
612
Aerial view of site NE2 on (f) July 12, 2020 and (g) October 9, 2020. Diurnal cycles of sensible
613
and latent heat averaged across all ISFS sites for the weeks of (h) Oct. 4 – 11 and (i) July 7 – 14.
614
615
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Continuous data collection throughout the campaign linked the energy balance components to
616
the remotely sensed atmospheric environment (Fig. 8). As is typical for EC measurements, we
617
observed energy fluxes that were lower in magnitude than the net incoming energy (RN – G),
618
when averaged across all sites. The magnitudes of the energy balance residual (CEB) was largest
619
during the daytime, when incoming solar radiation was highest. The opposite sign of CEB from
620
day to night in part can be attributed to heat storage in the canopy. However, the magnitudes of
621
the daytime values are larger than the nighttime values, which results in a daily mean imbalance.
622
623
624
Fig. 8. (a) Stacked energy balance components: net radiation minus ground heat
625
flux (RN - G), sensible and latent heat flux (HS and HL), and energy balance residual
626
(CEB) on Sep 24, 2019; (b) radiosonde profiles of potential temperature (θ); and (c)
627
time series of wind speed profile with overlaid ABL height from ceilometer (black
628
dots) and radiosondes (colored diamonds and dashed lines correspond to
629
630
631
The energy balance residual peaked under conditions of low turbulence (Fig. 9). It is during such
632
periods of calm wind and strongly unstable stratification in which thermally-induced mesoscale
633
eddies resulting from landscape-scale heterogeneity are expected (Steinfeld et al. 2007). This
634
lends support to the hypothesis that mesoscale eddies are responsible for the energy balance non-
635
closure.
636
637
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
638
Fig. 9. Daily mean energy balance residuals (CEB) normalized by net radiation minus
639
ground heat flux (RN - G) plotted against friction velocity (u*) for all ISFS EC towers
640
for the entire CHEESEHEAD19 dataset (excludes individual towers on days
641
without complete quality-controlled data).
642
643
Tower measurements, combined with in-situ measurements of air temperature and land surface
644
temperature from the DJI S-1000, were used to quantify variability in surface HS following Lee
645
et al. [2017], as shown in an example from 12 Jul 2019 (Fig. 10). On this day, as well as others,
646
there was significant temperature and HS variability; temperature (HS) differences were ~ 10°C
647
(100 W m−2) over the ~ 500 × 500 m area surrounding the SW2 tower. Fig. 10 illustrates even
648
finer scale resolution of surface temperature than the measures shown in Fig. 5. Such spatial
649
variation is directly related to underlying surface characteristics.
650
651
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
652
Fig. 10. Surface temperature (a) and Hs (b) from a downward-pointing infrared
653
camera flown on the DJI S-1000 sUAS surrounding the SW2 tower between 1504
654
and 1518 UTC 12 Jul 2019. Same for panels (c) and (d), but between 1614 and 1628
655
UTC 12 Jul 2019. Hs computed following Lee et al. (2017). As the technique
656
requires an initial Hs to derive the variability in Hs and Hs was unavailable from
657
SW2 on 12 Jul, Hs at SW2 was estimated using a linear regression with data from
658
nearby towers. Mean ± 1 standard deviation shown at the bottom of each panel.
659
660
Landscape heterogeneity was observed for a range of environmental variables, including
661
vegetation spectral characteristics and canopy height captured from downward-looking airborne
662
remote sensing instruments (Fig. 11). False color HySpex imagery is being used to differentiate
663
plant functional types at 1 m2 resolution. Additional information on leaf-on canopy structure,
664
obtained from the Routescene LiDAR at 11 flux sites and across the entire domain from the State
665
of Wisconsin leaf-off LiDAR dataset, are being used to identify surface roughness in the flux
666
footprints of the EC towers. In addition, these spatial data are being used as input drivers within
667
the ERF-VCV machine learning approach.
668
669
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
670
Fig. 11. Surface maps showing spatial variation around tower site SW2 in (a)
671
surface temperature measured by the DJI S-1000 (same as Fig. 10a), (b) vegetation
672
spectral characteristics measured by the HySpex shown as a false color image (849
673
nm – red, 1650 nm – green, 2217 nm – blue), and (c) surface/canopy height
674
measured by the sUAS Routescene lidar.
675
676
There is also spatial variation in the energy balance components across the domain on a typical
677
day (Fig. 12a). This variability includes the relative weighting of latent and sensible heat fluxes,
678
as well as the magnitude of the energy balance residual. The mean energy balance closure
679
(calculated as [HS + HL]/[RN – G]) across all the sites over the entire study period was 0.8 This is
680
typical for EC towers and supports the need for the advanced methods put forth by this study.
681
682
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
683
Fig. 12. Average daily mean energy balance pie charts for the flux towers over the
684
entire study period. The pie chart with the cyan outline (bottom center) was a
685
buoy EC system deployed on a small lake.
686
687
To address this spatial and temporal variability we are testing different types of spatial EC
688
techniques, which have been suggested as a means of mitigating errors arising from single-site
689
EC (Steinfeld et al. 2007; Mauder et al. 2008b). Using LES, Xu et al. (2020) found that standard
690
spatial EC improved closure over standard temporal EC, while a combined spatio-temporal
691
method performed better still. Further, by applying the ERF-VCV approach, the energy balance
692
was found to be almost completely closed.
693
694
Here we had the ability to calculate spatial fluxes from two different sources. First, the spatial
695
fluxes were calculated using a wavelet decomposition on the aircraft EC datasets. This dataset
696
has good spatial coverage but limited temporal resolution, though, with 72 flight hours spread
697
across 12 days, it is one of the largest airborne EC datasets ever collected.
698
699
The second data source for spatial EC was the set of 20 flux towers spread across the domain.
700
Calculations for flux footprints on Sep 26, 2019 (Fig. 13) show that spatial coverage of the
701
towers (including WLEF) covered roughly 8% of the domain (using Kljun et al. [2015]). This is
702
a significant increase compared to a single tower set up (typically <<1% of a 10 × 10 km area).
703
An additional benefit from the experiment design is that the towers cover a range of physical
704
environments. These data are being used to confirm the LES model results for improvements to
705
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
energy balance closure. By combining the tower and aircraft EC datasets we have excellent
706
coverage (~80%) of the study domain on flight days (Fig. 13).
707
708
709
Fig. 13. Flux footprint climatologies from the 20 flux towers and aircraft on the
710
morning of September 26, 2019. Tower footprints extend to the 90% footprint with
711
10% contour lines shown down to 10% (calculated based on Kljun et al. [2015]). The
712
heat map shows aircraft flux footprints with areas of strongest flux contribution in
713
red, grading to blue where there was no contribution (calculated based on Metzger
714
et al. [2013]). UWKA flight tracks shown as dashed black lines.
715
716
The characterization of the ABL and identification of mesoscale eddies will be performed using
717
lidar measurements of wind, water vapor, temperature, and backscatter. Figure 14 shows an
718
example of this on September 24. Increasing water vapor through the day is representative of a
719
large-scale warm, wet airmass entering the domain (Fig. 14c,d; Fig. S2a). This characterizes the
720
variation in water vapor throughout the collection of the morning UWKA CRL dataset (Fig.
721
14a). The afternoon CRL dataset (Fig. 14b) shows a more evenly mixed ABL, with variation in
722
water vapor due to local pockets of relatively moist and dry air. These two examples show the
723
varying applications of the CRL data depending on the atmospheric environment, with the
724
afternoon flight illustrating the potential of the dataset for determining the degree of ABL
725
heterogeneity arising from surface heterogeneity. Further analysis will investigate relationships
726
with underlying vegetation and LST.
727
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
728
Around 1200 UTC (7am local time) net radiation becomes positive (Fig. 8a) and soon after we
729
see the breakup of the surface inversion (Fig. 14d). Around 1400 – 1500 UTC we see the ABL
730
grow (Fig. 8c) followed by development of large-scale structures revealed by strong oscillations
731
in vertical wind speed (±2 m s−1; Fig. 14e). During peak hours the angle of attack of the wind
732
vectors oscillate between roughly -30º to 50º degrees on time scales of 10 minutes to an hour.
733
These angles far exceed those of the underlying terrain, suggesting that these periodic updrafts
734
and downdrafts are the result of mesoscale eddies.
735
736
Around 1900 UTC the domain clouds over, seen in RN and backscatter (Fig. 8a; Fig. 14f; Fig.
737
S2b). This causes the strength of the oscillation in vertical wind to decrease (Fig. 14e), which
738
coincides with a change in the relative weighting of the different energy balance components,
739
with both RN and HS decreasing strongly, while HL decreases only slightly (Fig. 8a). An increase
740
in RN around 2000 UTC corresponds to strengthening vertical wind speed oscillations. Further
741
analyses will investigate the prevalence of this result across the entire dataset and examine
742
specific drivers and possible implications for EC energy balance closure. These datasets show
743
that changes in ABL development are closely tied to changes in the surface energy fluxes,
744
highlighting the potential research applications of the CHEESEHEAD19 data.
745
746
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
747
Fig. 14. (a) and (b) show CRL cross sections of H2O mixing ratio (cut to domain
748
size; panel c colorbar represents panels a - c) for each of 10 legs on Research Flights
749
17 and 18 (Sep 24 at 13:51 - 16:26 and 19:11 - 21:31 UTC); time series profiles of
750
(c) H2O mixing ratio and (d) T measured by the ground-based MWR, (e) vertical
751
wind speed calculated using the ground-based RHI scanning wind lidars (LA, LB) for
752
the column above LVS, and (f) 532 nm backscatter from the ground-based HSRL at
753
WLEF tall tower on Sep 24, 2019.
754
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
EDUCATIONAL OUTREACH
755
756
Several public events were conducted to introduce and communicate the science goals and
757
objectives of the project. These include a pre-experiment community-wide public presentation at
758
the Park Falls Public Library and a summer open house at several sites, enabling members of the
759
community to visit data collection locations, meet CHEESEHEAD19 team members, and
760
participate in demonstrations of the instruments. CHEESEHEAD19 team members also
761
participated in surveys and in training on fieldwork bullying and sexual harassment prevention
762
(Fischer et al. in review).
763
764
The project also worked with two local school groups, one from Butternut, Wisconsin K-12
765
School and another from Chequamegon High School of Park Falls, WI, to include them as
766
supporting data collectors. The GLOBE (Global Learning and Observations to Benefit the
767
Environment) program trained Butternut K-12 students and a teacher to collect land cover
768
classification data, soil properties, and atmospheric data at seven of the tower sites at multiple
769
times throughout the summer. The high school group installed ten tree temperature sensors at
770
five of the forest flux tower sites, which are being used to estimate biomass heat storage. We
771
772
which conducted independent research projects on micrometeorology and carbon cycling.
773
774
775
Fig. 15. Bill Brown (just right of the radiosonde balloon) describing the capabilities
776
of the ISS facility during the community open house.
777
778
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
DATA AND CODE AVAILABILITY
779
780
The database of observations and models is currently online and freely available to the
781
community and public for general use or for further scientific investigation. The datasets and
782
supporting information have been gathered together in the NCAR Earth Observatory Laboratory
783
(EOL) data repository which can be accessed through the project web page at
784
https://www.eol.ucar.edu/field_projects/cheesehead. The project has open data and code policies,
785
in which other researchers are encouraged to use CHEESEHEAD19 resources for their own
786
research. The policies can be accessed through the above web page.
787
788
Additionally, data are stored and are being used for in-depth analysis and modeling purposes on
789
the NSF-funded cloud computing platform CyVerse, with the goal of having a central location
790
for users to bring their code to the data in a way that maintains data and code provenance for
791
792
descriptions of the sites, photographs, and data plots can be found on the CHEESEHEAD19
793
794
795
CONCLUSIONS
796
797
The data collected during the CHEESEHEAD19 field campaign show a distinct seasonal shift in
798
surface energy fluxes, as well as spatial patterning that appears to be directly related to the
799
characteristics of the underlying surface environment. Consequently, the imbalance in the energy
800
budget displays both temporal and spatial variability, with the imbalance becoming larger under
801
periods of low turbulence. The broad coverage of the measured fluxes using the 20-tower
802
network and airborne EC, combined with the collection of spatial data of surface characteristics
803
like LST, vegetation type, and canopy structure, will enable thorough investigation of the causes
804
of energy balance non-closure. Additionally, the suite of atmospheric profiling instrumentation
805
characterizes the mesoscale structure of atmospheric flows over the study domain to an
806
unprecedented degree, helping to determine how mesoscale eddies contribute to measured
807
imbalances. The observational dataset provided by CHEESEHEAD19 will also enable the use of
808
machine-learning approaches and LES for testing hypotheses on scaling and parameterization of
809
sub-grid processes in mesoscale meteorological models. Findings emerging from this project are
810
expected to have broad implications for heterogeneous terrestrial regions beyond the specific
811
study domain.
812
813
814
APPENDIX: LIST OF ACRONYMS
815
AGL – above ground level
816
ARL - Air Resources Laboratory (NOAA)
817
ATDD – Atmospheric Turbulence and Diffusion Division (NOAA)
818
CHEESEHEAD19 – Chequamegon Heterogeneous Ecosystem Energy-balance Study Enabled
819
by a High-density Extensive Array of Detectors 2019
820
CLAMPS – Collaborative Lower Atmospheric Mobile Profiling System (NOAA NSSL)
821
CRL – Compact Raman Lidar
822
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
EC – Eddy Covariance
823
GML – Global Monitoring Laboratory (NOAA)
824
IOP – Intensive Observation Period
825
LES – Large Eddy Simulation
826
lidar – light detection and ranging
827
LSM – Land Surface Model
828
NCAR – National Center for Atmospheric Research
829
NOAA – National Atmospheric and Oceanic Administration
830
NSF – National Science Foundation
831
NSSL – National Severe Storms Laboratory (NOAA)
832
PALM – Parallelized LES Model
833
PSL – Physical Sciences Laboratory (NOAA)
834
835
RASS – Radio Acoustic Sounding System
836
sodar – sonic detection and ranging
837
sUAS - small Unmanned Aircraft System
838
839
UWKA – University of Wyoming King Air
840
841
APPENDIX: LIST OF VARIABLES
842
FCO2 – CO2 flux (μmol m−2 s−1)
843
G – Ground heat flux (W m−2)
844
H2O – water vapor mixing ratio (g kg−1)
845
HS – sensible heat flux (W m−2)
846
HL – latent heat flux (W m−2)
847
LST – land surface temperature (C)
848
P – Pressure (mbar)
849
RN – Net surface radiation (W m−2)
850
T – temperature (C)
851
Tv – virtual temperature (C)
852
U – horizontal wind speed (m s−1)
853
u* – friction velocity (m s−1)
854
w – vertical wind speed (m s−1)
855
θ – potential temperature (C)
856
θv – virtual potential temperature (C)
857
τ – momentum flux (N m−2)
858
859
ACKNOWLEDGEMENTS
860
861
We would like to acknowledge operational, technical, and scientific support provided by
862
NCAR’s Earth Observing Laboratory, sponsored by the National Science Foundation. We also
863
acknowledge the technical and scientific contributions from teams at NOAA, National
864
Ecological Observatory Network (NEON), KIT, University of Wyoming, and University of
865
Wisconsin. We are grateful for the support provided by the US Forest Service and Wisconsin
866
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Educational Communication Board. CHEESEHEAD19 would not have been possible without
867
the contributions from many additional groups and individuals, who are listed in the
868
supplemental materials. This project was financially supported by NSF Award #1822420,
869
Deutsche Forschungsgemeinschaft (DFG) Award #406980118, Wisconsin Alumni Research
870
Foundation WARF UW Fall Competition Award, the Department of Energy American Network
871
Management Project support of the ChEAS core site cluster, and NSF Award #1918850. The
872
cross-lab NOAA contribution was supported by the Weather Program Office in the Office of
873
Oceanic and Atmospheric Research. We acknowledge that this project occurred on the
874
traditional territory of the Ojibwe people.
875
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
REFERENCES
876
877
Anderson, M., J. Norman, W. Kustas, R. Houborg, P. Starks, and N. Agam, 2008: A thermal-
878
based remote sensing technique for routine mapping of land-surface carbon, water and
879
energy fluxes from field to regional scales. Remote Sens. Environ., 112, 4227–4241,
880
https://doi.org/10.1016/j.rse.2008.07.009.
881
Anderson, M. C., and Coauthors, 2012: Mapping daily evapotranspiration at Landsat spatial
882
scales during the BEAREX’08 field campaign. Adv. Water Resour., 50, 162–177,
883
884
Antonarakis, A. S., J. W. Munger, and P. R. Moorcroft, 2014: Imaging spectroscopy- and lidar-
885
derived estimates of canopy composition and structure to improve predictions of forest
886
carbon fluxes and ecosystem dynamics. Geophys. Res. Lett., 41, 2535–2542,
887
https://doi.org/10.1002/2013GL058373.
888
Aubinet, M., and Coauthors, 2010: Direct advection measurements do not help to solve the
889
night-time CO2 closure problem: Evidence from three different forests. Agric. For. Meteor.,
890
150, 655–664, https://doi.org/10.1016/j.agrformet.2010.01.016.
891
Avissar, R., 1995: Scaling of land-atmosphere interactions: An atmospheric modelling
892
perspective. Hydrol. Process., 9, 679–695, https://doi.org/doi:10.1002/hyp.3360090514.
893
Bakwin, P. S., P. P. Tans, D. F. Hurst, and C. Zhao, 1998: Measurements of carbon dioxide on
894
very tall towers: results of the NOAA/CMDL program. Tellus B Chem. Phys. Meteor., 50,
895
401–415, https://doi.org/10.3402/tellusb.v50i5.16216.
896
Barnard, J. C., and C. N. Long, 2004: A Simple Empirical Equation to Calculate Cloud Optical
897
Thickness Using Shortwave Broadband Measurements. J. Appl. Meteor., 43, 1057–1066,
898
https://doi.org/10.1175/1520-0450(2004)043<1057:ASEETC>2.0.CO;2.
899
Barr, J. G., J. D. Fuentes, M. S. DeLonge, T. L. O’Halloran, D. Barr, and J. C. Zieman, 2013:
900
Summertime influences of tidal energy advection on the surface energy balance in a
901
mangrove forest. Biogeosciences, 10, 501–511, https://doi.org/10.5194/bg-10-501-2013.
902
Bertram, T. H., and Coauthors, 2011: A field-deployable, chemical ionization time-of-flight
903
mass spectrometer. Atmos. Meas. Tech., 4, 1471–1479, https://doi.org/10.5194/amt-4-1471-
904
2011.
905
Betts, A. K., J. H. Ball, A. C. M. Beljaars, M. J. Miller, and P. A. Viterbo, 1996: The land
906
surface-atmosphere interaction: A review based on observational and global modeling
907
perspectives. J. Geophys. Res. Atmos., 101, 7209–7225, https://doi.org/10.1029/95JD02135.
908
Beyrich, F., and Coauthors, 2006: Area-Averaged Surface Fluxes Over the Litfass Region Based
909
on Eddy-Covariance Measurements. Bound.-Layer Meteor., 121, 33–65,
910
https://doi.org/10.1007/s10546-006-9052-x.
911
Blumberg, W. G., D. D. Turner, U. Löhnert, and S. Castleberry, 2015: Ground-Based
912
Temperature and Humidity Profiling Using Spectral Infrared and Microwave Observations.
913
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Part II: Actual Retrieval Performance in Clear-Sky and Cloudy Conditions. J. Appl. Meteor.
914
Climatol., 54, 2305–2319, https://doi.org/10.1175/JAMC-D-15-0005.1.
915
Calhoun, R., R. Heap, M. Princevac, R. Newsom, H. Fernando, and D. Ligon, 2006: Virtual
916
Towers Using Coherent Doppler Lidar during the Joint Urban 2003 Dispersion Experiment.
917
J. Appl. Meteor. Climatol., 45, 1116–1126, https://doi.org/10.1175/JAM2391.1.
918
Cammalleri, C., M. C. Anderson, G. Ciraolo, G. D’Urso, W. P. Kustas, G. La Loggia, and M.
919
Minacapilli, 2012: Applications of a remote sensing-based two-source energy balance
920
algorithm for mapping surface fluxes without in situ air temperature observations. Remote
921
Sens. Environ., 124, 502–515, https://doi.org/10.1016/j.rse.2012.06.009.
922
Charney, J. G., 1975: Dynamics of deserts and drought in the Sahel. Q. J. R. Meteor. Soc., 101,
923
193–202, https://doi.org/doi:10.1002/qj.49710142802.
924
Charuchittipan, D., W. Babel, M. Mauder, J.-P. Leps, and T. Foken, 2014: Extension of the
925
Averaging Time in Eddy-Covariance Measurements and Its Effect on the Energy Balance
926
Closure. Bound.-Layer Meteor., 152, 303–327, https://doi.org/10.1007/s10546-014-9922-6.
927
Chen, B., and Coauthors, 2011: Assessing eddy-covariance flux tower location bias across the
928
Fluxnet-Canada Research Network based on remote sensing and footprint modelling. Agric.
929
For. Meteor., 151, 87–100, https://doi.org/10.1016/j.agrformet.2010.09.005.
930
Cook, B. D., and Coauthors, 2004: Carbon exchange and venting anomalies in an upland
931
deciduous forest in northern Wisconsin, USA. Agric. For. Meteor., 126, 271–295,
932
https://doi.org/10.1016/j.agrformet.2004.06.008.
933
Davis, K. J., P. S. Bakwin, C. Yi, B. W. Berger, C. Zhao, R. M. Teclaw, and J. G. Isebrands,
934
2003: The annual cycles of CO2 and H2O exchange over a northern mixed forest as
935
observed from a very tall tower. Glob. Chang. Biol., 9, 1278–1293,
936
https://doi.org/10.1046/j.1365-2486.2003.00672.x.
937
Desai, A. R., P. V. Bolstad, B. D. Cook, K. J. Davis, and E. V. Carey, 2005: Comparing net
938
ecosystem exchange of carbon dioxide between an old-growth and mature forest in the
939
upper Midwest, USA. Agric. For. Meteor., 128, 33–55,
940
https://doi.org/10.1016/j.agrformet.2004.09.005.
941
——, K. J. Davis, C. J. Senff, S. Ismail, E. V. Browell, D. R. Stauffer, and B. P. Reen, 2006: A
942
Case Study on the Effects of Heterogeneous Soil Moisture on Mesoscale Boundary-Layer
943
Structure in the Southern Great Plains, U.S.A. Part I: Simple Prognostic Model. Bound.-
944
Layer Meteor., 119, 195–238, https://doi.org/10.1007/s10546-005-9024-6.
945
Desai, A. R., B. R. Helliker, P. R. Moorcroft, A. E. Andrews, and J. A. Berry, 2010: Climatic
946
controls of interannual variability in regional carbon fluxes from top-down and bottom-up
947
perspectives. J. Geophys. Res. Biogeosciences, 115,
948
https://doi.org/doi:10.1029/2009JG001122.
949
Dürr, B., and R. Philipona, 2004: Automatic cloud amount detection by surface longwave
950
downward radiation measurements. J. Geophys. Res., 109, D05201,
951
https://doi.org/10.1029/2003JD004182.
952
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Eder, F., M. Schmidt, T. Damian, K. Träumner, and M. Mauder, 2015: Mesoscale eddies affect
953
near-surface turbulent exchange: Evidence from lidar and tower measurements. J. Appl.
954
Meteor. Climatol., 54, 189–206, https://doi.org/10.1175/JAMC-D-14-0140.1.
955
Eloranta, E. W., 2005: High spectral resolution lidar. Lidar: Range-Resolved Optical Remote
956
Sensing of the Atmosphere, K. Weitkamp, Ed., Springer-Verlag, 143–163.
957
Engelmann, C., and C. Bernhofer, 2016: Exploring Eddy-Covariance Measurements Using a
958
Spatial Approach: The Eddy Matrix. Bound.-Layer Meteor., 161, 1–17,
959
https://doi.org/10.1007/s10546-016-0161-x.
960
Finnigan, J., 2008: An introduction to flux measurements in difficult conditions. Ecol. Appl., 18,
961
1340–1350, https://doi.org/10.1890/07-2105.1.
962
Emily V. Fischer, Brittany Bloodhart, Kristen Rasmussen, Ilana B. Pollack, Meredith G.
963
Hastings, Erika Marin-Spiotta, Ankur R. Desai, Joshua P. Schwarz, Stephen Nesbitt, and
964
Deanna Hence, 2020. Leveraging Field-Campaign Networks to Effect Collaborative
965
Change on Sexual Harassment, Bulletin of the AMS, submitted.
966
Foken, T., M. Aubinet, J. J. Finnigan, M. Y. Leclerc, M. Mauder, and K. T. Paw U, 2011:
967
Results Of A Panel Discussion About The Energy Balance Closure Correction For Trace
968
Gases. Bull. Amer. Meteor. Soc., 92, ES13–ES18,
969
https://doi.org/10.1175/2011BAMS3130.1.
970
Frank, J. M., W. J. Massman, and B. E. Ewers, 2013: Underestimates of sensible heat flux due to
971
vertical velocity measurement errors in non-orthogonal sonic anemometers. Agric. For.
972
Meteor., 171172, 72–81, https://doi.org/https://doi.org/10.1016/j.agrformet.2012.11.005.
973
Gantner, L., V. Maurer, N. Kalthoff, and O. Kiseleva, 2017: The Impact of Land-Surface
974
Parameter Properties and Resolution on the Simulated Cloud-Topped Atmospheric
975
Boundary Layer. Bound.-Layer Meteor., 165, 475–496, https://doi.org/10.1007/s10546-017-
976
0286-6.
977
Gao, Z., H. Liu, E. S. Russell, J. Huang, T. Foken, and S. P. Oncley, 2016: Large eddies
978
modulating flux convergence and divergence in a disturbed unstable atmospheric surface
979
layer. J. Geophys. Res. Atmos., 121, 1475–1492, https://doi.org/10.1002/2015JD024529.
980
Garratt, J. R., 1993: Sensitivity of Climate Simulations to Land-Surface and Atmospheric
981
Boundary-Layer Treatments-A Review. J. Clim., 6, 419–448, https://doi.org/10.1175/1520-
982
0442(1993)006<0419:SOCSTL>2.0.CO;2.
983
Gerace, A., T. Kleynhans, R. Eon, and M. Montanaro, 2020: Towards an Operational, Split
984
Window-Derived Surface Temperature Product for the Thermal Infrared Sensors Onboard
985
Landsat 8 and 9. Remote Sens., 12, 224, https://doi.org/10.3390/rs12020224.
986
van Heerwaarden, C. C., J. P. Mellado, and A. De Lozar, 2014: Scaling Laws for the
987
Heterogeneously Heated Free Convective Boundary Layer. J. Atmos. Sci., 71, 3975–4000,
988
https://doi.org/10.1175/jas-d-13-0383.1.
989
Hilker, T., F. G. Hall, N. C. Coops, J. G. Collatz, T. A. Black, C. J. Tucker, P. J. Sellers, and N.
990
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Grant, 2013: Remote sensing of transpiration and heat fluxes using multi-angle
991
observations. Remote Sens. Environ., 137, 31–42, https://doi.org/10.1016/j.rse.2013.05.023.
992
Hill, T. C., M. Williams, and J. B. Moncrieff, 2008: Modeling feedbacks between a boreal forest
993
and the planetary boundary layer. J. Geophys. Res. Atmos., 113,
994
https://doi.org/doi:10.1029/2007JD009412.
995
——, ——, F. I. Woodward, and J. B. Moncrieff, 2011: Constraining ecosystem processes from
996
tower fluxes and atmospheric profiles. Ecol. Appl., 21, 1474–1489,
997
https://doi.org/doi:10.1890/09-0840.1.
998
Kampe, T. U., J. McCorkel, L. Hamlin, R. O. Green, K. S. Krause, and B. R. Johnson, 2011:
999
Progress in the development of airborne remote sensing instrumentation for the National
1000
Ecological Observatory Network. W. Gao, T.J. Jackson, J. Wang, and N.-B. Chang, Eds.,
1001
81560A.
1002
Kang, S.-L., K. J. Davis, and M. LeMone, 2007: Observations of the ABL Structures over a
1003
Heterogeneous Land Surface during IHOP_2002. J. Hydrometeor., 8, 221–244,
1004
https://doi.org/10.1175/JHM567.1.
1005
Kattge, J., W. Knorr, T. Raddatz, and C. Wirth, 2009: Quantifying photosynthetic capacity and
1006
its relationship to leaf nitrogen content for global-scale terrestrial biosphere models. Glob.
1007
Chang. Biol., 15, 976–991, https://doi.org/10.1111/j.1365-2486.2008.01744.x.
1008
Klein, P., and Coauthors, 2015: LABLE: A Multi-Institutional, Student-Led, Atmospheric
1009
Boundary Layer Experiment. Bull. Amer. Meteor. Soc., 96, 1743–1764,
1010
https://doi.org/10.1175/BAMS-D-13-00267.1.
1011
Kljun, N., P. Calanca, M. W. Rotach, and H. P. Schmid, 2015: A simple two-dimensional
1012
parameterisation for Flux Footprint Prediction (FFP). Geosci. Model Dev., 8, 3695–3713,
1013
https://doi.org/10.5194/gmd-8-3695-2015.
1014
Klugmann, D., K. Heinsohn, and H. J. Kirtzel, 1996: A low cost 24 GHz FM-CW Doppler radar
1015
rain profiler. Contrib. Atmos. Phys., 61, 247–253.
1016
Kneifel, S., M. Maahn, G. Peters, and C. Simmer, 2011: Observation of snowfall with a low-
1017
power FM-CW K-band radar (Micro Rain Radar). Meteor. Atmos. Phys., 113, 75–87,
1018
https://doi.org/10.1007/s00703-011-0142-z.
1019
Knuteson, R. O., and Coauthors, 2004: Atmospheric Emitted Radiance Interferometer. Part II:
1020
Instrument Performance. J. Atmos. Ocean. Technol., 21, 1777–1789,
1021
https://doi.org/10.1175/JTECH-1663.1.
1022
Kobayashi, H., D. D. Baldocchi, Y. Ryu, Q. Chen, S. Ma, J. L. Osuna, and S. L. Ustin, 2012:
1023
Modeling energy and carbon fluxes in a heterogeneous oak woodland: A three-dimensional
1024
approach. Agric. For. Meteor., 152, 83–100,
1025
https://doi.org/10.1016/j.agrformet.2011.09.008.
1026
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Launiainen, S., and Coauthors, 2016: Do the energy fluxes and surface conductance of boreal
1027
coniferous forests in Europe scale with leaf area? Glob. Chang. Biol., 22, 4096–4113,
1028
https://doi.org/10.1111/gcb.13497.
1029
Lee, T. R., M. Buban, E. Dumas, and C. B. Baker, 2017: A New Technique to Estimate Sensible
1030
Heat Fluxes around Micrometeorological Towers Using Small Unmanned Aircraft Systems.
1031
J. Atmos. Ocean. Technol., 34, 2103–2112, https://doi.org/10.1175/JTECH-D-17-0065.1.
1032
——, ——, ——, and ——, 2019: On the use of rotary-wing aircraft to sample near-surface
1033
thermodynamic fields: results from recent field campaigns. Sensors, 19 (1), 10,
1034
https://doi.org/10.3390/s19010010.
1035
Leuning, R., E. van Gorsel, W. J. Massman, and P. R. Isaac, 2012: Reflections on the surface
1036
energy imbalance problem. Agric. For. Meteor., 156, 65–74,
1037
https://doi.org/https://doi.org/10.1016/j.agrformet.2011.12.002.
1038
Liu, S., and Coauthors, 2016: Upscaling evapotranspiration measurements from multi-site to the
1039
satellite pixel scale over heterogeneous land surfaces. Agric. For. Meteor., 230231, 97–
1040
113, https://doi.org/https://doi.org/10.1016/j.agrformet.2016.04.008.
1041
Liu, S. M., Z. W. Xu, W. Z. Wang, Z. Z. Jia, M. J. Zhu, J. Bai, and J. M. Wang, 2011: A
1042
comparison of eddy-covariance and large aperture scintillometer measurements with respect
1043
to the energy balance closure problem. Hydrol. Earth Syst. Sci., 15, 1291–1306,
1044
https://doi.org/10.5194/hess-15-1291-2011.
1045
Löhnert, U., D. D. Turner, and S. Crewell, 2009: Ground-Based Temperature and Humidity
1046
Profiling Using Spectral Infrared and Microwave Observations. Part I: Simulated Retrieval
1047
Performance in Clear-Sky Conditions. J. Appl. Meteor. Climatol., 48, 1017–1032,
1048
https://doi.org/10.1175/2008JAMC2060.1.
1049
Long, C. N., and T. P. Ackerman, 2000: Identification of clear skies from broadband
1050
pyranometer measurements and calculation of downwelling shortwave cloud effects. J.
1051
Geophys. Res. Atmos., 105, 15609–15626, https://doi.org/10.1029/2000JD900077.
1052
Long, C. N., and Y. Shi, 2008: An Automated Quality Assessment and Control Algorithm for
1053
Surface Radiation Measurements. Open Atmos. Sci. J., 2, 23–37,
1054
https://doi.org/10.2174/1874282300802010023.
1055
——, and D. D. Turner, 2008: A method for continuous estimation of clear-sky downwelling
1056
longwave radiative flux developed using ARM surface measurements. J. Geophys. Res.,
1057
113, D18206, https://doi.org/10.1029/2008JD009936.
1058
——, T. P. Ackerman, K. L. Gaustad, and J. N. S. Cole, 2006: Estimation of fractional sky cover
1059
1060
https://doi.org/10.1029/2005JD006475.
1061
Mahrt, L., 2000: Surface Heterogeneity and Vertical Structure of the Boundary Layer. Bound.-
1062
Layer Meteor., 96, 33–62.
1063
Maronga, B., and Coauthors, 2015: The Parallelized Large-Eddy Simulation Model (PALM)
1064
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
version 4.0 for atmospheric and oceanic flows: model formulation, recent developments,
1065
and future perspectives. Geosci. Model Dev., 8, 2515–2551, https://doi.org/10.5194/gmd-8-
1066
2515-2015.
1067
Maronga, B., and Coauthors, 2020: Overview of the PALM model system 6.0. Geosci. Model
1068
Dev., 13, 1335–1372, https://doi.org/10.5194/gmd-13-1335-2020.
1069
Mauder, M., R. L. Desjardins, and I. Macpherson, 2007a: Scale analysis of airborne flux
1070
measurements over heterogeneous terrain in a boreal ecosystem. J. Geophys. Res., 112,
1071
D13112, https://doi.org/10.1029/2006JD008133.
1072
Mauder, M., O. O. Jegede, E. C. Okogbue, F. Wimmer, and T. Foken, 2007b: Surface energy
1073
balance measurements at a tropical site in West Africa during the transition from dry to wet
1074
season. Theor. Appl. Climatol., 89, 171–183, https://doi.org/10.1007/s00704-006-0252-6.
1075
Mauder, M., R. L. Desjardins, and I. Macpherson, 2008a: Creating Surface Flux Maps from
1076
Airborne Measurements: Application to the Mackenzie Area GEWEX Study MAGS 1999.
1077
Bound.-Layer Meteor., 129, 431–450, https://doi.org/10.1007/s10546-008-9326-6.
1078
Mauder, M., R. L. Desjardins, E. Pattey, Z. Gao, and R. van Haarlem, 2008b: Measurement of
1079
the Sensible Eddy Heat Flux Based on Spatial Averaging of Continuous Ground-Based
1080
Observations. Bound.-Layer Meteor., 128, 151–172, https://doi.org/10.1007/s10546-008-
1081
9279-9.
1082
Mcgloin, R., L. Šigut, K. Havránková, J. Dušek, M. Pavelka, and P. Sedlák, 2018: Energy
1083
balance closure at a variety of ecosystems in Central Europe with contrasting topographies.
1084
Agric. For. Meteor., 248, 418–431, https://doi.org/10.1016/j.agrformet.2017.10.003.
1085
Metzger, S., Durden, D., Paleri, S., Sühring, M., Butterworth, B., Desai, A., Florian, C., Mauder,
1086
M., Wanner, L., and Xu, K.: Numerical experiment design doubles scientific return of
1087
surface-atmosphere synthesis, in preparation.
1088
Metzger, S., 2018: Surface-atmosphere exchange in a box: Making the control volume a suitable
1089
representation for in-situ observations. Agric. For. Meteor., 255, 68–80,
1090
https://doi.org/10.1016/j.agrformet.2017.08.037.
1091
Metzger, S., and Coauthors, 2013: Spatially explicit regionalization of airborne flux
1092
measurements using environmental response functions. Biogeosciences, 10, 2193–2217,
1093
https://doi.org/10.5194/bg-10-2193-2013.
1094
Nakai, T., H. Iwata, Y. Harazono, and M. Ueyama, 2014: An inter-comparison between Gill and
1095
Campbell sonic anemometers. Agric. For. Meteor., 195196, 123–131,
1096
https://doi.org/10.1016/j.agrformet.2014.05.005.
1097
Newman, A. J., P. A. Kucera, and L. F. Bliven, 2009: Presenting the Snowflake Video Imager
1098
(SVI). J. Atmos. Ocean. Technol., 26, 167–179,
1099
https://doi.org/10.1175/2008JTECHA1148.1.
1100
Niinemets, Ü., 2001: Global-scale Climatic Controls of Leaf Dry Mass per Area, Density, and
1101
Thickness in Trees and Shrubs. Ecology, 82, 453–469, https://doi.org/10.1890/0012-
1102
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
9658(2001)082[0453:GSCCOL]2.0.CO;2.
1103
Niple, E. R., H. E. Scott, J. A. Conant, S. H. Jones, F. J. Iannarilli, and W. E. Pereira, 2016:
1104
Application of oxygen A-band equivalent width to disambiguate downwelling radiances for
1105
cloud optical depth measurement. Atmos. Meas. Tech., 9, 4167–4179,
1106
https://doi.org/10.5194/amt-9-4167-2016.
1107
Novak, G. A., M. P. Vermeuel, and T. H. Bertram, 2020: Simultaneous detection of ozone and
1108
nitrogen dioxide by oxygen anion chemical ionization mass spectrometry: a fast-time-
1109
response sensor suitable for eddy covariance measurements. Atmos. Meas. Tech., 13, 1887–
1110
1907, https://doi.org/10.5194/amt-13-1887-2020.
1111
Oncley, S. P., and Coauthors, 2007: The Energy Balance Experiment EBEX-2000. Part I:
1112
overview and energy balance. Bound.-Layer Meteor., 123, 1–28,
1113
https://doi.org/10.1007/s10546-007-9161-1.
1114
Peters, G., B. Fischer, and T. Andersson, 2002: Rain observations with a vertically looking Micro
1115
Rain Radar (MRR). Boreal Environ. Res., 7, 353–362, http://www.borenv.net/BER/pdfs/
1116
ber7/ber7-353.pdf.
1117
Pettersen, C., M. S. Kulie, L. F. Bliven, A. J. Merrelli, W. A. Petersen, T. J. Wagner, D. B.
1118
Wolff, and N. B. Wood, 2020a: A Composite Analysis of Snowfall Modes from Four
1119
Winter Seasons in Marquette, Michigan. J. Appl. Meteor. Climatol., 59, 103–124,
1120
https://doi.org/10.1175/JAMC-D-19-0099.1.
1121
Pettersen, C.; Bliven, L.F.; von Lerber, A.; Wood, N.B.; Kulie, M.S.; Mateling, M.E.; Moisseev,
1122
D.M.; Munchak, S.J.; Petersen, W.A.; Wolff, D.B., 2020b: Introducing the Precipitation
1123
Imaging Package: Assessment of microphysical and bulk characteristics of snow.
1124
Atmosphere, in review.
1125
Pielke, R. A., Sr, R. Avissar, M. Raupach, A. J. Dolman, X. Zeng, and A. S. Denning, 1998:
1126
Interactions between the atmosphere and terrestrial ecosystems: influence on weather and
1127
climate. Glob. Chang. Biol., 4, 461–475, https://doi.org/doi:10.1046/j.1365-2486.1998.t01-
1128
1-00176.x.
1129
Platis, A., A. F. Moene, D. M. Villagrasa, F. Beyrich, D. Tupman, and J. Bange, 2017:
1130
Observations of the Temperature and Humidity Structure Parameter Over Heterogeneous
1131
Terrain by Airborne Measurements During the LITFASS-2003 Campaign. Bound.-Layer
1132
Meteor., 165, 447–473, https://doi.org/10.1007/s10546-017-0290-x.
1133
Poulos, G. S., and Coauthors, 2002: CASES-99: A Comprehensive Investigation of the Stable
1134
Nocturnal Boundary Layer. Bull. Amer. Meteor. Soc., 83, 555–581,
1135
https://doi.org/10.1175/1520-0477(2002)083<0555:CACIOT>2.3.CO;2.
1136
Raasch, S., and M. Schröter, 2001: PALM - A large-eddy simulation model performing on
1137
massively parallel computers. Meteor. Zeitschrift, 10, 363–372,
1138
https://doi.org/10.1127/0941-2948/2001/0010-0363.
1139
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Ran, Y., X. Li, R. Sun, N. Kljun, L. Zhang, X. Wang, and G. Zhu, 2016: Spatial
1140
representativeness and uncertainty of eddy covariance carbon flux measurements for
1141
upscaling net ecosystem productivity to the grid scale. Agric. For. Meteor., 230231, 114–
1142
127, https://doi.org/https://doi.org/10.1016/j.agrformet.2016.05.008.
1143
Raupach, M. R., and J. J. Finnigan, 1995: Scale issues in boundary-layer meteorology: Surface
1144
energy balances in heterogeneous terrain. Hydrol. Process., 9, 589–612,
1145
https://doi.org/doi:10.1002/hyp.3360090509.
1146
Reen, B. P., D. R. Stauffer, and K. J. Davis, 2014: Land-Surface Heterogeneity Effects in the
1147
Planetary Boundary Layer. Bound.-Layer Meteor., 150, 1–31,
1148
https://doi.org/10.1007/s10546-013-9860-8.
1149
Reichstein, M., G. Camps-Valls, B. Stevens, M. Jung, J. Denzler, N. Carvalhais, and Prabhat,
1150
2019: Deep learning and process understanding for data-driven Earth system science.
1151
Nature, 566, 195–204, https://doi.org/10.1038/s41586-019-0912-1.
1152
Rihani, J. F., F. K. Chow, and R. M. Maxwell, 2015: Isolating effects of terrain and soil moisture
1153
heterogeneity on the atmospheric boundary layer: Idealized simulations to diagnose land-
1154
atmosphere feedbacks. J. Adv. Model. Earth Syst., 7, 915–937,
1155
1156
Sellers, P., and Coauthors, 1995: The Boreal Ecosystem–Atmosphere Study (BOREAS): An
1157
Overview and Early Results from the 1994 Field Year. Bull. Amer. Meteor. Soc., 76, 1549–
1158
1577, https://doi.org/10.1175/1520-0477(1995)076<1549:TBESAO>2.0.CO;2.
1159
Serbin, S. P., A. Singh, B. E. McNeil, C. C. Kingdon, and P. A. Townsend, 2014: Spectroscopic
1160
determination of leaf morphological and biochemical traits for northern temperate and
1161
boreal tree species. Ecol. Appl., 24, 1651–1669, https://doi.org/10.1890/13-2110.1.
1162
Steinfeld, G., M. O. Letzel, S. Raasch, M. Kanda, and A. Inagaki, 2007: Spatial
1163
representativeness of single tower measurements and the imbalance problem with eddy-
1164
covariance fluxes: results of a large-eddy simulation study. Bound.-Layer Meteor., 123, 77–
1165
98, https://doi.org/10.1007/s10546-006-9133-x.
1166
——, S. Raasch, and T. Markkanen, 2008: Footprints in Homogeneously and Heterogeneously
1167
Driven Boundary Layers Derived from a Lagrangian Stochastic Particle Model Embedded
1168
into Large-Eddy Simulation. Bound.-Layer Meteor., 129, 225–248,
1169
https://doi.org/10.1007/s10546-008-9317-7.
1170
Stoy, P. C., and Coauthors, 2013: A data-driven analysis of energy balance closure across
1171
FLUXNET research sites: The role of landscape scale heterogeneity. Agric. For. Meteor.,
1172
171172, 137–152, https://doi.org/10.1016/j.agrformet.2012.11.004.
1173
Sulman, B. N., A. R. Desai, B. D. Cook, N. Saliendra, and D. S. Mackay, 2009: Contrasting
1174
carbon dioxide fluxes between a drying shrub wetland in Northern Wisconsin, USA, and
1175
nearby forests. Biogeosciences, 6, 1115–1126, https://doi.org/10.5194/bg-6-1115-2009.
1176
Turner, D. D., and W. G. Blumberg, 2019: Improvements to the AERIoe Thermodynamic Profile
1177
Retrieval Algorithm. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 12, 1339–1354,
1178
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
https://doi.org/10.1109/JSTARS.2018.2874968.
1179
Turner, D. D., and U. Löhnert, 2014: Information Content and Uncertainties in Thermodynamic
1180
1181
Emitted Radiance Interferometer (AERI). J. Appl. Meteor. Climatol., 53, 752–771,
1182
https://doi.org/10.1175/JAMC-D-13-0126.1.
1183
Twine, T. E., and Coauthors, 2000: Correcting eddy-covariance flux underestimates over a
1184
grassland. Agric. For. Meteor., 103, 279–300, https://doi.org/https://doi.org/10.1016/S0168-
1185
1923(00)00123-4.
1186
Vogelmann, H., Speidel, J., and Perfahl, M.: Laser concept of the mobile ATMONSYS-lidar and
1187
its application during CHEESEHEAD, EGU General Assembly 2020,
1188
https://doi.org/10.5194/egusphere-egu2020-16517
1189
Wagner, T. J., P. M. Klein, and D. D. Turner, 2019: A New Generation of Ground-Based Mobile
1190
Platforms for Active and Passive Profiling of the Boundary Layer. Bull. Amer. Meteor. Soc.,
1191
100, 137–153, https://doi.org/10.1175/BAMS-D-17-0165.1.
1192
Wang, J., J. Zhuang, W. Wang, S. Liu, and Z. Xu, 2015: Assessment of Uncertainties in Eddy
1193
Covariance Flux Measurement Based on Intensive Flux Matrix of HiWATER-MUSOEXE.
1194
IEEE Geosci. Remote Sens. Lett., 12, 259–263,
1195
https://doi.org/10.1109/LGRS.2014.2334703.
1196
Wang, Y. Q., Y. J. Xiong, G. Y. Qiu, and Q. T. Zhang, 2016: Is scale really a challenge in
1197
evapotranspiration estimation? A multi-scale study in the Heihe oasis using thermal remote
1198
sensing and the three-temperature model. Agric. For. Meteor., 230231, 128–141,
1199
https://doi.org/https://doi.org/10.1016/j.agrformet.2016.03.012.
1200
Wang, Z., P. Wechsler, W. Kuestner, J. French, A. Rodi, B. Glover, M. Burkhart, and D. Lukens,
1201
2009: Wyoming Cloud Lidar: instrument description and applications, Optics Express Vol.
1202
17, Iss. 16, pp. 13576–13587.
1203
Wang, Z., A. Chlus, R. Geygan, Z. Ye, T. Zheng, A. Singh, J.J. Couture, J. Cavender-Bares, E.L.
1204
Kruger and P.A. Townsend, accepted: Foliar functional traits from imaging spectroscopy
1205
across biomes in eastern North America. New Phytol.
1206
Wisconsin Department of Natural Resources. (2019). Wisconsin Wiscland 2 Land Cover
1207
1208
Wolf, B., and Coauthors, 2017: The SCALEX Campaign: Scale-Crossing Land Surface and
1209
Boundary Layer Processes in the TERENO-preAlpine Observatory. Bull. Amer. Meteor.
1210
Soc., 98, 1217–1234, https://doi.org/10.1175/BAMS-D-15-00277.1.
1211
Wu, D., Z. Wang, P. Wechsler, N. Mahon, M. Deng, B. Glover, M. Burkhart, W. Kuestner, and
1212
B. Heesen, 2016: Airborne compact rotational Raman lidar for temperature measurement,
1213
Opt. Express, 24, A1210-A1223.
1214
Wu, P., H. Shen, T. Ai, and Y. Liu, 2013: Land-surface temperature retrieval at high spatial and
1215
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
temporal resolutions based on multi-sensor fusion. Int. J. Digit. Earth, 6, 113–133,
1216
https://doi.org/10.1080/17538947.2013.783131.
1217
Wulfmeyer, V., S. K. Muppa, A. Behrendt, E. Hammann, F. Späth, Z. Sorbjan, D. D. Turner, and
1218
R. M. Hardesty, 2016: Determination of Convective Boundary Layer Entrainment Fluxes,
1219
Dissipation Rates, and the Molecular Destruction of Variances: Theoretical Description and
1220
a Strategy for Its Confirmation with a Novel Lidar System Synergy. J. Atmos. Sci., 73, 667–
1221
692, https://doi.org/10.1175/JAS-D-14-0392.1.
1222
Wulfmeyer, V., and Coauthors, 2018: A New Research Approach for Observing and
1223
Characterizing Land–Atmosphere Feedback. Bull. Amer. Meteor. Soc., 99, 1639–1667,
1224
https://doi.org/10.1175/BAMS-D-17-0009.1.
1225
Xu, K., S. Metzger, and A. R. Desai, 2017a: Upscaling tower-observed turbulent exchange at
1226
fine spatio-temporal resolution using environmental response functions. Agric. For.
1227
Meteor., 232, 10–22, https://doi.org/10.1016/j.agrformet.2016.07.019.
1228
——, ——, and A. R. Desai, 2018: Surface-atmosphere exchange in a box: Space-time resolved
1229
storage and net vertical fluxes from tower-based eddy covariance. Agric. For. Meteor., 255,
1230
81–91, https://doi.org/10.1016/j.agrformet.2017.10.011.
1231
Xu, Z., Y. Ma, S. Liu, W. Shi, and J. Wang, 2017b: Assessment of the Energy Balance Closure
1232
under Advective Conditions and Its Impact Using Remote Sensing Data. J. Appl. Meteor.
1233
Climatol., 56, 127–140, https://doi.org/10.1175/JAMC-D-16-0096.1.
1234
Xu, K., M. Sühring, S. Metzger, D. Durden, and A. R. Desai, 2020: Can Data Mining Help Eddy
1235
Covariance See the Landscape? A Large-Eddy Simulation Study. Bound.-Layer Meteor.,
1236
176, 85–103, https://doi.org/10.1007/s10546-020-00513-0.
1237
Zhu, X., G. Ni, Z. Cong, T. Sun, and D. Li, 2016: Impacts of surface heterogeneity on dry
1238
planetary boundary layers in an urban-rural setting. J. Geophys. Res. Atmos., 121, 164–179,
1239
https://doi.org/10.1002/2016JD024982.
1240
Zitouna-Chebbi, R., L. Prévot, F. Jacob, R. Mougou, and M. Voltz, 2012: Assessing the
1241
consistency of eddy covariance measurements under conditions of sloping topography
1242
within a hilly agricultural catchment. Agric. For. Meteor., 164, 123–135,
1243
https://doi.org/10.1016/j.agrformet.2012.05.010.
1244
1245
1246
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
SUPPLEMENTAL MATERIAL
1247
1248
1249
1250
Fig. S1. The (a) DJI S-1000 and (b) Meteodrone SSE.
1251
1252
Table S1. List of the flux towers in operation during the CHEESEHEAD19 field campaign.
Site
#
Site
Name
Ameriflu
x ID
Latitude
(° N)
Longitud
e
(° W)
Tower
height
(m)
Canopy height
(m)
Vegetation
Contact
1
NW1
US-PFb
45.97200
90.32317
32
25
pine
Oncley
2
NW2
US-PFc
45.96773
90.30878
12
3
aspen
Oncley
3
NW3
US-PFd
45.96892
90.30103
3
0.3
wetland
Oncley
4
NW4
US-PFe
45.97925
90.30042
32
20.1
lake
Oncley
5
NW5
US-PFf
45.94583
90.29437
2
0
grass
Stoy
6
NE1
US-PFg
45.97348
90.27230
32
33.2
pine
Oncley
7
NE2
US-PFh
45.95573
90.24060
32
19.2
pine
Oncley
8
NE3
US-PFi
45.97490
90.23273
32
18.3
hardwood
Oncley
9
NE4
US-PFj
45.96187
90.22703
32
18.3
maple
Oncley
10
SW1
US-PFk
45.91490
90.34250
32
24.4
aspen
Oncley
11
SW2
US-PFl
45.94090
90.31773
25
19.2
aspen
Oncley
12
SW3
US-PFm
45.92067
90.30990
32
15
hardwood
Oncley
13
SW4
US-PFn
45.93922
90.28232
32
25.9
hardwood
Oncley
14
SE1
US-PFo
45.92288
90.27283
1.5
0
lake
Stoy
15
SE2
US-PFp
45.93652
90.26408
32
24.4
hardwood
Oncley
16
SE3
US-PFq
45.92715
90.24750
32
14.3
aspen
Oncley
17
SE4
US-PFr
45.92448
90.24745
3
0.3
wetland
Oncley
18
SE5
US-PFs
45.93808
90.23818
12
3.1
aspen
Oncley
19
SE6
US-PFt
45.91973
90.22883
32
21.6
pine
Oncley
20
WLEF
US-PFa
45.94590
90.27230
396
n/a
mixed
Desai
21
WCR
US-WCr
45.80600
90.07980
30
24
hardwood
Desai
22
LOS
US-Los
46.08270
89.97920
10
2
wetland
Desai
1253
1254
1255
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Table S2. Instrumentation deployed at the CHEESEHEAD19 campaign by the NOAA Global
1256
Monitoring Laboratory
1257
Instrument
Measurement
Dates (2019)
Resolution
Central ISS Station
Eppley Pyrgeometer
06/29 - 10/22
1 min avg.
Eppley Precision Spectral
Pyranometer (PSP)
06/29 - 10/22
1 min avg.
Pyranometer
06/29 - 10/22
1 min avg.
Eppley Normal Incidence
Pyrheliometer (NIP)
06/29 - 10/22
1 min avg.
Total Sky Imager (TSI)
Images/movies of sky cover, fractional sky cover
07/05 - 10/22
15 sec
Vaisala CL51 Ceilometer
Cloud base height, boundary layer height
06/29 - 10/22
16 sec
Multi Filter Rotating
(MFRSR)
hemispheric total and diffuse spectral irradiance
at 6 bands: 415, 500, 670, 868, 940, 1625 nm;
retrievals of aerosol optical depth
06/29 - 10/22
20 sec
hemispheric total spectral irradiance at 6 bands:
415, 500, 670, 868, 940, 1625 nm. Spectral surface
albedo and NDVI (with MFRSR).
06/29 - 10/22
20 sec
LICOR Quantum 190R
06/29 - 10/22
1 min avg.
Aerodyne Three-Waveband
Spectrally-agile Technique
(TWST)
Cloud optical depth, spectral SW zenith radiance
(350-1000 nm, ~2.5 nm resolution)
09/20 - 10/22
1 sec
Vaisala HMP60
Temperature and Relative Humidity
06/29 - 10/22
1 min avg.
RM Young, Model 05103
Wind direction and speed at 10 m
06/29 - 10/22
1 min avg.
Prentice and Lakeland Airports
Eppley Pyrgeometer
06/28 - 10/23
1 min avg.
Kipp & Zonen CMP11
Pyranometer
06/28 - 10/23
1 min avg.
SW diffuse and total broadband hemispheric
06/28 - 10/23
1 min avg.
Vaisala HMP60
Temperature and Relative Humidity
06/28 - 10/23
1 min avg.
Vaisala CL-51 Ceilometer
Cloud Base Height (CBH)
06/28 - 10/23
16 sec
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
1258
1259
Table S3. The dates, times, and flight patterns of the UWKA flights.
Date
Flight
Number
Takeoff
(UTC)
Landing
(UTC)
Entered
Domain
(UTC)
Exited
Domain (UTC)
Flight
Pattern
7/9/19
RF01
13:57
16:47
14:06
16:16
WE2
7/9/19
RF02
19:02
21:32
19:11
21:22
WE2
7/11/19
RF03
14:10
17:00
14:22
16:36
WE1
7/11/19
RF04
19:00
21:40
19:15
21:28
WE1
7/12/19
RF05
13:40
16:45
13:49
16:06
WE2
7/12/19
RF06
17:52
21:00
18:04
20:46
WE2
7/13/19
RF07
14:05
16:52
14:22
16:32
SE2
7/13/19
RF08
18:56
21:30
19:12
21:16
SW1
8/20/19
RF09
13:40
16:23
13:51
16:12
SE1
8/20/19
RF10
19:12
22:22
19:23
21:51
SE1
8/21/19
RF11
13:54
16:50
14:08
16:36
SW1
8/21/19
RF12
18:55
21:50
19:11
21:38
SW1
8/22/19
RF13
13:57
17:15
14:11
16:55
SW2
8/22/19
RF14
19:00
22:01
19:13
21:46
SW2
8/23/19
RF15
13:57
16:48
14:07
16:38
WE2
8/23/19
RF16
19:07
22:03
19:17
21:46
WE2
9/24/19
RF17
13:37
17:00
13:53
16:24
SE1
9/24/19
RF18
18:57
21:49
19:10
21:39
SE1
9/25/19
RF19
14:20
17:22
14:41
17:09
SW1
9/25/19
RF20
19:12
22:06
19:29
21:53
SW1
9/26/19
RF21
13:52
16:46
14:05
16:35
SE1
9/26/19
RF22
18:31
21:40
18:45
21:14
SE1
9/28/19
RF23
14:14
17:30
14:37
17:17
WE1
9/28/19
RF24
18:50
21:50
19:07
21:36
WE1
1260
1261
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
1262
Fig S2. Preliminary profiles of (a) H2O and (b) backscatter measured by the ATMONSYS
1263
lidar. These are complementary, collocated datasets with the HSRL, AERI, and MWR at
1264
WLEF. The high resolution data (vertical resolution of 110 m and 7.5 m for H2O and
1265
backscatter, respectively; 20 second temporal resolution for both) are capable of being
1266
combined with collocated Doppler wind lidar data to calculate flux profiles.
1267
1268
1269
ACKNOWLEDGEMENTS
1270
1271
We thank the many groups and individuals whose contributions made this project possible. From
1272
the US Forest Service we thank Linda Parker, Melanie Fullman, Jonathan McNeill, and Jim
1273
Mineau. We thank Linda Cully at NCAR EOL for creating the project web page and data
1274
repository. Thank you to the colleagues who helped deploy the NOAA instruments: Ed Dumas,
1275
This work has been submitted to the Bulletin of the American Meteorological Society.
Copyright in this work may be transferred without further notice.
Gary Hodges, Emiel Hall, Christian Herrera, Hagen Telg, Jim Wendell, Herman Scott, Irina
1276
Djalalova, Laura Bianco, Tom Ayers, Matt Carney, Doug Kennedy, Sean Waugh, Petra Klein,
1277
and Tyler Bell. Thank you to the UWKA support crew: Matt Burkhart, Min Deng,
1278
Tom Drew, Brent Glover, Zane Little, Austin Morgan, Larry Oolman, Ed Sigel, and
1279
Brett Wadsworth. We thank the UW administrative, communications, and technical support from
1280
Wayne Feltz, Brad Pierce, Jenny Hackel, Chelsea Dahmen, Christi Levenson, Sue Foldy, Bailey
1281
Murphy, Leo Mikula, Ryan Clare, Kelly Tyrell, and Jeff Miller. Thank you to the team from
1282
UW-Madison FWE who assisted in the field work, processing, and administrative activities,
1283
including Jacob May in lidar drone deployment, field sampling by Ella Norris, Ben Sellers, Sam
1284
Jaeger, Nanfeng Liu, Ben Spaier, Josh Phillips; and airborne data collection and processing Erin
1285
Hokanson Wagner, Brendan Heberlein, and Nanfeng Liu. Thank you to Karla Ortman and Scott
1286
Bowe at Kemp Natural Resources Station for assisting with lodging needs of the project. We
1287
appreciate the on-going support of Jeff Ayers, Steve Bauder, Jeff Ohnstad, Doug Siroin, and
1288
Marta Bechtol of the Wisconsin Educational Communication Board who own and operate the
1289
WLEF tall tower. Thank you to Noah R. Lottig, Paul Schram, and Emily Stanley of the UW
1290
Center for Limnology for providing equipment and processing water pCO2 samples. Thank you
1291
to Matthias Perfahl of KIT for technical operation of the ATMONSYS lidar, Gabe Bromley for
1292
EC installation, and Ke Xu for project lead-up research. Thank you to UW-Eau Claire students
1293
Josie Radtke and Whitney Mottishaw for technical support. We also thank those involved in the
1294
UW-Madison AOS404 class: Susi Weisner, Jess Turner, Sophie Hoffman, Juliet Pilewskie, Iman
1295
Nasif, Peter Janssen, as well as April Hiscox and the U South Carolina class. Thank you to
1296
Laurie Fox and the students at Butternut School and Travis Augustine and the students at
1297
Chequamegon High School Class ACT charter who helped collect data. We acknowledge
1298
Christopher Bosma, who helped deploy the MRRPro. Thank you to Tania Kleynhans and Aaron
1299
Gerace for processing Landsat 8 LST data. Thank you to the NOAA Carbon Cycle and
1300
Greenhouse Gas group: Arlyn Andrews, Jonathan Kofler, Colm Sweeney, and Issac Vimont.
1301
Thank you to the group operating TCCON: Paul Wennberg and Debra Wunch. Thank you to the
1302
team from NASA: Joel McCorkel, Eric Vermote, and Bill Rountree. We thank Ann Wojcieszak
1303
for allowing instruments to be installed on her property (the ISS field). We thank Brittany
1304
Bloodhart, Emily Fischer, Erika Marin-Spiotta for organizing the fieldwork surveys and bullying
1305
and sexual harassment prevention training. Lastly, we thank our friends at Park Falls Gastropub -
1306
the official pub of the CHEESEHEAD19 project.
1307
... However, in order to fully complement these forthcoming approaches to the scales of NWP models, spatially distributed sensors over a larger experimental domain may be required. Experiments such as the Chequamegon Heterogeneous Ecosystem Energy-balance Study Enabled by a High-density Extensive Array of Detectors (CHEESEHEAD) (Butterworth et al. 2020(Butterworth et al. , 2021, which captured data over O 10 km scales with a large array of towers over a forested canopy, may be the step forward in understanding this issue at larger scales. Lastly, expansion of similar three-dimensional analysis over a variety of surfaces may shed light on energy-balance residuals over vegetated surfaces or within vegetated canopies, where energy storage and moisture fluxes may be more relevant (Oncley et al. 2007). ...
Article
Full-text available
Data from the Idealized Planar-Array experiment for Quantifying Spatial heterogeneity are used to perform a control volume analysis (400 ×\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\times$$\end{document} 400 ×\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\times$$\end{document} 2 m3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$^3$$\end{document}) on the total derivative of the temperature tendency equation. Analysis of the heat-flux imbalance, which is defined as the ratio of the sum of advective, dispersive, and turbulence-flux terms to the turbulence-flux term, are presented. Results are divided amongst free-convective and forced-convective days, as well as high-wind-speed and quiescent nocturnal periods. Findings show that the median flux imbalance is greater on forced-convective days (a 168% turbulence-flux overestimation, or relative importance of the advection to dispersive flux to the turbulence flux) when compared to free-convective periods (79% turbulence-flux overestimation). During nocturnal periods, a median turbulence-flux underestimation of 146% exists for quiescent nights and a 43% underestimation of the flux for high-wind-speed nights. These results support the existing literature, suggesting that mean air-temperature heterogeneities lead to strong bulk advection and dispersive fluxes. A discussion of the impact of the flux imbalance on the surface energy balance and numerical-weather-prediction modelling is presented.
... This EBC is not ideal but close to the typical range EC of studies globally, approximately 80% (Foken, 2008;Twine et al., 2000;Wilson et al., 2002). Lack of EBC at the half-hourly scale could result from multiple sources, including not resolving mesoscale eddies, errors in radiation measurements, and heat storage in biomass (Butterworth et al., 2020). Since the EC community has not widely adopted corrections for carbon fluxes due to lack of EBC, we report our results here solely for transparency. ...
Article
Full-text available
Bioenergy has been identified as a key component of climate change mitigation. Therefore, quantifying the net carbon balance of bioenergy feedstocks is crucial for accurate projections of climate mitigation benefits. Switchgrass (Panicum virgatum) has many characteristics of an ideal bioenergy crop with high yields, low maintenance, and deep roots with potential for below ground carbon sequestration. However, assessments of net annual carbon exchange between switchgrass fields and the atmosphere are rare. Here we present observations of net carbon fluxes in a minimally managed switchgrass field in Virginia (Ameriflux site US‐SB2) over 5 years (three to seven years since establishment). Average annual net ecosystem exchange (NEE) of carbon was near zero (60 g C m−2 yr−1) but the net ecosystem carbon balance (NECB) that includes harvested carbon was a net source of carbon to the atmosphere (313 g C m−2 yr−1). The field alternated between a large and small source of carbon annually, with the inter‐annual variability most strongly correlated with day of the last frost and the interaction of temperature and precipitation. Overall, the consistent source of carbon to the atmosphere at US‐SB2 differs substantially from other EC studies that report switchgrass fields to be either neutral or a sink of carbon when accounting for both NEE and harvested carbon. This study illustrates that predictions of net carbon climate benefits from bioenergy crops can not assume that the ecosystem will be a net sink of carbon from the atmosphere. Background climate, management, and land‐use history may determine whether widespread deployment of switchgrass as a bioenergy feedstock results in realized climate change mitigation.
Article
Full-text available
Remote-sensing observations are needed to estimate the regional and global impacts of snow. However, to retrieve accurate estimates of snow mass and rate, these observations require augmentation through additional information and assumptions about hydrometeor properties. The Precipitation Imaging Package (PIP) provides information about precipitation characteristics and can be utilized to improve estimates of snowfall rate and accumulation. Here, the goal is to demonstrate the quality and utility of two higher-order PIP-derived products: liquid water equivalent snow rate and an approximation of volume-weighted density called equivalent density. Accuracy of the PIP snow rate and equivalent density is obtained through intercomparison with established retrieval methods and through evaluation with colocated ground-based observations. The results confirm the ability of the PIP-derived products to quantify properties of snow rate and equivalent density, and demonstrate that the PIP produces physically realistic snow characteristics. When compared to the National Weather Service (NWS) snow field measurements of six-hourly accumulation, the PIP-derived accumulations were biased only +2.48% higher. Additionally, this work illustrates fundamentally different microphysical and bulk features of low and high snow-to-liquid ratio events, through assessment of observed particle size distributions, retrieved mass coefficients, and bulk properties. Importantly, this research establishes the role that PIP observations and higher-order products can serve for constraining microphysical assumptions in ground-based and spaceborne remotely sensed snowfall retrievals.
Article
Full-text available
Foliar functional traits are widely used to characterize leaf and canopy properties that drive ecosystem processes and to infer physiological processes in Earth system models. Imaging spectroscopy provides great potential to map foliar traits to characterize continuous functional variation and diversity, but few studies have demonstrated consistent methods for mapping multiple traits across biomes. With airborne imaging spectroscopy data and field data from 19 sites, we developed trait models using partial least squares regression, and mapped 26 foliar traits in 7 NEON (National Ecological Observatory Network) ecoregions (domains) including temperate and subtropical forests and grasslands of eastern North America. Model validation accuracy varied among traits (normalized RMSE, 9.1‐19.4%; R2, 0.28‐0.82), with phenolic concentration, leaf mass per area and equivalent water thickness performing best across domains. Across all trait maps, 90% of vegetated pixels had reasonable values for one trait, and 28‐81% provided high confidence for multiple traits concurrently. Maps of 26 traits and their uncertainties for eastern U.S. NEON sites are available for download, and are being expanded to the western U.S. and tundra/boreal zone. These data enable better understanding of trait variations and relationships over large areas, calibration of ecosystem models, and assessment of continental‐scale functional diversity.
Article
Full-text available
We report on the development, characterization, and field deployment of a fast-time-response sensor for measuring ozone (O3) and nitrogen dioxide (NO2) concentrations utilizing chemical ionization time-of-flight mass spectrometry (CI-ToFMS) with oxygen anion (O2-) reagent ion chemistry. We demonstrate that the oxygen anion chemical ionization mass spectrometer (Ox-CIMS) is highly sensitive to both O3 (180 counts s−1 pptv−1) and NO2 (97 counts s−1 pptv−1), corresponding to detection limits (3σ, 1 s averages) of 13 and 9.9 pptv, respectively. In both cases, the detection threshold is limited by the magnitude and variability in the background determination. The short-term precision (1 s averages) is better than 0.3 % at 10 ppbv O3 and 4 % at 10 pptv NO2. We demonstrate that the sensitivity of the O3 measurement to fluctuations in ambient water vapor and carbon dioxide is negligible for typical conditions encountered in the troposphere. The application of the Ox-CIMS to the measurement of O3 vertical fluxes over the coastal ocean, via eddy covariance (EC), was tested during the summer of 2018 at Scripps Pier, La Jolla, CA. The observed mean ozone deposition velocity (vd(O3)) was 0.013 cm s−1 with a campaign ensemble limit of detection (LOD) of 0.0027 cm s−1 at the 95 % confidence level, from each 27 min sampling period LOD. The campaign mean and 1 standard deviation range of O3 mixing ratios was 41.2±10.1 ppbv. Several fast ozone titration events from local NO emissions were sampled where unit conversion of O3 to NO2 was observed, highlighting instrument utility as a total odd-oxygen (Ox=O3+NO2) sensor. The demonstrated precision, sensitivity, and time resolution of this instrument highlight its potential for direct measurements of O3 ocean–atmosphere and biosphere–atmosphere exchange from both stationary and mobile sampling platforms.
Article
Full-text available
In this paper, we describe the PALM model system 6.0. PALM (formerly an abbreviation for Parallelized Large-eddy Simulation Model and now an independent name) is a Fortran-based code and has been applied for studying a variety of atmospheric and oceanic boundary layers for about 20 years. The model is optimized for use on massively parallel computer architectures. This is a follow-up paper to the PALM 4.0 model description in Maronga et al. (2015). During the last years, PALM has been significantly improved and now offers a variety of new components. In particular, much effort was made to enhance the model with components needed for applications in urban environments, like fully interactive land surface and radiation schemes, chemistry, and an indoor model. This paper serves as an overview paper of the PALM 6.0 model system and we describe its current model core. The individual components for urban applications, case studies, validation runs, and issues with suitable input data are presented and discussed in a series of companion papers in this special issue.
Conference Paper
Full-text available
Surface-atmosphere interactions are traditionally in the vanguard of inter-disciplinary research, with efforts ranging from empirical studies over theoretical generalizations to computational simulations. More recently, data-intensive information discovery promises to further expand our insight into momentum, energy, water, and trace gas cycling. However, “data deluge” rather than the next inter-disciplinary breakthrough can result from poor information overlap among ground, airborne and satellite observations, as well as numerical simulations: scientific return hinges on our ability to merge information among these perspectives, for which the pre-experiment stage provides the unique opportunity to bake it into the overall study design. We present an extensible design approach, which we use to optimize the Chequamegon Heterogeneous Ecosystem Energy-balance Study Enabled by a High-density Extensive Array of Detectors (CHEESEHEAD). In an initial science traceability assessment, we explicitly map how hypotheses on energy balance closure and spatial patterning depend on requirements such as observation type, performance and placement. In the following, we consider 20 National Center for Atmospheric Research Integrated Surface Flux Facility towers, the University of Wyoming King Air flux aircraft, and various remote sensing data products. Next, we determine the trade-space for meeting discipline-specific criteria by evaluating existing parameterizations, such as measurement heights as a function of surface roughness and atmospheric blending. In the final step, high-resolution Large Eddy Simulations yield a super-sample of virtual ground, airborne and satellite observations for various scenarios. We analyze these virtual observations through Environmental Response Functions to yield optimal aircraft sampling patterns for augmenting a stratified random flux tower grid in combination with satellite retrievals. We show how this design approach doubles our ability to test energy balance closure and spatial patterning hypotheses while simplifying logistics by an order of magnitude. Our companion presentations Butterworth et al., Desai et al., Paleri et al. provide initial field-data results from the CHEESEHEAD Intensive Observation Periods.
Article
Full-text available
The split window technique has been used for over thirty years to derive surface temperatures of the Earth with image data collected from spaceborne sensors containing two thermal channels. The latest NASA/USGS Landsat satellites contain the Thermal Infrared Sensor (TIRS) instruments that acquire Earth data in two longwave infrared bands, as opposed to a single band with earlier Landsats. The United States Geological Survey (USGS) will soon begin releasing a surface temperature product for Landsats 4 through 8 based on the single spectral channel methodology. However, progress is being made toward developing and validating a more accurate and less computationally intensive surface temperature product based on the split window method for Landsat 8 and 9 datasets. This work presents the progress made towards developing an operational split window algorithm for TIRS. Specifically, details of how the generalized split window algorithm was tailored for the TIRS sensors are presented, along with geometric considerations that should be addressed to avoid spatial artifacts in the final surface temperature product. Validation studies indicate that the proposed algorithm is accurate to within 2 K when compared to land-based measurements and to within 1 K when compared to water-based measurements, highlighting the improved accuracy that may be achieved over the current single-channel methodology being used to derive surface temperature in the Landsat Collection 2 surface temperature product. Surface temperature products using the split window methodologies described here can be made available upon request for testing purposes.
Article
Presented are four winter seasons of data from an enhanced precipitation instrument suite based at the National Weather Service (NWS) Office in Marquette (MQT), Michigan (250–500 cm of annual snow accumulation). In 2014 the site was augmented with a Micro Rain Radar (MRR) and a Precipitation Imaging Package (PIP). MRR observations are utilized to partition large-scale synoptically driven (deep) and surface-forced (shallow) snow events. Coincident PIP and NWS MQT meteorological surface observations illustrate different characteristics with respect to snow event category. Shallow snow events are often extremely shallow, with MRR-indicated precipitation heights of less than 1500 m above ground level. Large vertical reflectivity gradients indicate efficient particle growth, and increased boundary layer turbulence inferred from observations of spectral width implies increased aggregation in shallow snow events. Shallow snow events occur 2 times as often as deep events; however, both categories contribute approximately equally to estimated annual accumulation. PIP measurements reveal distinct regime-dependent snow microphysical differences, with shallow snow events having broader particle size distributions and comparatively fewer small particles and deep snow events having narrower particle size distributions and comparatively more small particles. In addition, coincident surface meteorological measurements indicate that most shallow snow events are associated with surface winds originating from the northwest (over Lake Superior), cold temperatures, and relatively high surface pressures, which are characteristics that are consistent with cold-air outbreaks. Deep snow events have meteorologically distinct conditions that are accordant with midlatitude cyclones and frontal structures, with mostly southwest surface winds, warmer temperatures approaching freezing, and lower surface pressures.
Article
Machine learning approaches are increasingly used to extract patterns and insights from the ever-increasing stream of geospatial data, but current approaches may not be optimal when system behaviour is dominated by spatial or temporal context. Here, rather than amending classical machine learning, we argue that these contextual cues should be used as part of deep learning (an approach that is able to extract spatio-temporal features automatically) to gain further process understanding of Earth system science problems, improving the predictive ability of seasonal forecasting and modelling of long-range spatial connections across multiple timescales, for example. The next step will be a hybrid modelling approach, coupling physical process models with the versatility of data-driven machine learning.
Article