Content uploaded by Heather A. Shupe
Author content
All content in this area was uploaded by Heather A. Shupe on Oct 25, 2016
Content may be subject to copyright.
Bulletin of the American Meteorological Society
EARLY ONLINE RELEASE
This is a preliminary PDF of the author-produced
manuscript that has been peer-reviewed and
accepted for publication. Since it is being posted
so soon after acceptance, it has not yet been
copyedited, formatted, or processed by AMS
Publications. This preliminary version of the
manuscript may be downloaded, distributed, and
cited, but please be aware that there will be visual
differences and possibly some content differences
between this version and the final published version.
The DOI for this manuscript is doi: 10.1175/BAMS-D-15-00277.1
The final published version of this manuscript will replace the
preliminary version at the above DOI once it is available.
If you would like to cite this EOR in a separate work, please use the following full
citation:
Wolf, B., C. Chwala, B. Fersch, J. Garvelmann, W. Junkermann, M. Zeeman, A.
Angerer, B. Adler, C. Beck, C. Brosy, P. Brugger, S. Emeis, M. Dannenmann, F.
De Roo, E. Diaz-Pines, E. Haas, M. Hagen, I. Hajnsek, J. Jacobeit, T. Jadghuber,
N. Kalthoff, R. Kiese, H. Kunstmann, O. Kosak, R. Krieg, C. Malchow, M. Mauder,
AMERICAN
METEOROLOGICAL
SOCIETY
R. Merz, C. Notarnicola, A. Philipp, W. Reif, S. Reineke, T. Rödiger, N. Ruehr, K.
Schäfer, M. Schrön, A. Senatore, H. Shupe, I. Voelksch, C. Wanninger, S.
Zacharias, and H. Schmid, 2016: The ScaleX campaign: scale-crossing land-
surface and boundary layer processes in the TERENO-preAlpine observatory.
Bull. Amer. Meteor. Soc. doi:10.1175/BAMS-D-15-00277.1, in press.
© 2016 American Meteorological Society
1
The ScaleX campaign: scale-crossing land-surface and boundary layer
1
processes in the TERENO-preAlpine observatory
2
3
B. Wolf1, C. Chwala1, B. Fersch1, J. Garvelmann1, W. Junkermann1, M. J. Zeeman1, A.
4
Angerer3, B. Adler2, C. Beck4, C. Brosy1, P. Brugger1, S. Emeis1, M. Dannenmann1, F. De Roo1,
5
E. Diaz-Pines1, E. Haas1, M. Hagen11, I. Hajnsek7,9, J. Jacobeit4, T. Jagdhuber7, N. Kalthoff2, R.
6
Kiese1, H. Kunstmann1,4, O. Kosak3, R. Krieg6, C. Malchow1, M. Mauder1, R. Merz6, C.
7
Notarnicola8, A. Philipp4, W. Reif3, S. Reineke1, T. Rödiger6, N. Ruehr1, K. Schäfer1, M.
8
Schrön5, A. Senatore10, H. Shupe1, I. Völksch1, C. Wanninger3, S. Zacharias5, and H. P.
9
Schmid1
10
1 Institute of Meteorology and Climate Research (IMK-IFU), Karlsruhe Institute of Technology (KIT), 82467 Garmisch-
11
Partenkirchen, Germany
12
2 Institute of Meteorology and Climate Research (IMK-TRO), Karlsruhe Institute of Technology (KIT), 76021 Karlsruhe,
13
Germany
14
3 Institute for Software & Systems Engineering (ISSE), University of Augsburg, Germany
15
4 Institute of Geography (IGUA), University of Augsburg, Germany
16
5 Department Monitoring & Exploration Technologies, Helmholtz-Centre for Environmental Research (UFZ), 04318 Leipzig,
17
Germany
18
6 Department Catchment Hydrology, Helmholtz-Centre for Environmental Research (UFZ), 06120 Halle/Saale, Germany
19
7 Department of Radar Concepts, German Aerospace Center (DLR), 82234 Oberpfaffenhofen, Germany
20
8 Institute for Applied Remote Sensing, European Academy of Bolzano (EURAC), 39100 Bolzano, Italy
21
9 Institute of Environmental Engineering, ETH Zürich, Switzerland
22
10 Department of Civil and Chemical Engineering, University of Calabria, Rende/Cosenza, Italy
23
11 Institute of Atmospheric Physics, German Aerospace Center (DLR), 82234 Oberpfaffenhofen, Germany
24
25
Corresponding author: Benjamin Wolf
26
Institute of Meteorology and Climate Research (IMK-IFU)
27
Kreuzeckbahnstrasse 19
28
82467 Garmisch-Partenkirchen, Germany
29
E-mail: benjamin.wolf@kit.edu
30
31
Manuscript (non-LaTeX) Click here to download Manuscript (non-LaTeX)
BAMS_WolfEtAl_ScaleX_Rev2_vfinal.doc
2
32
Capsule
33
Augmenting long-term ecosystem-atmosphere observations with multidisciplinary intensive
34
campaigns aims at closing gaps in spatial and temporal scales of observation for energy- and
35
biogeochemical cycling, and at stimulating collaborative research.
36
Abstract
37
ScaleX is a collaborative measurement campaign, co-located with a long-term
38
environmental observatory of the German TERENO (TERrestrial ENvironmental
39
Observatories) network in mountainous terrain of the Bavarian Prealps, Germany. The aims
40
of both TERENO and ScaleX include the measurement and modeling of land-surface
41
atmosphere interactions of energy, water, and greenhouse gases. ScaleX is motivated by the
42
recognition that long-term intensive observational research over years or decades must be
43
based on well-proven, mostly automated measurement systems, concentrated on a small
44
number of locations. In contrast, short-term intensive campaigns offer the opportunity to
45
assess spatial distributions and gradients by concentrated instrument deployments, and by
46
mobile sensors (ground/airborne) to obtain transects and three-dimensional patterns of
47
atmospheric, surface, or soil variables and processes. Moreover, intensive campaigns are
48
ideal proving grounds for innovative instruments, methods and techniques to measure
49
quantities that cannot (yet) be automated or deployed over long time-periods. ScaleX is
50
distinctive in its design that combines the benefits of a long-term environmental monitoring
51
approach (TERENO) with the versatility and innovative power of a series of intensive
52
campaigns, to bridge across a wide span of spatial and temporal scales. This contribution
53
presents the concept and first data products of ScaleX-2015. The second installment of
54
3
ScaleX is set for the summer 2016 and periodic further ScaleX campaigns are planned
55
throughout the life-time of TERENO. This paper calls for collaboration in future ScaleX
56
campaigns, or by using our data in modeling studies. It is also an invitation to emulate the
57
ScaleX concept at other long-term observatories.
58
59
4
60
1. Introduction
61
ScaleX is an intensive interdisciplinary observation campaign in a region of complex
62
topography and land-use/land-cover variations in Southern Germany. It explores the
63
question how well measured and modeled components of biogeochemical and biophysical
64
cycles match at the interfaces of soils, vegetation and the atmosphere, and across various
65
spatial and temporal scales. This type of lead question is not new: scale-integration in
66
observation and modeling for land surface – atmosphere exchange processes was one of
67
the principal motivations for past large-scale field programs, such as FIFE (First ISLSCP Field
68
Experiment, Kansas, USA; e.g., Sellers et al., 1988, 1992), BOREAS (Boreal Ecosystem-
69
Atmosphere Exchange Study, Canada; e.g., Hall, 1999; Sellers et al., 1995 – and articles in
70
the same issue), or CME (Carbon in the Mountains Experiment; e.g., Sun et al. 2010; Desai et
71
al. 2011) to name just three prominent examples. These (and other) field programs have
72
resulted in numerous publications, have spawned research ideas, and led to new
73
observation and modeling techniques in ecosystem-atmosphere science. Data from these
74
programs have served as valuable benchmarks for model development and measurement
75
inter-comparisons, and have contributed significantly to progress in scale-integration and
76
matching of observations and modeling. So why should we endeavor on yet other field
77
campaigns with similar objectives?
78
This question has many answers. Firstly, despite the progress achieved by past field
79
campaigns, the mismatch between observations of land-surface processes and their
80
modeled equivalents is still so large that it constitutes a major source of uncertainty in
81
climate models (e.g., Best et al., 2015). Secondly, new knowledge in science invariably gives
82
5
rise to new questions (Firestein 2012). As we learn more about dominant processes and
83
feedback relations, we discover patterns of discrepancy and unexplained deviations at
84
previously disregarded scales that are potentially responsible for long-term trends. Thirdly,
85
progress in instrumentation and data communications allow us to close gaps in the
86
temporal and spatial coverage of observations that previous field campaigns were limited
87
by. Lastly, experience shows that, whenever scientists from various backgrounds work
88
together, on the same objectives, and on the same field sites, collaboration fosters new
89
ideas and thinking-outside-the-box that gives rise to new knowledge (Hall 1999; Goring et al.
90
2014).
91
In our view, these points alone justify a new scale-crossing field campaign such as ScaleX.
92
However, in a number of ways ScaleX is different from previous field programs. As
93
presented below, ScaleX is directed at a range of spatial scales that is generally smaller, but
94
with higher measurement and modeling resolution and more complex topography than
95
considered in previous land surface – atmosphere processes campaigns.
96
Yet the most important novelty of ScaleX probably lies in its infrastructural setting and
97
temporal outlook. The backbone of micrometeorological, hydrological and ecosystem-
98
atmosphere exchange instrumentation used by ScaleX is formed by the permanent
99
environmental TERENO-preAlpine observatory (Zacharias et al. 2011), with stations
100
distributed along an elevation gradient in the pre-Alpine region of Germany (see Section 3).
101
The ScaleX campaign builds on this research infrastructure with a multitude of additional
102
instruments and observation platforms (ground based in-situ, remote sensing, and
103
airborne), to enhance spatial and temporal measurement resolutions, and to complement
104
the permanent suite of measurements with additional observed variables and processes.
105
6
The first campaign (June-July 2015) was run by KIT/IMK-IFU Garmisch-Partenkirchen (the
106
institute that operates the backbone infrastructure) with collaborating partners from the
107
region (see list of co-author affiliations). The second campaign, set for June-July 2016,
108
includes a larger number of national and international partners and collaborators, who are
109
invited to use the permanent research infrastructure, with data- and power connectivity.
110
Because TERENO-preAlpine is set to be operated for the next two decades or longer, it will
111
be possible to re-visit the same sites periodically again in future editions of ScaleX. In our
112
view this long-term continuity is a valuable opportunity to expand the usual narrow
113
temporal constraint of intensive measurement campaigns toward time-scales that are
114
important for land-use change, climate change and ecosystem renewal. The ScaleX concept
115
can likely serve as a model for similar combinations of long-term backbone observatories
116
and periodic intensive campaigns in other permanently operated ecosystem-atmosphere
117
observatories, such as in the AmeriFlux network (Baldocchi 2003; Boden et al. 2013) and the
118
National Ecological Observatory Network (NEON; Kampe et al. 2010) of the United States, or
119
the Integrated Carbon Observation System (ICOS; https://icos-ri.eu) in Europe.
120
In short, the general idea of ScaleX is to introduce a concept that combines the objectives of
121
long-term ecosystem research with those of intensive campaigns; to expand the scale and
122
resolution of observations; to stimulate collaborative, interdisciplinary research and
123
synergistic interactions.
124
The purpose of the present article is to provide some background on the rationale,
125
organization and specific research goals of ScaleX (Section 2); to briefly introduce the
126
TERENO-preAlpine observatory with its principal site and the long-term backbone
127
observation program (Sections 3 and 4); to give an overview of the instrumentation
128
7
deployed during ScaleX-2015 (Section 5); and to present examples of derived data products
129
(Section 6). Lastly, but most importantly, this article hopes to attract interested research
130
groups as collaborating partners in future campaigns of ScaleX (Section 7).
131
2. Background
132
In the biogeochemical and biophysical cycles that shape our world, terrestrial and aquatic
133
ecosystems are the most important brokers for energy and matter exchanges between the
134
atmosphere, oceans and continents. They provide natural resources, are mediators of
135
climate change, and contribute to water availability and soil conservation. Terrestrial
136
ecosystems in particular are extremely variable over a wide range of scales both in space
137
and time, and yet they form the most direct foundation for the majority of food production,
138
water- and air-quality that humanity depends on. Processes, such as the flows of energy,
139
water, oxygen (O), carbon (C), nitrogen (N), and other essential trace substances in and
140
between ecosystems and their environment indicate the vibrance and variability of
141
ecosystems, and underline the inter-dependency of supporting, provisioning and regulating
142
services that ecosystems provide (e.g., Reid et al., 2005).
143
In terrestrial ecosystems, important exchange fluxes occur at the interfaces of the Earth-
144
system compartments atmosphere, biosphere, pedosphere, and hydrosphere that each act
145
as reservoirs and sites of transformation in biogeochemical and energy cycling. Given their
146
different nature, chemical and physical transformation and transport processes within these
147
compartments act on vastly different temporal and spatial scales (temporally from fractions
148
of a second for turbulence and biochemical light-responses, to decades or longer for climate
149
trends and soil development; spatially from soil microbes to hydrological catchments or
150
landscape units; e.g., Ehleringer and Field 1993), and interactions between them are
151
8
typically characterized by highly non-linear feedback dynamics. Thus, no single natural scale
152
of study exists that can adequately represent the manifold interplay of ecosystem-
153
atmosphere processes (e.g., Levin, 1992). Scaling errors typically arise from inconsistencies
154
or nonlinear behavior when observations or models at one scale are transferred or
155
aggregated to another, or when model or measurement resolutions filter out temporal or
156
spatial interactions (e.g., Mahrt 1987; Bünzli and Schmid 1998; Schmid and Lloyd 1999).
157
From this perspective, any activity aiming to understand interactions between Earth-system
158
compartments requires a scale-integrative observation strategy and needs to go beyond
159
simply assigning aggregated measured values to a larger spatial or temporal domain
160
(Osmond et al. 2004; May 1999; Caldwell et al. 1993).
161
One pertinent example for which a scale-integrative observation approach is considered to
162
be essential is the observation of the energy balance at the land surface. The turbulent
163
components of the relevant exchange fluxes (i.e., sensible and latent heat fluxes) are
164
commonly determined by the eddy-covariance (EC) method. In typical deployments, EC
165
measurements capture turbulent surface-atmosphere interactions on spatial scales of a few
166
hundred meters or less, and over time scales of an hour or less (e.g., Baldocchi 2003).
167
However, microscale atmospheric processes (e.g., Orlanski, 1975) can be influenced by
168
circulation patterns at scales of up to several 10s of kilometers, persisting for hours (sub-
169
meso- to mesoscale; e.g., Emeis 2015). This kind of scale-interaction is now widely
170
recognized as a principal cause for the so-called energy balance closure problem (Mauder et
171
al. 2010): in most energy balance observations worldwide, the turbulent components are
172
seen to underestimate the sum of their radiative and conductive counterparts by 10-20%
173
(e.g., Stoy et al. 2013), likely due to unaccounted for sub-meso- and mesoscale contributions
174
to sensible and latent heat transport.
175
9
Scale related complications are of particular concern in complex and fragmented landscapes
176
such as mountain regions, where high spatial variability of land use and topography typically
177
entail abrupt changes in available energy, precipitation, soil moisture, vegetation, or soils
178
(Beniston 2006; Poulos et al. 2012). Thus, ecosystem research in complex environments
179
especially demands scale-integrative approaches for observations and modeling.
180
The ScaleX-2015 campaign was motivated by far-reaching research questions and topics,
181
including (1) how do mesoscale structures in the atmospheric boundary layer (ABL)
182
influence EC-derived surface fluxes; (2) interaction of trace-gas plumes from strong
183
(anthropogenic) point sources with natural background fluxes; (3) development of
184
instrumentation and methods to use unmanned aerial vehicles (UAV) for ABL
185
characterization of scalars and turbulence; (4) how do patterns of precipitation relate to soil
186
moisture and runoff over different temporal and spatial scales.
187
Examples of observations in ScaleX-2015 motivated by these questions are presented in
188
Section 6a-e.
189
190
3. The TERENO-preAlpine Observatory
191
TERENO (TERrestrial ENvironmental Observatories) is a German network of observatories
192
investigating the ecological and climatic impact of global environmental change on
193
terrestrial systems (Zacharias et al. 2011). The TERENO-preAlpine observatory is located in
194
the Bavarian foothills of the Alps (i.e., the Bavarian Prealps), with elevations from 450 m up
195
to 2000 m above sea level (a.s.l.) roughly to the west of an axis between Munich, Germany,
196
and Innsbruck, Austria (Fig. 1). At its core is an extensively instrumented site cluster in the
197
10
catchments of the rivers Ammer (709 km2) and Rott (55 km2). With dairy farming as the
198
dominant land use in the valleys of this region, the preAlpine observatory includes the
199
grassland sites Fendt, Rottenbuch and Graswang (www.europe-fluxdata.eu station codes:
200
DE-Fen, DE-RbW, and DE-Gwg) at elevations of 595, 769 and 864 m a.s.l., respectively (Fig.
201
1).
202
The climate change sensitivity of mountain regions, such as the TERENO-preAlpine
203
observatory, is seen to be amplified compared to global averages (Böhm et al. 2001;
204
Smiatek et al. 2009; Calanca 2007), with expected strong consequences in the regional
205
thermal and precipitation regimes, C- and N-dynamics, and thus nutrient cycling and
206
ecosystem functioning (Mills et al. 2014). To study the impact of climate change on
207
ecosystem functioning and services, regional circulation and precipitation patterns, the
208
continuously operated backbone infrastructure of TERENO-preAlpine includes ecosystem-
209
atmosphere flux stations along an elevation gradient, micrometeorology and boundary layer
210
sounding systems, and a hydro-meteorological mesoscale network with precipitation gauge
211
transects and a rain radar (Fig. 1, right). The ScaleX campaign 2015 focused primarily on DE-
212
Fen, which is described in detail in the next section.
213
4. The DE-Fen site and its permanent backbone instrumentation
214
DE-Fen is located at the head of a small tributary stream to the river Rott (Fig. 2). The land
215
use at the bottom of this shallow valley is dominated by grassland, sometimes with small
216
patches of cropland, mostly maize. Three dairy farms are located within a distance of less
217
than 1 km to the south and west of the site. To the west, a plateau parallels the valley
218
approx. 100 to 130 m above its floor. The plateau’s shoulder is covered predominately with
219
mixed forest. About 5 km south-west of the site the German Weather Service (DWD)
220
11
operates its Meteorological Observatory Hohenpeissenberg (MOHP, 988 m a.s.l.). The
221
northern rim of the Alps lies approx. 30 km to the south.
222
The permanent backbone instrumentation at DE-Fen includes a micrometeorology station,
223
hydro-meteorological installations, and a lysimeter cluster containing the principal local soil
224
types for the measurement of biosphere-atmosphere-hydrosphere exchange processes
225
(specifics of instrumentation are given in Table 1). The core micrometeorology
226
instrumentation is an EC-system (for momentum, CO2, water vapor, and heat exchange
227
fluxes), a multi-component surface radiation balance system (including direct/diffuse
228
incoming shortwave radiation), photosynthetically active radiation (PAR), soil heat-flux
229
plates and profiles of soil temperatures and soil moisture, as well as other standard
230
meteorological instruments. This array of in-situ instruments (Fig. 2) is augmented by a
231
ceilometer for the determination of boundary layer height.
232
To quantify the grassland water balance at high temporal resolution (30 minutes) the
233
lysimeter cluster (Fig. 2 and Table 1) contains 18 weighable large (1.0 m2, 1.4 m height)
234
grassland-soil monoliths, equipped with soil temperature and moisture sensors. Over each
235
monolith, soil-atmosphere exchange fluxes of CO2, CH4 and N2O are determined through
236
sequential sampling by an automated static chamber system in conjunction with a quantum
237
cascade laser absorption spectrometer.
238
The heart of hydro-meteorological measurements at DE-Fen is a wireless sensor network
239
(nicknamed SoilNetFen, following Bogena 2010) which covers an area of approx. 400 m x
240
330 m in the footprint of the EC station. SoilNetFen measures soil moisture, soil
241
temperature, and matrix potential every 15 minutes at 5, 20, and 50 cm depth at 55
242
locations (Fig. 2 and Table 1). A Cosmic Ray Neutron Sensor (CRNS; Zreda et al. 2008)
243
12
monitors the field-integrated variations of the soil water content. SoilNetFen is augmented
244
by three discharge gauges, five groundwater wells and one precipitation gauge (Fig. 2 and
245
Table 1).
246
5. Additional Instrumentation and Measurements during ScaleX
247
To extend spatial and temporal scales of observation beyond the range covered by the
248
permanent backbone setup at DE-Fen, the measurement program was complemented
249
during ScaleX-2015 by a combination of additional measurement locations, remote sensing
250
instruments, and airborne platforms (visit www.scalex.imk-ifu.kit.edu for illustrations).
251
Instruments that could not be operated in a continuous mode were integrated by means of
252
intensive observation periods. Specifics of all instruments or installations mentioned in this
253
section are summarized in Table 1.
254
Boundary layer remote sensing was conducted by three high-resolution scanning Doppler-
255
LIDAR (light detection and ranging) systems for vertical profiles of wind and turbulence
256
(1000 m max.), as well as by a radio acoustic-sounding system (RASS; Emeis et al., 2009) to
257
determine vertical profiles of wind and temperature (560 m max.). Resulting data products
258
include the characterization of the turbulence and thermal structure in the boundary layer,
259
as well as the detection of low level jets. In addition, a ground-based scanning
260
microwaveradiometer was operated to obtain integrated water vapor (IWV), liquid water
261
path (LWP), and temperature and humidity profiles.
262
The remote sensing measurements were complemented by airborne observations. A swarm
263
of unmanned aerial vehicles (UAV) was jointly operated by the IMK-IFU, ISSE and IGUA in
264
different experiments and coordinated flight patterns (four copters, three fixed-wing), each
265
13
equipped with temperature, humidity and pressure sensors. Due to legal provisions, the
266
maximum ascent of the copters above ground level was limited to 150 m. In addition to the
267
UAVs, the microlight aircraft D-MIFU (see Junkermann 2001; Junkermann et al. 2011;
268
Metzger et al. 2013) was deployed to provide wind, temperature, moisture, turbulent fluxes
269
and radiation measurements at a larger spatial extent of about 12 km by 12 km around DE-
270
Fen, from 50 m up to 2.5 km above ground level (a.g.l.).
271
To explore the spatial- and temporal variability of precipitation, rain gauges were installed
272
at 5 locations within the SoilNetFen area and at 17 additional locations in the Rott
273
catchment. This gauge network, as well as a micro rain radar (MRR) and two disdrometers,
274
augmented and provided ground-truth for the DWD C-band radar at MOHP and the TERENO
275
X-band rain radar (Fig. 1). Furthermore, the chemistry and isotopic composition of
276
precipitation, surface- and subsurface water was tracked by water samples taken both
277
manually and automatically throughout the campaign (using a cavity ring down, CRD,
278
spectrometer; Table 1). To link the soil moisture measurements at the point and catchment
279
scales, Mobile CRNS (TERENO Rover), air-borne (synthetic aperture radar; F-SAR) sensors
280
were used, and linked to satellite derived data (RADARSAT 2).
281
Greenhouse gas (GHG) flux measurements at the lysimeter cluster were complemented by a
282
large static chamber (Schäfer et al. 2012, in conjunction with a trace gas analyzer) for CH4
283
flux measurements on a patch of grassland that is frequently flooded, and by atmospheric
284
CH4 concentration measurements. To evaluate the regional CH4 sink- or source strength,
285
profiles of atmospheric CH4 concentrations (using a CRD) were determined on a tower at
286
heights of 1, 5 and 10 m above ground, and up- and downwind of a dairy farm (using an
287
14
open-path methane analyzer; range ~100 m), along with wind speed and direction (wind
288
sensor network).
289
6. Some first data products
290
The activities in ScaleX-2015 were organized along the overarching research questions of
291
land-surface-atmosphere interactions in the atmospheric boundary layer discussed in
292
Section 2. Here, a selection of first data products is presented, to illustrate the range of
293
intensive observations conducted in the ScaleX campaigns.
294
a. High resolution ABL motion structure by a LIDAR cluster
295
Standard observations of biosphere-atmosphere exchange (e.g., using the EC method)
296
assume horizontal homogeneity of the turbulence structure and generally ignore the
297
contributions of ABL-scale or mesoscale motions on exchange fluxes. In fragmented
298
landscapes with topography and mixed land use, secondary circulations can develop that
299
affect the validity of standard exchange observations. To account for the effects of such
300
non-local motions on turbulent exchange near the surface is difficult, but their exclusion
301
introduces bias in long-term fluxes (Mahrt 1987, 2010).
302
In ScaleX-2015 a cluster of three Doppler boundary layer LIDARs was used, in conjunction
303
with a network of sonic anemometers to characterize the motion structure over the entire
304
ABL continuously, and at high temporal and spatial resolution, over the duration of the
305
campaign. The Doppler LIDARs (Table 1) recorded three-dimensional wind vectors (u, v, and
306
w) in a vertical scanning profile arrangement that served as a virtual tower up to
307
approximately 1000 m above the surface (Fig. 3, left panel).
308
15
The observations revealed flow features over a range of time and length scales. The right
309
panel of Fig. 3 illustrates a representative day (1 July 2015): The development of thermally
310
driven activity in the ABL at around 07:00 UTC (08:00 local standard time) was visible first as
311
a change of wind direction and vertical wind speed, starting at the surface and rising rapidly.
312
Daytime flow was dominated by northerly to easterly wind throughout the boundary layer.
313
In addition, the daytime boundary layer was characterized by typical convective motion
314
features over scales of several minutes and vertical extents of several hundred meters. After
315
sunset, the wind direction shifted to the east and a low-level easterly jet formed around
316
19:00 UTC between 200 and 500 m a.g.l., but decayed in magnitude around 23:30 UTC as
317
shown by the horizontal wind speed. Nighttime wind direction above 200 m stayed mostly
318
southeasterly to easterly, in contrast to layers below 200 m, which showed low wind speed,
319
but directional shear up to 180 degrees, even below the low-level jet.
320
On this particular day, more than 82% of the recorded nocturnal half-hourly EC observations
321
of CO2 and heat exchange were rejected, based on standard quality control criteria including
322
stationarity, turbulence characteristics and signal noise (Mauder et al. 2013). The remaining
323
nocturnal surface flux observations coincided with the presence of the low-level jet after
324
sunset. Between 9:00 and 19:00 UTC no data were rejected or flagged. This nighttime bias
325
of missing turbulence data underlines the difficulty of obtaining nighttime trace-gas flux and
326
transport information, discussed in the next Section. High resolution ABL motion data, such
327
as those presented here, are anticipated to be valuable to evaluate (e.g.) Large Eddy
328
Simulation (LES) models for the assessment of typically unresolved non-local contributions
329
to surface fluxes.
330
16
b. Variability of methane concentration in the nocturnal boundary layer (NBL)
331
Methane (CH4) is an important GHG of predominantly biogenic origin, with ecosystems
332
acting either as net sources or sinks. Wetlands and water-logged soils emit CH4 due to
333
activity of methanogenic microbes, while upland and well-aerated soils are usually net sinks
334
for atmospheric CH4, because of the predominance of CH4-oxidizing microbes. Although
335
methane sensors fast enough for eddy-covariance are available, CH4 fluctuations often
336
range near the limit of sensitivity and EC-signals tend to be noisy (e.g., Hommeltenberg et al.
337
2014). A convenient alternative method is the static chamber method (see Pihlatie et al.
338
2013 for a review). This method determines surface exchange fluxes over a well-defined
339
area of ground (commonly < 1 m2), by measuring trace gas accumulation or depletion over a
340
given time, referenced to the chamber volume. Because variability in soils (e.g., moisture
341
and substrate availability) typically ranges down to similar spatial scales as the chamber
342
dimensions, the scaling up from chambers to a scale comparable to an EC-flux footprint
343
(e.g., Schmid 2002), or to the resolution at which ecosystem exchange models are
344
commonly run, is a formidable problem (e.g., Pihlatie et al. 2010). An additional
345
confounding problem is posed by larger scale spatial heterogeneity, e.g., in mixed-land use
346
areas with upland grassland or crops, patches of wetlands, and pastures or barns with
347
methane-producing cattle. Particularly in night-time stable conditions, plumes of methane
348
enriched air (for example) can be transported over considerable distances with very little
349
mixing: if such a transient plume increases the local atmospheric CH4 concentration, the
350
higher CH4 supply may lead to extra stimulation of the methane consuming microbes in the
351
soil. The resulting increased uptake rate needs to be quantified and considered when
352
calibrating and validating biogeochemical models designed to simulate CH4 exchange based
353
on local soil properties and soil environmental conditions. Transient plumes may also affect
354
17
night-time EC-measurements of methane: shallow CH4 plumes may lead to spurious vertical
355
gradients that, in the presence of weak turbulence, introduce a contribution to the EC-flux
356
signal that has no linkage to surface sources or sinks in the flux footprint (Finnigan 2004).
357
Alternately, larger-scale spatially averaged trace gas fluxes (e.g., at the scale of a model grid-
358
cell) can in theory be derived by the boundary layer budget method (Denmead et al. 1996;
359
Emeis 2008), an inverse method, where surface fluxes are the residual result of
360
concentration changes and transport terms observed and modeled over a hypothetical box
361
bounded by the height of the ABL.
362
However, nighttime application of direct or inverse trace-gas flux estimates is a challenge,
363
because very little is known about spatial and temporal CH4 variability above the surface in
364
the NBL, and observations are difficult and rare. In ScaleX a new approach was explored to
365
assess plumes and gradients of methane in the NBL that may make such observations more
366
accessible in future. Measurements of atmospheric CH4 concentration were performed by
367
pumping ambient air through a sampling tube (Teflon®, outer diameter: 1/8”, 3.2 mm) to a
368
CRD-spectrometer. To extend nighttime vertical CH4 profiles to beyond the 10 m tower at
369
DE-Fen (location H in Fig. 2), the end of a sampling line (70 m length) was mounted to the
370
hexacopter F550 and periodically raised to heights of 10, 25, and 50 m a.g.l. Data are
371
reported as one minute averages for all heights.
372
Observations from 21 July 2015 (Fig. 4) showed that atmospheric CH4 concentrations at all
373
measurement heights increased well above background concentration (1.9 ppm,
374
determined as the average ABL concentration in well-mixed daytime conditions). The lower
375
sampling heights exhibited strong variations, whereas fluctuations were much reduced
376
above the 10 m tower height. Fig. 4 also show considerable negative vertical CH4
377
18
concentration gradients which start to decrease in the second half of the night, indicating
378
slow vertical mixing in the NBL. These findings suggest shallow advection from areas with
379
strong CH4 sources, because the local grassland soils were sinks for atmospheric CH4.
380
Concurrent wind directions point to dairy barns nearby as the likely culprit. Observations
381
from other nights confirmed this night-time advection to be a regular occurrence. The
382
observations also show that the measured values by the hexacopter method agree well with
383
the tower measurements at 10 m. They indicate that, using UAVs to carry trace gas intake
384
lines to heights beyond the reach of common instrument masts, is promising as a low-cost
385
and flexible way to explore the GHG, or other trace gas structure in the nocturnal boundary
386
layer.
387
c. Use of UAVs and microlight aircraft for three-dimensional boundary layer
388
characterization
389
One of the biggest challenges for projections of regional climate or ecosystem-atmosphere
390
interactions is to get the hydrometeorology right (Clark et al. 2015). Without knowledge of
391
how much water is transpired or evaporated in a region, we cannot predict how much CO2
392
the plants will assimilate. The amount of water vapor that is transported from one meso-
393
scale model grid-cell to another is crucial information for predictions of when and where
394
that water will fall as rain. In complex terrain, it is important to know on which side of a
395
ridge rain falls, to infer whether vegetation will be water stressed, or whether a river might
396
flood. However, the evaluation of regional scale hydro-meteorological models by
397
observation is challenged by (i) unresolved spatial variability of atmospheric temperature
398
and humidity, and (ii) a lack of adequate experimental tools to determine the balances of
399
water and heat over model grid cells or model sub-domains (Lorenz and Kunstmann 2012).
400
19
In ScaleX an attempt is made to tackle this problem by combining the hydrometeorological
401
in-situ observations of TERENO-preAlpine with a suite of remote sensing techniques
402
(ground-based and airborne), instrumented UAVs, and a microlight aircraft, to capture the
403
three-dimensional variability of atmospheric state variables in the ABL at high resolution.
404
ScaleX-2015 included a proof-of-concept campaign to coordinate the flight patterns of a
405
swarm of UAVs and the microlight aircraft.
406
The microlight aircraft D-MIFU was used to assess 3D distributions of winds, air
407
temperature, dewpoint, latent and sensitive heat fluxes, surface temperature, radiation
408
balance and aerosol size distributions. Flights included horizontal tracks as well as vertical
409
“spiral staircase” profiles from about 50 m up to 2000 m a.g.l. directly above and at the
410
vertices of a 12 km by 12 km rectangle around DE-Fen. At a radius of about 500 m around
411
the DE-Fen EC-station, small UAVs were operated to determine the small-scale spatial and
412
temporal variability of the thermal structure in the ABL.
413
Battery operated UAVs are constrained by flight duration, horizontal distance (~300 m) and
414
maximum ascent, while aircraft are limited by the lowest legally possible flight level (50 m).
415
To capture the thermal structure in the boundary layer over DE-Fen, several vertical profiles
416
of air temperature were determined with the hexacopter, one fixed wing UAV and the D-
417
MIFU microlight on 15 July (Fig. 5), each set within about 15-30 minutes. Though the
418
measurements were not taken at exactly the same time and location, the temperature
419
measurements of all three systems mostly agreed within 0.5 °C for the overlapping heights.
420
Therefore, the aerial vehicles complemented each other to obtain a seamless
421
representation of the vertical structure from the ground up to the free troposphere.
422
20
Together with the use of the hexacopter in trace gas measurements, these first results are
423
encouraging for the use of lightweight UAVs as an emerging technology in atmospheric
424
boundary layer research. Battery operated UAVs have no exhaust, and very little heat
425
emissions, and can be programmed to perform complex flight patterns or (for copters) hold
426
a given position even in convectively turbulent conditions. The 2015 campaign also
427
established that a swarm of 3 copters and 3 fixed-wing UAVs can be deployed together, to
428
perform complex coordinated sensing patterns in a small boundary layer volume (ca. 300 m
429
wide and high; not shown), e.g., to perform in-situ measurements at exactly the same time
430
and height at different locations. The UAVs used here are lightweight (below 2.5 kg) and
431
thus the instrument payload is very limited (currently temperature, humidity, pressure, and
432
wind velocity; see Table 1). With progressive developments in sensor miniaturization, rapid
433
expansion of further research applications of UAVs can be expected in future campaigns.
434
d. Soil moisture and precipitation patterns at a range of scales
435
Precipitation and soil moisture are the fundamental hydrologic quantities required for a
436
more profound understanding of runoff- and flood generation, but they are also essential
437
for plant-physiological and biogeochemical processes (Ruehr et al. 2014; Yao et al. 2010;
438
Clough et al. 2004). At the same time, the measurement of precipitation and soil moisture
439
beyond the point scale is one of the most critical challenges in hydrological sciences.
440
Therefore, a major focus within the ScaleX campaign concerned the characterization of the
441
spatial and temporal variability of rainfall and soil moisture at DE-Fen and within the Rott
442
catchment.
443
For precipitation, this objective is accomplished using weather radar data and a dense
444
network of rain gauges at 22 locations (Fig. 6) in the Rott catchment region. The average
445
21
distance between gauges was 250 m at DE-Fen and 2.5 km in the catchment. To handle
446
random errors in the rain gauge data, each of the 22 locations was equipped with a set of
447
three tipping bucket rain gauges (Krajewski et al. 2003). With this level of redundancy
448
spurious outliers and instrumental errors could be identified and the faulty sensor excluded
449
from estimates of precipitation, resulting in quality controlled and mostly gap-free
450
precipitation time series.
451
Though the average distance between the rain gauge sites is only 2.5 km, local convective
452
events may remain concealed. To cover the whole target region with high spatial resolution,
453
data from the polarimetric C-band weather radar at MOHP (see Fig. 1) are used. Fig. 6 shows
454
an example of the high spatial variability of hourly precipitation during a convective event.
455
While the rain gauge and radar data is in good agreement at the gauge locations, the gauges
456
alone cannot resolve the spatial variability at an hourly scale. The combination of radar and
457
gauges facilitates validation and adjustment of the radar field at the gauge locations, and
458
correcting it for inherent radar errors. The corrected radar field can then serve as additional
459
high resolution rainfall information to be used in hydrological modeling.
460
Soil moisture patterns were identified based on SoilNetFen measurements (Fig. 2).
461
Individual point-measurements of soil volumetric water content (VWC) at 5, 20 and 50 cm
462
depth were interpolated to maps for each depth, using a simple inverse distance weighting
463
scheme (Pebesma 2004). Fig. 7 a-c to illustrate resulting moisture fields for July 15, 2015.
464
ScaleX-2015 included a first comparison of soil moisture distributions determined by the
465
SoilNetFen capacitance-based sensors and by the TERENO Rover, a mobile cosmic-ray
466
neutron sensor system mounted on a pick-up truck. CRNS is a relatively new technique to
467
estimate spatially integrated soil moisture, introduced by Zreda et al. (2008), but based on
468
22
theory largely from the 1950s. Primary cosmic rays enter Earth from galactic origins mainly
469
as protons. Collisions with nuclei in the atmosphere, and later in soils, generate cascades of
470
neutrons with decreasing levels of energy (secondary cosmic rays). In the words of Zreda et
471
al. (2008), “soil moisture content on a horizontal scale of hectometers and at depths of
472
decimeters can be inferred from measurements of low-energy cosmic-ray neutrons that are
473
generated within soil, moderated mainly by hydrogen atoms, and diffused back to the
474
atmosphere. These neutrons are sensitive to water content changes, but largely insensitive
475
to variations in soil chemistry, and their intensity above the surface is inversely correlated
476
with hydrogen content of the soil”. However, according to Köhli et al. (2015), the spatial
477
sensitivity of the sensor decreases sharply with distance, and the effective measurement
478
depth depends on soil type and moisture content, typically ranging from 10 to 40 cm.
479
Stationary CRNS are commonly used for monitoring of soil moisture variations over time,
480
while the mobile CRNS TERENO Rover can detect spatial variations along transect paths,
481
filtered by a footprint size on the order of several hundred meters diameter (Zreda et al.
482
2008).
483
The TERENO Rover was repeatedly employed at DE-Fen during ScaleX-2015, and also along
484
tracks throughout the Rott catchment. The lower right panel of Fig. 7 shows VWC derived
485
from 127 data points observed by the CRNS along the Rover tracks at DE-Fen (transect
486
velocity of approx. 2 km h-1). Because the southern part of the SoilNetFen area could not be
487
accessed with the vehicle, data are missing there. Considering the large footprint of CRNS,
488
the gradient of dry to moist conditions from west to east in SoilNetFen, is captured well by
489
the Rover, both qualitatively and quantitatively. The “eye” structures of apparent high VWC
490
in the plot are likely artifacts of the basic interpolation method used in these preliminary
491
results.
492
23
e. Runoff generation mechanisms and storage interactions
493
Surface water – groundwater interactions are complex (Sophocleous 2002) and crucial for
494
the functioning of riparian ecosystems (Kalbus et al. 2006; Jones and Holmes 1996) and the
495
hyporheic zone (Sophocleous 2002; Kalbus et al. 2006; Jones and Holmes 1996). Because
496
groundwater is usually depleted in heavier stable isotopes compared to surface water
497
bodies (Uhlenbrook et al. 2002; Tetzlaff et al. 2009; Coplen et al. 2000; Hinkle et al. 2001),
498
the stable isotope abundances of oxygen-18 and deuterium in water have been used widely
499
as natural tracers to explore hydrological processes and interactions between surface- and
500
groundwater.
501
As DE-Fen is located at the bottom of a shallow valley, the hydrodynamic gradient is weak
502
and it can be expected that groundwater – surface water interactions are an important
503
mechanism in the study area. So, not surprisingly, hydrochemical analysis and groundwater
504
level measurements indicate the existence of exchanges between groundwater and surface
505
water (not shown here). However, the detailed mechanisms of runoff generation and
506
storage system interactions are not satisfyingly understood in this region.
507
To explore runoff generation dynamics and the connections of stream water to the local
508
aquifer system, the water isotopic composition was analyzed during the ScaleX campaign
509
(Table 1) by automatically drawing stream water samples every 6 hours at the outlet of the
510
headwater catchment (Fig. 2, location D). In addition, groundwater was manually sampled
511
bi-weekly at the same location, and batch samples of precipitation were collected weekly
512
close to the EC-station.
513
Figure 8 shows an overview of observed precipitation, stream discharge, groundwater table
514
variations and the δ18O isotopic composition of water from these three hydrological
515
24
compartments in the Rott catchment over the ScaleX-2015 period. The top panel
516
demonstrates that the relatively strong rainfall events in the first half of the campaign were
517
closely traced by peaks in discharge. The middle panel indicates that precipitation water
518
tends to exhibit higher δ18O values than groundwater (which varies only very little). In
519
contrast, the stream water isotope signature is evidently reacting rapidly to inputs from
520
precipitation (with its higher isotope signature). Despite different sampling frequencies, the
521
isotopic enrichment of stream water after strong rainfall events suggests that infiltration or
522
drainage of excess water was the dominant runoff mechanism during rainfall events. During
523
the recession of stream flow, the contribution of groundwater to runoff increased, resulting
524
in very similar isotopic composition of stream water and groundwater in the low flow period
525
observed during the comparatively dry second half of the campaign.
526
The data in Fig. 8 illustrate the usefulness of continuous time series of isotopic water
527
signatures to identify response times of flow regimes, recharge of water storage bodies and
528
mixing processes. Such data, in conjunction with regional scale hydrometeorological and soil
529
moisture information presented above (Fig. 6 and 7), form a valuable test-bed for model
530
evaluation and as ground-truth for satellite based estimates of the land surface water
531
balance. To this end, and to integrate local observations of soil moisture over the region,
532
airborne and satellite remote sensing methods are used. During ScaleX-2015, an airborne
533
synthetic aperture radar mission (F-SAR, fully polarimetric L-band) was conducted by DLR
534
(German Aerospace) over the ScaleX area, and a space-borne SAR (RADARSAT 2) scene was
535
acquired for the entire TERENO-preAlpine region.
536
25
7. Discussion and outlook
537
Integrated observation programs of ecosystem – atmosphere interactions are always
538
intensive in instrumentation and labor. To conserve costs, long-term observatory operations
539
are commonly based on well-proven, mostly automated measurement systems,
540
concentrated on a small number of locations. Such systems constitute the long-term
541
backbone to build understanding of interactions and feedbacks between the atmosphere
542
(from turbulence to climatic scales) and ecosystems (from photosynthesis to the life-cycle of
543
vegetation). In contrast, short-term intensive campaigns are useful to pursue specific
544
research goals with an all-out and focused effort. Past examples of intensive campaigns
545
have shown them to be fertile spawning-grounds for collaboration and research innovation.
546
The ScaleX concept combines the benefits of both long-term monitoring and short-term
547
intensive approaches. It uses an integrated TERENO long-term observatory site, with its
548
backbone infrastructure, logistics and long-term expertise, as the staging area for repeated
549
short intensive campaigns. The continuity of the backbone measurements and the broad
550
spectrum of campaign observations complement each other.
551
In the coming months and years, more comprehensive and interdisciplinary analyses of
552
ScaleX-2015 data are anticipated, beyond the few examples presented here, and these will
553
likely lead to new insights on scale interactions of ecosystem – atmosphere processes in
554
complex terrain. In such analyses, modeling activities from single process models to fully
555
coupled regional climate models or Large Eddy Simulation systems will play important roles
556
as scale-integrators, and to pinpoint process interrelations and feedback mechanisms that
557
are reflected in the data. Vice-versa, enhanced ScaleX and TERENO data products will likely
558
serve for model performance evaluations at a wide range of scales and applications. For
559
26
example, Hingerl et al. (2016) used energy balance measurements from TERENO-preAlpine
560
for the evaluation and analysis of the distribution of water- and energy fluxes over the Rott
561
catchment, computed by GEOtop, a distributed water- and energy-balance model for
562
complex terrain (http://www.geotop.org).
563
An important aspect of the ScaleX concept is that the same study area will be re-visited by
564
recurrent campaigns. After its inception in 2015, the second installment of ScaleX is set for
565
the summer of 2016 and periodic further ScaleX campaigns are planned throughout the life-
566
time of TERENO. Data from these campaigns are stored in an on-line ScaleX cloud that is
567
freely available to all partners collaborating in measurements, modeling or data-mining. For
568
access to the cloud, visit the ScaleX web site (www.scalex.imk-ifu.kit.edu), or contact the
569
corresponding author.
570
As we progress, we hope that ScaleX will become more integrated in terms of in-situ
571
observations, remote sensing and modeling, and that the spectrum of observations
572
continues to grow towards a more comprehensive modeling test-bed for processes at the
573
interface of soils, vegetation and the atmosphere. To follow up and go beyond the specific
574
research questions of ScaleX-2015 (see Section 2), we invite expertise on specific topics for
575
future ScaleX campaigns, including (i) the contribution of advective terms to EC flux
576
measurements; (ii) simulation of farm-animal related emissions to derive CH4 source
577
strengths at catchment scale; (iii) atmospheric transport modeling to link point sources to
578
background emission, and NBL concentration profiles of trace gases; (iv) miniaturisation of
579
trace gas sensors for UAVs; (v) space-borne and airborne remote sensing estimates of soil
580
moisture and precipitation, land cover, elevation and biomass productivity; (vi) advanced,
581
coupled soil-vegetation-atmosphere exchange modeling. Over time, the group of scientists
582
27
and institutions that participate in ScaleX is expected to evolve as well as the topical foci,
583
and thus, this paper is an invitation to collaborate in future ScaleX campaigns, and to
584
emulate the ScaleX concept at other long-term observatory sites.
585
586
28
587
Acknowledgements
588
Research at KIT/IMK-IFU for TERENO-preAlpine and ScaleX is funded, in part, by the
589
Helmholtz Association and its program ATMO (Atmosphere and Climate), through grants
590
from the German Federal Ministry of Education and Research (BMBF). The TERENO activities
591
of ATMO are integrated in the German climate research initiative REKLIM. MM, CM, FDR
592
and MZ were supported by the Helmholtz Young Investigator Group “Capturing all relevant
593
scales of biosphere-atmosphere exchange – the enigmatic energy balance closure problem”,
594
and by KIT. We want to thank Dr. Jörg Seltmann and Dr. Michael Frech (both of DWD) for
595
assistance with the C-band weather radar data, Dr. Christoph Münkel (of Vaisala) for
596
support with the ceilometer data processing, and the IMK-IFU technical staff and numerous
597
student field assistants for their engaged work in the field at all hours. We are indebted to
598
Mr. Anton Jungwirth, farmer at Fendt, for providing access to the field site DE-Fen, and for
599
his great flexibility and tolerance during the ScaleX campaign.
600
601
29
References
602
Baldocchi, D., 2003: Assessing the eddy covariance technique for evaluating carbon dioxide
603
exchange rates of ecosystems: past, present and future. Glob. Chang. Biol., 9, 479–492,
604
doi:10.1046/j.1365-2486.2003.00629.x.
605
Beniston, M., 2006: Mountain weather and climate: A general overview and a focus on
606
climatic change in the Alps. Hydrobiologia, 562, 3–16, doi:10.1007/s10750-005-1802-0.
607
Best, M. J., and Coauthors, 2015: The plumbing of land surface models: benchmarking
608
model performance. J. Hydrometeorol., 16, 1425–1442, doi:10.1175/JHM-D-14-0158.1.
609
Boden, T. A., M. Krassovski, and B. Yang, 2013: Geoscientific Instrumentation Methods and
610
Data Systems The AmeriFlux data activity and data system: an evolving collection of
611
data management techniques, tools, products and services. Geosci. Instrum. Method.
612
Data Syst. Discuss, 3, 59–85, doi:10.5194/gid-3-59-2013.
613
Bogena, H., 2010: Potential of wireless sensor networks for measuring soil water content
614
variability. Vadose Zo. Journal2, 9, 1002–1013, doi:10.2136/vzj2009.0173.
615
Böhm, R., I. Auer, M. Brunetti, M. Maugeri, T. Nanni, and W. Schöner, 2001: Regional
616
temperature variability in the european Alps : 1760 – 1998 from homogenized
617
instrumental time series. Int. J. Climatol., 21, 1779–1801, doi:10.1002/joc.689.
618
Bünzli, D., and H. P. Schmid, 1998: The influence of surface texture on regionally aggregated
619
evaporation and energy partitioning. J. Atmos. Sci., 55, 961–972, doi:10.1175/1520-
620
0469(1998)055<0961:TIOSTO>2.0.CO;2.
621
Calanca, P., 2007: Climate change and drought occurrence in the Alpine region: How severe
622
are becoming the extremes? Glob. Planet. Change, 57, 151–160,
623
30
doi:10.1016/j.gloplacha.2006.11.001.
624
Caldwell, M. M., P. A. Matson, C. Wessman, and J. Gamon, 1993: Prospects for scaling.
625
Scaling Physiological Processes: Leaf to Globe, J.R. Ehleringer and C.B. Field, Eds.,
626
Academic Press, San Diego, 223–230.
627
Clark, M. P., and Coauthors, 2015: Improving the representation of hydrologic processes in
628
Earth System Models. Water Resour. Res., 51, 5929–5956,
629
doi:10.1002/2015WR017096.
630
Clough, T. J., F. M. Kelliher, R. R. Sherlock, and C. D. Ford, 2004: Lime and Soil Moisture
631
Effects on Nitrous Oxide Emissions from a Urine Patch. Soil Sci. Soc. Am. J., 68, 1600–
632
1609, doi:10.2136/sssaj2004.1600.
633
Coplen, T., A. Herczeg, and C. Barnes, 2000: Isotope Engineering—Using Stable Isotopes of
634
the Water Molecule to Solve Practical Problems. Environ. Tracers Subsurf. Hydrol., 79–
635
110, doi:10.1007/978-1-4615-4557-6_3.
636
Denmead, O. T., M. R. Raupach, F. X. Dunin, H. A. Cleugh, and R. Leuning, 1996: Boundary
637
layer budgets for regional estimates of scalar fluxes. Glob. Chang. Biol., 2, 255–264,
638
doi:10.1111/j.1365-2486.1996.tb00077.x.
639
Desai, A. R., and Coauthors, 2011: Seasonal pattern of regional carbon balance in the central
640
Rocky Mountains from surface and airborne measurements. J. Geophys. Res.
641
Biogeosciences, 116, 1–17, doi:10.1029/2011JG001655.
642
Ehleringer, J. R., and C. B. Field, eds., 1993: Scaling physiological processes: leaf to globe.
643
Academic Press,.
644
Emeis, S., 2008: Examples for the determination of turbulent (sub-synoptic) fluxes with
645
31
inverse methods. Meteorol. Zeitschrift, 17, 3–11, doi:10.1127/0941-2948/2008/0265.
646
——, 2015: Observational techniques to assist the coupling of CWE/CFD models and meso-
647
scale meteorological models. J. Wind Eng. Ind. Aerodyn., 144, 24–30,
648
doi:10.1016/j.jweia.2015.04.018.
649
——, K. Schäfer, and C. Münkel, 2009: Observation of the structure of the urban boundary
650
layer with different ceilometers and validation by RASS data. Meteorol. Zeitschrift, 18,
651
149–154, doi:10.1127/0941-2948/2009/0365.
652
Finnigan, J. J., 2004: The footprint concept in complex terrain. Agric. For. Meteorol., 127,
653
117–129, doi:10.1016/j.agrformet.2004.07.008.
654
Firestein, S., 2012: Ignorance: How it drives science. Oxford University Press, USA, 208 pp.
655
Goring, S. J., and Coauthors, 2014: Improving the culture of interdisciplinary collaboration in
656
ecology by expanding measures of success. Front. Ecol. Environ., 12, 39–47,
657
doi:10.1890/120370.
658
Hall, F. G., 1999: Introduction to special section: BOREAS in 1999: Experiment and science
659
overview. J. Geophys. Res., 104, 27627–27639, doi:10.1029/1999JD901026.
660
Hingerl, L., H. Kunstmann, S. Wagner, M. Mauder, J. Bliefernicht, and R. Rigon, 2016:
661
Spatiotemporal variability of water and energy fluxes - A case study for a mesoscale
662
catchment in pre-alpine environment. Hydrol. Process., accepted.
663
Hinkle, S. R., J. H. Duff, F. J. Triska, A. Laenen, E. B. Gates, K. E. Bencala, D. A. Wentz, and S.
664
R. Silva, 2001: Linking hyporheic flow and nitrogen cycling near the Willamette River - A
665
large river in Oregon, USA. J. Hydrol., 244, 157–180, doi:10.1016/S0022-
666
1694(01)00335-3.
667
32
Hommeltenberg, J., M. Mauder, M. Drösler, K. Heidbach, P. Werle, and H. P. Schmid, 2014:
668
Ecosystem scale methane fluxes in a natural temperate bog-pine forest in southern
669
Germany. Agric. For. Meteorol., 198-199, 273–284,
670
doi:10.1016/j.agrformet.2014.08.017.
671
Jones, J. B., and R. M. Holmes, 1996: Surface-subsurface interactions in stream ecosystems.
672
Trends Ecol. Evol., 11, 239–242, doi:10.1016/0169-5347(96)10013-6.
673
Junkermann, W., 2001: An ultralight aircraft as platform for research in the lower
674
troposphere: System performance and first results from radiation transfer studies in
675
stratiform aerosol layers and broken cloud conditions. J. Atmos. Ocean. Technol., 18,
676
934–946, doi:10.1175/1520-0426(2001)018<0934:AUAAPF>2.0.CO;2.
677
——, R. Hagemann, and B. Vogel, 2011: Nucleation in the Karlsruhe plume during the
678
COPS/TRACKS-Lagrange experiment. Q. J. R. Meteorol. Soc., 137, 267–274,
679
doi:10.1002/qj.753.
680
Kalbus, E., F. Reinstorf, and M. Schirmer, 2006: Measuring methods for groundwater -
681
surface water interactions: a review. Hydrol. Earth Syst. Sci., 10, 873–887,
682
doi:10.5194/hess-10-873-2006.
683
Kampe, T., B. R. Johnson, M. Kuester, and M. Keller, 2010: NEON: the first continental-scale
684
ecological observatory with airborne remote sensing of vegetation canopy
685
biochemistry and structure. J. Appl. Remote Sens., 4, 043510, doi:10.1117/1.3361375.
686
Köhli, M., M. Schrön, M. Zreda, U. Schmidt, P. Dietrich, and S. Zacharias, 2015: Footprint
687
characteristics revised for field-scale soil moisture minitoring with cosmic-ray neutrons.
688
Water Resour. Res., 51, 5772–5790, doi:10.1002/2014WR016259.
689
33
Krajewski, W. F., G. J. Ciach, and E. Habib, 2003: An analysis of small-scale rainfall variability
690
in different climatic regimes. Hydrol. Sci. J., 48, 151–162,
691
doi:10.1623/hysj.48.2.151.44694.
692
Levin, S. A., 1992: The problem of pattern and scale in ecology. Ecology, 73, 1943–1967,
693
doi:10.2307/1941447.
694
Lorenz, C., and H. Kunstmann, 2012: The hydrological cycle in three state-of-the-art
695
reanalyses: Intercomparison and performance analysis. J. Hydrometeorol., 13, 1397–
696
1420, doi:10.1175/JHM-D-11-088.1.
697
Mahrt, L., 1987: Grid-Averaged Surface Fluxes. Mon. Weather Rev., 115, 1550–1560,
698
doi:10.1175/1520-0493(1987)115<1550:GASF>2.0.CO;2.
699
Mahrt, L., 2010: Computing turbulent fluxes near the surface: Needed improvements. Agric.
700
For. Meteorol., 150, 501–509, doi:10.1016/j.agrformet.2010.01.015.
701
Mauder, M., R. L. Desjardins, E. Pattey, and D. Worth, 2010: An Attempt to Close the
702
Daytime Surface Energy Balance Using Spatially-Averaged Flux Measurements.
703
Boundary-Layer Meteorol., 136, 175–191, doi:10.1007/s10546-010-9497-9.
704
——, M. Cuntz, C. Drüe, A. Graf, C. Rebmann, H. P. Schmid, M. Schmidt, and R. Steinbrecher,
705
2013: A strategy for quality and uncertainty assessment of long-term eddy-covariance
706
measurements. Agric. For. Meteorol., 169, 122–135,
707
doi:10.1016/j.agrformet.2012.09.006.
708
May, R., 1999: Unanswered questions in ecology. Philos. Trans. R. Soc. Lond. B. Biol. Sci.,
709
354, 1951–1959, doi:10.1098/rstb.1999.0534.
710
Metzger, S., and Coauthors, 2013: Spatially explicit regionalization of airborne flux
711
34
measurements using environmental response functions. Biogeosciences, 10, 2193–
712
2217, doi:10.5194/bg-10-2193-2013.
713
Mills, R. T. E., K. S. Gavazov, T. Spiegelberger, D. Johnson, and A. Buttler, 2014: Diminished
714
soil functions occur under simulated climate change in a sup-alpine pasture, but
715
heterotrophic temperature sensitivity indicates microbial resilience. Sci. Total Environ.,
716
473-474, 465–472, doi:10.1016/j.scitotenv.2013.12.071.
717
Orlanski, I., 1975: A rational subdivision of scales for atmospheric processes. Bull. Am.
718
Meteorol. Soc., 56, 527–530.
719
Osmond, B., and Coauthors, 2004: Changing the way we think about global change research:
720
Scaling up in experimental ecosystem science. Glob. Chang. Biol., 10, 393–407,
721
doi:10.1111/j.1529-8817.2003.00747.x.
722
Pebesma, E. J., 2004: Multivariable geostatistics in S: The gstat package. Comput. Geosci.,
723
30, 683–691, doi:10.1016/j.cargo.2004.03.012.
724
Pihlatie, M., and Coauthors, 2013: Comparison of static chambers to measure CH4 emissions
725
from soils. Agric. For. Meteorol., 171-172, 124–136,
726
doi:10.1016/j.agrformet.2012.11.008.
727
Pihlatie, M. K., and Coauthors, 2010: Greenhouse gas fluxes in a drained peatland forest
728
during spring frost-thaw event. Biogeosciences, 7, 1715–1727, doi:10.5194/bg-7-1715-
729
2010.
730
Poulos, M. J., J. L. Pierce, A. N. Flores, and S. G. Benner, 2012: Hillslope asymmetry maps
731
reveal widespread, multi-scale organization. Geophys. Res. Lett., 39, 1–6,
732
doi:10.1029/2012GL051283.
733
35
Reid, W. V, and Coauthors, 2005: Millenium Ecosystem Assessment: Ecosystems and Well-
734
being: Synthesis. Island Press, Washington, DC, 155 pp.
735
Ruehr, N. K., B. E. Law, D. Quandt, and M. Williams, 2014: Effects of heat and drought on
736
carbon and water dynamics in a regenerating semi-arid pine forest: A combined
737
experimental and modeling approach. Biogeosciences, 11, 4139–4156, doi:10.5194/bg-
738
11-4139-2014.
739
Schäfer, K., J. Böttcher, D. Weymann, C. von der Heide, and W. H. M. Duijnisveld, 2012:
740
Evaluation of a Closed Tunnel for Field-Scale Measurements of Nitrous Oxide Fluxes
741
from an Unfertilized Grassland Soil. J. Environ. Qual., 41, 1383,
742
doi:10.2134/jeq2011.0475.
743
Schmid, H. P., 2002: Footprint modeling for vegetation atmosphere exchange studies: a
744
review and perspective. Agric. For. Meteorol., 113, 159–183, doi:10.1016/S0168-
745
1923(02)00107-7.
746
——, and C. R. Lloyd, 1999: Spatial representativeness and the location bias of flux
747
footprints over inhomogeneous areas. Agric. For. Meteorol., 93, 195–209,
748
doi:10.1016/S0168-1923(98)00119-1.
749
Sellers, P. J., F. G. Hall, G. Asrar, D. E. Strebel, and R. E. Murphy, 1988: The first ISLSCP Field
750
Experiment (FIFE). Bull. Am. Meteorol. Soc., 69, 22–27, doi:10.1007/BF00138905.
751
——, ——, ——, ——, and ——, 1992: An Overview of the First International Satellite Land
752
Surface Climatology Project (ISLSCP) Field Experiment ( FIFE ). J. Geophys. Res., 97,
753
18345–18371.
754
——, and Coauthors, 1995: The Boreal Ecosystem-Atmosphere Study (BOREAS): An
755
36
Overview and Early Results form the 1994 Field Year. Bull. Am. Meteorol. Soc., 76,
756
1549–1577, doi:10.1175/1520-0477(1995)076<1549:TBESAO>2.0.CO;2.
757
Smiatek, G., H. Kunstmann, R. Knoche, and A. Marx, 2009: Precipitation and temperature
758
statistics in high-resolution regional climate models: Evaluation for the European Alps.
759
J. Geophys. Res., 114, 1–16, doi:D19107\r10.1029/2008jd011353.
760
Sophocleous, M., 2002: Interactions between groundwater and surface water: The state of
761
the science. Hydrogeol. J., 10, 52–67, doi:10.1007/s10040-001-0170-8.
762
Stoy, P. C., and Coauthors, 2013: A data-driven analysis of energy balance closure across
763
FLUXNET research sites: The role of landscape scale heterogeneity. Agric. For.
764
Meteorol., 171-172, 137–152, doi:10.1016/j.agrformet.2012.11.004.
765
Sun, J., and Coauthors, 2010: A multiscale and multidisciplinary investigation of ecosystem-
766
atmosphere CO2 exchange over the rocky mountains of colorado. Bull. Am. Meteorol.
767
Soc., 91, 209–230, doi:10.1175/2009BAMS2733.1.
768
Tetzlaff, D., J. Seibert, K. J. McGuire, H. Laudon, D. A. Burns, S. M. Dunn, and C. Soulsby,
769
2009: How does landscape strucutre influcence catchment transit tiem across different
770
geomorphic provinces. Hydrol. Process., 23, 945–953, doi:10.1002/hyp.
771
Uhlenbrook, S., M. Frey, C. Leibundgut, and P. Maloszewski, 2002: Hydrograph separations
772
in a mesoscale mountainous basin at event and seasonal timescales. Water Resour.
773
Res., 38, 31–1 – 31–14, doi:10.1029/2001WR000938.
774
Yao, Z., X. Wu, B. Wolf, M. Dannenmann, K. Butterbach-Bahl, N. Brueggemann, W. W. Chen,
775
and X. H. Zheng, 2010: Soil-atmosphere exchange potential of NO and N2O in different
776
land use types of Inner Mongolia as affected by soil temperature, soil moisture, freeze-
777
37
thaw, and drying-wetting events. J. Geophys. Res., 115, doi:doi:10.1029/2009JD013528.
778
Zacharias, S., and Coauthors, 2011: A Network of Terrestrial Environmental Observatories in
779
Germany. Vadose Zo. J., 10, 955, doi:10.2136/vzj2010.0139.
780
Zreda, M., D. Desilets, T. P. A. Ferre, and R. L. Scott, 2008: Measuring soil moisture content
781
non-invasively at intermediate spatial scale using cosmic-ray neutrons. Geophys. Res.
782
Lett., 35, L21402, doi:10.1029/2008GL035655.
783
784
38
785
Tables
786
Table 1: Permanent and campaign instrumentation at DE-Fen site, available during ScaleX-2015 (1 June to 31 July). Deployment dates are given
787
for intermittent measurements only. See Fig. 1 and 2 for deployment locations.
788
PERMANENT (principal components of TERENO-preAlpine backbone instrumentation at DE-Fen site)
Instrument / Installation
(number, if multiple)
Determined Quantity
Models &
Manufacturers
(principal components
only)
EC Flux station
(+ supporting micromet)
CO2, latent heat and sensible heat fluxes (supporting micromet including: short
and long wave radiation components, PAR, soil heat flux, soil moisture)
CSAT-3 (a); LI-7500 (b)
Ceilometer
aerosol backscatter for ABL-height estimation (15 min resolution)
CL51 (c)
SoilNetFen network
(55 locations, 3 depths)
soil volumetric water content (capacitance/frequency domain technology), soil
water potential, soil temperature (15 min resolution)
SMT-100 (d)
Cosmic ray neutron sensor
field scale top soil water content
CRS-2000/B (e)
Discharge gauges (2)
river discharge (Thomson V-notch weir)
(by IMK-IFU) (f)
Groundwater wells (5)
groundwater head
(by IMK-IFU) (f)
Rain gauge
Precipitation
Pluvio2 (g)
X-band radar
radar reflectivity and precipitation
RAINSCANNER (h)
Lysimeter cluster with
static dark chamber
system (18)
groundwater recharge, evapotranspiration, CO2, CH4 and N2O fluxes
Science Lysimeter (i);
Dual Laser Trace Gas
Monitor (j)
ADDITIONAL ScaleX-2015 Instrumentation
Radio acoustic sounding
(RASS)
wind & temperature profile, vertical velocity variance, range: 20 - 560 m, 10
min means
482 MHz RASS (k)
Doppler LIDAR (3)
3-D wind and turbulence profile, range up to 1000 m in 18 m increments, 1 to
Stream Line (l)
39
3 min means
Passive microwave and
infrared radiometer
Temperature and humidity profiles, integrated water vapor (IWV) and liquid
water path (LWP) and cloud base temperature
HATPRO (m)
Hexacopter
payload sensors for: relative humidity, air temperature, air pressure;
deployment dates: 24, 25 and 30 June, 1, 10, 15, 16 20, 21, 23 and 30 July
Hexacopter F550 (n)
Quadrocopter swarm (3)
payload sensors for: relative humidity, air temperature, air pressure;
deployment dates: 30 June, 1 July, 6 August
Saphira(v), Autoquad
M4(w)
Fixed wing UAVs (3)
payload sensors for: relative humidity, air temperature, air pressure, wind;
deployment dates: 30 June, 1 and 15 July, 6 August
(by IGUA) (f)
Microlight aircraft D-MIFU
Temperature, dewpoint and aerosol profiles, turbulent fluxes, radiation (UV-
IR); deployment dates: 5, 12, 25 and 26 June, 4, 7, 10, 15, 16 and 22 July
(for/by IMK-IFU) (f)
Rain gauges (groups of 3)
precipitation amount; 5 groups at DE-Fen, 17 groups within Rott catchment
Rain Collector (o)
DWD C-band radar
Spatial information on precipitation amount and hydrometeor types
Dual-Pole Doppler C-
band weather radar (by
DWD) (f)
Micro rain radar
vertical profiles of rain rate, drop size distribution
MRR (k)
Disdrometers (2)
drop size distribution, rain rate
Parsivel (g); and LNM (p)
Cavity ring down (CRD)
spectrometer
isotopic composition (18O-H2O and 2H-H2O) of precipitation, groundwater and
streamflow
L1102-I (q)
TERENO Rover
soil water content; vehicle-based CRNS
CRS-1000 (e)
F-SAR
top soil water content; (one overflight during ScaleX-2015)
L-band SAR (by DLR) (f)
Big chamber
CH4 soil flux; static chamber principle (dimensions: 10 m x 2.60 m, max. height
0.61 m); deployment dates: 9, 16, 25, 26 and 30 June, 10, 14, 20, 21, 23, 28
and 30 July
(by IMK-IFU) (f)
Trace Gas Analyzer
CH4 and H2O concentrations
Fast Methane Analyzer (r)
Wind sensor network
(3 locations)
Wind and turbulence (profile at 1(s), 5(s), 10 (a) m, location H in Figure 2; two
stations (s) (t) at 3 m height, locations A and K in Figure 2)
CSAT-3 (a); WindSonic (s);
81000 (t)
CRD spectrometer
CH4, N2O and CO2 concentrations
G2508 (q)
Open path methane
analyzer
Line averaged methane mixing ratios
Gas Finder 2 (u)
40
Manufacturers: (a): Campbell Scientific, Logan UT (USA); (b): LI-COR, Lincoln, NE (USA); (c): Vaisala, Helsinki (Finnland); (d): TRUEBNER Instruments, Neustadt (Germany);
789
(e): Hydroinnova LLC, Albuquerque, NM (USA); (f): in-house or custom built; (g): OTT Hydromet, Kempten (Germany); (h): Selex ES GmbH, Neuss (Germany); (i): UMS,
790
Munich (Germany); (j): Aerodyne Research, Billerica, MA (USA); (k): METEK GmbH, Elmshorn (Germany); (l): Halo Photonics, Worcestershire (UK); (m): Radiometer Physics
791
GmbH, Meckenheim (Germany); (n): DJI, Beijing (China); (o): Davis Instruments, Haward, CA (USA); (p): Thies Clima, Göttingen (Germany); (q): Picarro Inc., Santa Clara, CA
792
(USA); (r): Los Gatos Research, San Jose, CA (USA); (s): Gill Instruments, Lymington, UK; (t): RM Young, Traverse City, MI (USA); (u): Boreal Laser Inc., Edmonton, AB
793
(Canada); ; (v): rOsewhite Multicopter, Mauerstetten (Germany); (w): distributed by iRC-Electronic, Wehringen (Germany)
794
795
41
Figure Captions
796
797
Fig. 1. Location of the TERENO-preAlpine observatory between Innsbruck (Austria) and
798
Munich (Germany) (left). The map on the right shows the southern Ammer catchment (black
799
boundary) and the northern Rott catchment (grey boundary), with the three principal sites
800
(black rectangles), precipitation gauges (red dots), X-band rain radar (red triangle) and the
801
meteorological observatory MOHP (black asterisk). See text for details. The red square
802
indicates the ScaleX-2015 study area presented in Fig. 2. Color bars show elevation in m
803
a.s.l. The maps were produced using Copernicus data and information funded by the
804
European Union – EU-DEM layers (uploaded 10/08/2003) and the ATKIS stream network.
805
806
Fig. 2. (left panel) The ScaleX study area centered around DE-Fen (black square) with
807
topographic features (colors encode elevation in m a.s.l.), catchment boundaries (Rott in
808
grey and Ammer in black) and MOHP. Streams and lakes (blue) are shown for the Rott
809
catchment only. (right panel) Map (approx. 1000 m x 1000 m) of land cover (see map),
810
installations, water ways and roads at DE-Fen with additions of SoilNetFen nodes (black
811
crosses), precipitation gauges (red dots), groundwater wells (brown triangles) and discharge
812
weirs (brown dots). The marks A to K represent the locations of, A: the remote sensing hub,
813
B: CRNS, C: EC station, D: automatic stream water sampler, E: RASS, F: big chamber, G:
814
lysimeter cluster, H: 10 m tower, I: 3D-Doppler LIDARs, J: nearby farm, and K: open path
815
methane analyzer. Sonic anemometers at locations A, K, and a profile at H constituted the
816
wind sensor network (see Table 1). Abbreviations are explained in Table 1 and in the text.
817
The maps were produced using EU-DEM, Corine Land Cover 2006 (The European Topic
818
Centre on Spatial Information and Analysis, uploaded 8 April 2014, permalink SH04UZP80M)
819
42
and OpenStreetMap information (www.geofabrik.de, downloaded Jan 2015).
820
821
Fig. 3. (left panel) Schematic illustration of the 3D Doppler LIDAR setup used to observe a
822
vertical profile of 3-D wind vectors (virtual tower). The schematic is superimposed on a
823
terrain representation of the DE-Fen site. (right panel) Profiles of vertical and horizontal
824
wind speed and wind direction at the ScaleX virtual tower on 1 July 2015. Positive (negative)
825
vertical wind speed indicates upward (downward) motion. Vertical axes are in m a.g.l.
826
827
Fig. 4. CH4 concentrations (± std. dev., 1 minute means) measured in 1 and 10 m height on
828
the tower, and by the hexacopter at 10, 25 and 50 m on 21 July 2015. To improve legibility
829
of the data at 25 and 50 m, lines were added to connect the measurements at these levels.
830
Local standard time is UTC+1.
831
832
Fig. 5. (left panel) First 300 m of vertical air temperature (Ta) profiles determined by the
833
hexacopter (blue colors), fixed wing UAV (yellow and orange) and D-MIFU (grey) on 15 July
834
2015 (start times given in UTC, local standard time is UTC+1). The right panel shows profile
835
flight tracks of D-MIFU (white), fixed-wing (yellow) and hexacopter (blue).
836
837
Fig. 6. Example for the high spatial variability of hourly rainfall in the region of the Rott
838
catchment (1600 UTC 27 June 2015) recorded by the rain gauge network (color of filled
839
circles) and the DWD C-band weather radar (colored map). The color scale is given in mm of
840
hourly precipitation.
841
43
842
Fig. 7. Volumetric water content (VWC) for 1 July 2015, derived from SoilNetFen at the DE-
843
Fen site for different depths and from the CRNS Rover. Gray lines represent roads and
844
tracks, the Rott creek is printed in blue. The SoilNetFen profiles are marked by white
845
crosses. The southern part of the SoilNetFen area was not accessible to the Rover. The “eye”
846
structures in some regions of the maps are likely artifacts from the simple distance-
847
weighting interpolation method used.
848
849
Fig. 8. Rainfall intensity and stream discharge measured at the location of the automatic
850
water sampler (A), isotopic composition of precipitation, stream water and groundwater (B),
851
and groundwater level (C) during the ScaleX campaign 2015. Gaps in streamwater isotopic
852
composition were caused by instrumental failure.
853
854
44
855
Figures
856
Fig. 1. Location of the TERENO-preAlpine observatory between Innsbruck (Austria) and Munich
(Germany) (left). The map on the right shows the southern Ammer catchment (black boundary) and
the northern Rott catchment (grey boundary), with the three principal sites (black rectangles),
precipitation gauges (red dots), X-band rain radar (red triangle) and the meteorological observatory
MOHP (black asterisk). See text for details. The red square indicates the ScaleX-2015 study area
presented in Fig. 2. Color bars show elevation in m a.s.l. The maps were produced using Copernicus
data and information funded by the European Union – EU-DEM layers (uploaded 10/08/2003) and
the ATKIS stream network.
857
45
858
Fig. 2. (left panel) The ScaleX study area centered around DE-Fen (black square) with topographic
features (colors encode elevation in m a.s.l.), catchment boundaries (Rott in grey and Ammer in
black) and MOHP. Streams and lakes (blue) are shown for the Rott catchment only. (right panel) Map
(approx. 1000 m x 1000 m) of land cover (see map), installations, water ways and roads at DE-Fen
with additions of SoilNetFen nodes (black crosses), precipitation gauges (red dots), groundwater
wells (brown triangles) and discharge weirs (brown dots). The marks A to K represent the locations
of, A: the remote sensing hub, B: CRNS, C: EC station, D: automatic stream water sampler, E: RASS, F:
big chamber, G: lysimeter cluster, H: 10 m tower, I: 3D-Doppler LIDARs, J: nearby farm, and K: open
path methane analyzer. Sonic anemometers at locations A, K, and a profile at H constituted the wind
sensor network (see Table 1). Abbreviations are explained in Table 1 and in the text. The maps were
produced using EU-DEM, Corine Land Cover 2006 (The European Topic Centre on Spatial Information
and Analysis, uploaded 8 April 2014, permalink SH04UZP80M) and OpenStreetMap information
(www.geofabrik.de, downloaded Jan 2015).
859
46
860
Fig. 3. (left panel) Schematic illustration of the 3D Doppler LIDAR setup used to observe a vertical
profile of 3-D wind vectors (virtual tower). The schematic is superimposed on a terrain
representation of the DE-Fen site. (right panel) Profiles of vertical and horizontal wind speed and
wind direction at the ScaleX virtual tower on 1 July 2015. Positive (negative) vertical wind speed
indicates upward (downward) motion. Vertical axes are in m a.g.l.
861
862
47
863
864
Fig. 4. CH4 concentrations (± std. dev., 1 minute means) measured in 1 and 10 m height on the
tower, and by the hexacopter at 10, 25 and 50 m on 21 July 2015. To improve legibility of the data
at 25 and 50 m, lines were added to connect the measurements at these levels. Local standard
time is UTC+1.
865
866
48
867
868
Fig. 5. (left panel) First 300 m of vertical air temperature (Ta) profiles determined by the hexacopter
(blue colors), fixed wing UAV (yellow and orange) and D-MIFU (grey) on 15 July 2015 (start times
given in UTC, local standard time is UTC+1). The right panel shows profile flight tracks of D-MIFU
(white), fixed-wing (yellow) and hexacopter (blue).
869
870
49
871
872
Fig. 6. Example for the high spatial variability of hourly rainfall in the region of the Rott catchment
(1600 UTC 27 June 2015) recorded by the rain gauge network (color of filled circles) and the DWD
C-band weather radar (colored map). The color scale is given in mm of hourly precipitation.
873
50
Fig. 7. Volumetric water content (VWC) for 1 July 2015, derived from SoilNetFen at the DE-Fen site for
different depths and from the CRNS Rover. Gray lines represent roads and tracks, the Rott creek is
printed in blue. The SoilNetFen profiles are marked by white crosses. The southern part of the
SoilNetFen area was not accessible to the Rover. The “eye” structures in some regions of the maps
are likely artifacts from the simple distance-weighting interpolation method used.
874
875
51
Fig. 8. Rainfall intensity and stream discharge measured at the location of the automatic water
sampler (A), isotopic composition of precipitation, stream water and groundwater (B), and
groundwater level (C) during the ScaleX campaign 2015. Gaps in streamwater isotopic composition
were caused by instrumental failure.
876
