remote sensing  
Review
Understanding Forest Health with Remote
Sensing-Part II—A Review of Approaches
and Data Models
Angela Lausch 1,2,*, Stefan Erasmi 3, Douglas J. King 4, Paul Magdon 5 and Marco Heurich 6
1 Department Computational Landscape Ecology, Helmholtz Centre for Environmental Research—UFZ,
Permoserstr. 15, Leipzig D-04318, Germany
2 Department of Geography, Lab for Landscape Ecology, Humboldt Universität zu Berlin,
Rudower Chaussee 16, 12489 Berlin, Germany
3 Cartography GIS & Remote Sensing Section, Institute of Geography, Georg–August–University Göttingen,
Goldschmidtstr. 5, Göttingen D-37077, Germany; serasmi@gwdg.de
4 Geomatics and Landscape Ecology Lab, Department of Geography and Environmental Studies,
Carleton University, 1125 Colonel By Drive, Ottawa, ON K1S 5B6, Canada; doug.king@carleton.ca
5 Chair of Forest Inventory and Remote Sensing, Georg-August-University Göttingen, Büsgenweg 5,
Göttingen D-37077, Germany; pmagdon@gwdg.de
6 Bavarian Forest National Park, Department of Conservation and Research, Freyunger Straße 2,
Grafenau D-94481, Germany; marco.heurich@npv-bw.bayern.de
* Correspondence: angela.lausch@ufz.de; Tel.: +49-341-235-1961; Fax: +49-341-235-1939
Academic Editors: Lars T. Waser, Clement Atzberger, Richard Gloaguen and Prasad S. Thenkabail
Received: 11 September 2016; Accepted: 23 January 2017; Published: 5 February 2017
Abstract: Stress in forest ecosystems (FES) occurs as a result of land-use intensification, disturbances,
resource limitations or unsustainable management, causing changes in forest health (FH) at various
scales from the local to the global scale. Reactions to such stress depend on the phylogeny of forest
species or communities and the characteristics of their impacting drivers and processes. There are
many approaches to monitor indicators of FH using in-situ forest inventory and experimental
studies, but they are generally limited to sample points or small areas, as well as being time- and
labour-intensive. Long-term monitoring based on forest inventories provides valuable information
about changes and trends of FH. However, abrupt short-term changes cannot sufficiently be
assessed through in-situ forest inventories as they usually have repetition periods of multiple years.
Furthermore, numerous FH indicators monitored in in-situ surveys are based on expert judgement.
Remote sensing (RS) technologies offer means to monitor FH indicators in an effective, repetitive and
comparative way. This paper reviews techniques that are currently used for monitoring, including
close-range RS, airborne and satellite approaches. The implementation of optical, RADAR and
LiDAR RS-techniques to assess spectral traits/spectral trait variations (ST/STV) is described in detail.
We found that ST/STV can be used to record indicators of FH based on RS. Therefore, the ST/STV
approach provides a framework to develop a standardized monitoring concept for FH indicators
using RS techniques that is applicable to future monitoring programs. It is only through linking
in-situ and RS approaches that we will be able to improve our understanding of the relationship
between stressors, and the associated spectral responses in order to develop robust FH indicators.
Keywords: spectral traits (ST); spectral trait variations (STV); in-situ; remote sensing (RS) approaches;
plant phenomics facilities; wireless sensor networks (WSN); RADAR; optical; LiDAR; RS models
1. Introduction
There is a growing awareness of the multitude of ecosystem services that forests provide and their
importance for the wellbeing of humans and conservation of the environment. Forest ecosystems (FES)
Remote Sens. 2017, 9, 129; doi:10.3390/rs9020129 www.mdpi.com/journal/remotesensing
Remote Sens. 2017, 9, 129 2 of 33
are well known as hosts of biodiversity, carbon and oxygen pools, sources of timber, water purification
systems and providers of livelihoods, and yet an increasing number of anthropogenic threats endanger
FES. While direct stressors are a result of forest utilization or land use changes, indirect stress is caused
by drivers such as climate change, air pollution and globalization, which in turn may bring about an
additional impact from invasive species [1].
Due to the crucial importance of forests, it is of high relevance to better understand how different
stressors affect forest conditions and how these mechanisms interact with the type, structure and
history of the forest. It is only then that forest management and conservation strategies can be adapted
to maintain and improve forest health (FH) and to cope with new and upcoming threats. For both, it is
important to develop monitoring techniques, which provide information on the past and present state
of FH. Such information forms the basis for research on the interactions between forest conditions
and stressors. This is particularly relevant and urgent for FES where growth and regeneration can
take centuries. Thus, today’s management decisions will determine the shape of our forests for
future generations. To monitor and understand the complexity of FH, its entities, drivers, functional
mechanisms and impacts, it is imperative to develop effective standardized large scale long-term
monitoring systems.
FH monitoring (FHM) has a long history and is carried out by private forest owners, forest
administrations and other public entities at local, national and international levels. There are numerous
in-situ monitoring programs at the local and regional scales that increasingly utilize standardized
indicators of FH. Among the largest monitoring programs are the FOREST EUROPE program of the
European Union and the International Co-operative Programme on Assessment and Monitoring of
Air Pollution Effects on Forests (EU/ICP Forests) which started in 1990 [2], the FHM program of
the United States [3] and the Chinese national program on ecological functions [4]. While in some
countries, FHM belongs to independent monitoring programs, other countries include the assessment
of FH in their national forest inventories [5]. Table 1 provides an overview on current long-term FHM
monitoring programs that use in-situ methods.
Table 1. Long-term monitoring of forest health (FH) at country, European and global level.
Level/Scale Responsible Body Description Resources/Links
Country level:
Germany
Johann Heinrich von
Thünen-Institute
Forest condition monitoring (FCM).
[6]
Level-I-Monitoring.
Frequency: Annual.
Compilation of national reports on forest
conditions for Germany (FCA). FH assessed
using systematic sample grid of permanent plots.
Federal Research
Institute for countryside,
forests and fisheries
Intensive monitoring.
Level-II Monitoring.
Frequency: Continuous.
66 sites intensively monitored, partly through
continual sampling of relevant ecosystem
compartments in selected FES.
Federal Ministry for
Food and Agriculture
National forest inventory.
[7]
Level-III-Monitoring.
Frequency: Every 10 years.
Status and development of the forests of
Germany derived from a sample based large
scale forest inventory.
Country level:
USA
United States
Department of
Agriculture(USDA)
Forest Service
FH Monitoring (M).
[8]Frequency: Annual.
National program designed to determine the
status, changes, and trends in indicators of forest
condition on an annual basis.
Remote Sens. 2017, 9, 129 3 of 33
Table 1. Cont.
Level/Scale Responsible Body Description Resources/Links
Country level:
Canada
Canadian Forest
Service (CFS)
National FHM Network.
[9]Frequency: 5-years.
Established 1994 based on earlier Acid Rain
Monitoring Network. Plot-based Bi-annual to
5-year repetition depending on the variables.
National Forest
Inventory (NFI), Canada
National Forest Inventory.
[10]Frequency: 10-years.
Mixed plot and RS based. Includes inventory
parameters + assessment of insect, disease, fire
and other disturbance damage.
European level
United Nations
Economic Commission
for Europe (UNECE)
ICP.
[11]Frequency: Annual.
International Co-operative Programme on the
Assessment and Monitoring of Air Pollution
Effects on Forest. Developed to standardise the
recording of different FH indicators on three
levels of intensity.
Global level
Food and Agriculture
Organization of the
United Nations (FAO)
Forest Resources Assessment.
[12]Frequency: 5-years.
FH recorded by the FAO as part of the Forest
Resources Assessment (FRA). Individual
countries report their findings to the FAO, which
then compiles a report.
FH as such cannot be observed or monitored directly but needs to be assessed using indicators
which are then combined to create a holistic picture of the forest condition. Different initiatives and
programs have developed FH indicators. For example, FHM conducted by the US Forest Service
collects variables on lichen communities, forest soils, tree crown conditions, vegetation diversity and
structure, downed woody materials and ozone damage [13]. The FOREST EUROPE program defines
7 criteria to describe the status of European forests, where criterion 3 analyses the “Maintenance of
Forest Ecosystem Health and Vitality” [2]. Both programs use sample based in-situ assessments, where
the set of indicators is either directly observed (e.g., amount of deadwood, species composition) or
quantified by expert judgement (e.g., defoliation and discoloration classes). The quality and consistency
of such assessments depend on the expertise of the field teams, which can limit comparability between
assessments. Thus, for various important FH indicators such as tree crown density, defoliation,
“naturalness” or regeneration potential, there are still no standardized metrics nor direct measuring
procedures available. However, these in-situ FH assessments are regularly conducted and have proven
to be useful to detect changes in FH conditions over longer periods. Remote sensing (RS) techniques are
regularly used for forest monitoring. However, in the context of operational national and international
FHM programs, RS currently only plays a minor role. To provide but one example: in the 364-page
report for 46 European countries, the term “remote sensing” occurs only once where it is stated that
RS and geographic information systems are new technologies that will be used by some countries [2].
Pause et al. [14] concluded that a major advantage of implementing remote-sensing techniques is the
possibility of repeatedly acquiring standardized information over large areas at low costs with high
frequency. However, RS techniques are only helpful to improve the objectivity of FHM when they can
be reliably linked to FH indicators. Furthermore, a combined terrestrial and remote-sensing monitoring
system not only serves to monitor changes in forest conditions, but also to relate the changes to the
drivers [15].
In the first part of this paper series, Lausch et al. [16] provided a working definition of FH and
reviewed FH indicators, as well as their importance and relevance for the development of FHM
programs. They introduced the concept of Spectral Traits (ST) and spectral trait variations (STV) as a
Remote Sens. 2017, 9, 129 4 of 33
framework for FHM. In this second paper, we review existing remote-sensing systems and evaluate
their potential for monitoring specific FH indicators. This paper addresses the following questions:
1. Which factors are important when designing FHM programs that combine terrestrial and
remote-sensing data?
2. Which remote sensors and systems are suitable for monitoring which FH indicators?
3. Which new technologies and current developments are relevant for the design of future
FHM programs?
In the introduction, we provide an overview of the current close-range FHM programs to identify
potential links between in situ observations and RS data. Section 2 reviews the trends and new
technologies for close-range RS approaches. Section 3 analyses air- and space-borne RS systems
for estimating and monitoring FH indicators. In Section 4, we describe approaches of physical and
empirical models followed by conclusions in Section 5.
2. Trends in Close-Range RS Approaches for Assessing FH
Traits and trait variations in FES display specific spectral responses in RS data that are caused by
the taxonomy, phenology and phylogenetics of forest species and communities as well as by natural
and anthropogenic stressors [16,17]. The spectrum of responses ranges from additive reinforcing effects
to contrary effects, reductive or even exclusive signals. To correctly interpret these spectral response
signals in terms of disturbance or stress, as well as to gain sufficient information for calibrating and
validating airborne and space-borne RS data, extensive close-range RS measurements (Figure 1, Table 2)
as well as physically-based models (Section 5) are needed.
Remote Sens. 2017, 9, 129  4 of 32 
 
In the introduction, we provide an overview of the current close-range FHM programs to 
identify potential links between in situ observations and RS data. Section 2 reviews the trends and 
new technologies for close-range RS approaches. Section 3 analyses air- and space-borne RS systems 
for estimating and monitoring FH indicators. In Section 4, we describe approaches of physical and 
empirical models followed by conclusions in Section 5.  
2. Trends in Close-Range RS Approaches for Assessing FH 
Traits and trait va iations in FES display spec fic spectral responses in RS data that are caused 
by the taxo omy, phe ology and phylogenetics of f rest speci  and communities as well as by 
natural and anthropog nic stressors [16,17]. The spectrum of responses ranges from additive 
reinforcing effects to contrary effects, reductive or even exclusive signals. To correctly interpret these 
spectral response signals in terms of disturbance or stress, as well as to gain sufficient information 
for calibrating nd v lidating airborne and space-borne RS data, extensive clos -r ng  RS 
m asurements (Figure 1, Table 2) as well as physically-based models (Section 5) are needed. 
I -situ sampling of forest canopy elements for measurement of leaf properties such as chlorophyll, 
protein, water content, leaf area index (LAI) or functional plant traits such as photosynthesis activity is 
difficult compared to sampling of ground vegetation. Recording this kind of informatio  often 
requires tree climbers, lifting platforms or towers, Unmanned Arial Vehicle (UAV) with arms for 
cutting small branches or rifles or harpoons to shoot down leaf material (Figure 1). 
 
Figure 1. Methods and materials for sampling canopy materials in forest, (a) Tree climbers, photo by 
Michael Lender, (b) Cherry picker, photo by Franz Baierl, (c) Crossbow, photos by Zhihui Wang and, 
(d) Unmanned Aerial Vehicle (UAV). 
In addition to UAVs, aircrafts and space-borne RS applications, close-range RS approaches are 
increasingly being established to support the observation of FH indicators as shown in Figure 2. 
They range from spectral analyses of forest species to terrestrial wireless sensor networks as 
summarized in Figure 2 and Table 2. 
i . t t i l f li t i l i f t, ( ) li , t
i l ; i , i l; ; i i ,
ri l ehicle (UAV).
In-situ sampling of forest canopy elements for measurement of leaf properties such as chlorophyll,
protein, water content, leaf area index (LAI) or functional plant traits such as photosynthesis activity is
difficult compared to sampling of ground vegetation. Recording this kind of information often requires
tree climbers, lifting platforms or towers, Unmanned Arial Vehicle (UAV) with arms for cutting small
branches or rifles or harpoons to shoot down leaf material (Figure 1).
In addition to UAVs, aircrafts and space-borne RS applications, close-range RS approaches are
increasingly being established to support the observation of FH indicators as shown in Figure 2.
They range from spectral analyses of forest species to terrestrial wireless sensor networks as
summarized in Figure 2 and Table 2.
Remote Sens. 2017, 9, 129 5 of 33
Remote Sens. 2017, 9, 129  5 of 32 
 
 
Figure 2. Overview of different close-range RS methods to analyse indicators of FH, (a) Laboratory 
spectrometer, (b,c) Ash trees monitored in a close-range RS spectral laboratory (manual) with 
imaging hyperspectral sensors AISA EAGLE/HAWK (modified after Brosinksy et al. [18],  
(d) Automated plant phenomics facilities (e) Field-spectrometer measurements, (f) Wireless sensor 
networks—WSN, (g) One sensor node of the WSN (graphic, photo (f,g)) by Jan Bumberger and 
Hannes Mollenhauer), (h) instrumentation of the Hohes Holz forest site (modified by Wöllschläger et al. [19] 
(i) tower with different RS instruments, (j) mobile crane with RS measurement platform (modified 
after Clasen et al. [20]). 
Table 2. Technical and methodological options for close-range RS to observe FH indicators. 
Close-Range Measurement 
Approaches/References 
Advantages/Applications Disadvantages 
Field Spectrometers [21–24] 
Basis for research on spectral characteristics of 
biochemical-biophysical, morphological traits. Spectral 
databases for classification and validation. Basis for 
research on taxonomic, phylogenetic, genetic, epigenetic 
or morphological-functional features. 
Analysis at molecular 
level. Geometric, 
structural, distribution, 
population and 
community effects are not 
measurable. No 
standardized measurement 
protocols available 
Spectral laboratory (Manual 
operation) [18,25–29]. 
Seasonal, annual, long-term. Biochemical-biophysical, 
structural variables in organs (roots, leaf, stem) and whole 
tree. Experimental stress analyses (drought, heavy metals, 
tropospheric ozone, flooding, nitrogen loads, etc.). 
Extensive lab-based measurement program for biotic, 
abiotic, climate conditions. Comparative analyses can be 
conducted under natural or artificial conditions. 
Multi-sensor recording at specific plant development 
stages. Storage in spectral databases for validation and 
calibration. 
Development of measuring 
boxes for the sensors 
(automated). Age and 
development stages of the 
trees are a limiting factor 
(often only trees up to 5 
years old can be recorded) 
Plant phenomics facilities (Fully 
automatic operation), Ecotrons 
(Controlled environmental 
facility), [30–48]. 
Tower (eddy flux tower) with 
different non-invasive measuring 
technologies as well as RS 
technology (mobile, permanently 
installed), [49–52]. 
Long-term monitoring. International networks exist. 
Extensive multi sensor monitoring is possible for biotic 
and abiotic conditions (e.g., phenocams). Spectral 
measurements directly on canopy level. 
Local results for a 
particular site, not 
transferable 
Wireless sensor networks 
(WSN) [53–62] 
Long-term high frequency monitoring. Extensive multi 
sensor measurement is possible. Measuring various 
biochemical-biophysical, structural variables in organs 
(roots, leaf, stem) and whole tree. Enables results over 
more extensive areas. Easy to install in remote areas. 
Primarily non-imaging 
sensor technology can be 
implemented 
Figure 2. Overview of different close-range RS ethods to analyse indicators of FH, (a) Laboratory
spectrometer; (b,c) Ash trees monitored in a close-range RS spectral laboratory (manual) with imaging
hyperspectral sensors AISA EAGLE/HAWK (modified after Brosinksy et al. [18]; (d) Automated
plant phenomics facilities (e) Field-spectrometer measurements; (f) Wireless sensor networks—WSN;
(g) One sensor node of the WSN (graphic, photo (f;g)) by Jan Bumberger and Hannes Mollenhauer);
(h) instrumentation of the Hohes Holz forest site (modified by Wöllschläger et al. [19] (i) tower with
different RS instruments, (j) mobile crane with RS measurement platform (modified after Clasen et al. [20]).
Table 2. Technical and methodological options for close-range RS to observe FH indicators.
Close-Range Measurement
Approaches/R f rences Adv ntages/Applications Disadvantages
Field Spectrometers [21–24]
Basis for research on spectral characteristics of
biochemical-biophysical, morphological traits.
Spectral d tabases for cl ssification and
validation. Basis for resea c n taxonomic,
phylogenetic, genetic, epigenetic or
morphological-functional features.
Analysis at molecular level.
Geometric, structural, distribution,
population and community effects
re not measurable.
No standardized measurement
protocols available
Spectral laboratory (Manual
oper tion) [18,25–29].
Plant phenomics facilities (Fully
automatic operation), Ecotrons
(Controlled environmental
facility), [30–48].
Seasonal, annual, long-term.
Biochemical-biophysical, structural variables in
organs (roots, leaf, stem) and whole tree.
Experimental stress analyses (drought, heavy
metals, tropospheric ozone, flooding, nitrogen
loads, etc.). Extensive lab-based measurement
program for biotic, abiotic, climate conditions.
Comparative analyses can be conducted under
natural or artificial conditions. Multi-sensor
recording at specific plant development stages.
Storage in spectral databases for validati
and calibration.
Development of measuring boxes
for the sensors (automated).
Age and develop ent stag s of
the trees are a limiting factor
(often only trees up to 5 years old
can be recorded)
Tower (eddy flux tower) with
different non-invasive measuring
technologies as well as RS
technology (mobile, permanently
installed), [49–52].
Long-term monitoring. In ernational networks
exist. Extensive multi sensor monitoring is
possible for biotic and abiotic conditions
(e.g., phenocams). Spectral measurements
directly on canopy level.
Local results for a particular site,
not transferable
Wireless sensor networks
(WSN) [53–62]
Long-term high frequency monitoring.
Extensive multi sensor measurement is possible.
Measuring various biochemical-biophysical,
structural variables in organs (roots, leaf, stem)
and whole tree. Enables results over more
extensive areas. Easy to install in remote areas.
Primarily non-imaging sensor
technology can be implemented
Remote Sens. 2017, 9, 129 6 of 33
2.1. Close-Range RS Approaches—Spectral Laboratory, Plant Phenomics Facilities and Ecotrons
Close-range investigations are conducted to determine the spectral responses of FH traits and to
derive meaningful models from them (see Table 2). Reactions of woody plants to stress are species
specific [59,63]. Teodoro et al. [59] analysed different strategies of Brazilian forest tree species to cope
with drought stress and maintain plant functioning. As reactions of woody plants to stress factors like
drought can often only be observed years later in the form of biochemical, physiological or geometrical
changes to woody plant traits [27], the ability of different tree species to adapt to climate change is still
not well understood [64].
With the help of close-range RS approaches, extensive long-term stress monitoring can be
carried out that takes into account vegetation phenological cycles as well as inter-annual variations.
To specifically investigate different stress factors, experimental settings are best suited as they provide
reliable input for models and eliminate confounding factors. Brosinsky et al. [18] investigated the
spectral response of ash trees (Fraxinus excelsior L.) to physiological stress from flooding over a
3-month period, whereas Buddenbaum et al. [27] modelled the photosynthesis rate of young European
beech trees under drought stress following a two-year water stress treatment by using close-range
hyperspectral visible, near infrared and thermal sensors. Another common method to induce water
stress is the girdling of trees with differing degrees of intensity [65,66].
The Genotype comprises hereditary information about the DNA of a species while the phenotype
represents the physiological, morphological, anatomical, development characteristics and the
interactions of species with their environmental conditions, resource limitations and stress factors.
Insights into genotype and phenotype interactions in plant stress physiology are not only gained
from recording individual plant ST, but also by including the entire genotype-epigenetic-phenotypic-
environment matrix [67]. This can be achieved, for example, by recording phenotypical plant traits in
plant phenomic facilities [30,34] or controlled environmental facilities—Ecotron [38]. The acquisition
of phenotype plant traits in plant phenomics facilities has so far mainly been carried out for crop
vegetation [33,34,68]. For woody plants, preliminary investigations have been restricted to fruit
trees [36]. Phenotypical investigation of woody plants should therefore be augmented in the future, in
order to be able to better understand the mechanisms and stress factor interactions involved when
interpreting FH indicators observed using RS. It is these kinds of comparisons that support research
on the effects of the Bidirectional Reflectance Distribution Function (BRDF), scale and different RS
platforms as well as different sensor characteristics on model design [69].
The regulation of photosynthetic electron transport, as well as its feedback processes and the
assessment of photosynthetic efficiency and stress in crops, grassland and forest canopies, have been
investigated with sun-induced chlorophyll fluorescence methods at leaf [41], field, and regional
levels [42–45]. Spectroscopic techniques for the detection of chlorophyll fluorescence forms the basis
for the European Space Agency (ESA) Fluorescence Explorer Sensors (FLEX, [45,46,70]) to be launched
in 2018. With its very high spectral resolution across the 0.3–3.0 µm range, FLEX will be the first
satellite that is able to directly measure solar-induced chlorophyll fluorescence and thus forest stress
indicators and other vegetation parameters.
Disadvantages of close-range RS laboratory methods include the use of artificial light sources and
the short distance between the sensor and object. Substantial image pre-processing and calibration
is often required to compare lab and field measurements. However, given the experimental setting,
BRDF effects can be observed with high sensitivity and the spatial resolutions are much finer (mm)
than with airborne- and space-borne RS. Thus, experimental spectral laboratories allow continuous
investigation of phenotypical features as well as a comprehensive understanding of the spectral
response to stress-physiological parameters in forest vegetation. Table 3 summarizes findings from
exemplary close-range RS studies of FH that illustrate the points made above.
Remote Sens. 2017, 9, 129 7 of 33
Table 3. Example studies illustrating the use of spectral laboratory, plant phenomics and controlled
environmental facilities in common FH applications.
Application Example Studies Main Findings
Genotype-epigenetic and
phenotype interactions
[30,34]
Qualitative, quantitative and spectroscopic recording of plant
species phenotypes for better understanding of the link
between the genotype and the phenotype.
[67]
Understanding the impacts and resilience to stress,
disturbance or resourcelimitations of forest species and
ecosystems is crucial for understanding the
genotype-epigenetic-phenotypic-environment matrix.
Goal of the Plant
Phenotyping Network [32,71]
International Plant Phenotyping Network [72]: (1) Innovative
non-invasive techniques such as stereo systems, hyperspectral,
RGB, thermal, fluorescence cameras, laser scanners or X-ray
tomography; (2) Continuous, very high temporal resolution
acquisition of phenotypical traits that provides important
reference information for RS approaches; (3) ST/STV are
saved in databases; (4) Data can be used for calibration and
validation of air- and spaceborne RS data.
Spectral traits of leaves [27]
Laboratory-based imaging spectroscopy combined with the
inversion of a radiative transfer model is able to derive
biochemical spectral traits (N, chlorophyll content,
carotenoids, brown pigments, water content, dry mass) at the
leaf level on the sub-millimeter scale.
Development and testing
of new close-range RS
technologies
[41–45]
Investigations of sun-induced chlorophyll fluorescence
methods at the leaf level, field level, and the regional level are
only possible through fundamental research in Plant
Phenomics Facilities and Ecotrons.
[46–48]
Evaluated the effects of different plant stresses on
photosynthetic performance. This research on chlorophyll
fluorescence and its acquisition using spectroscopic techniques
forms the basis for developing the European Space Agency
(ESA) Fluorescence Explorer Sensors (FLEX, [45,46,70]).
[47]
Development of 3-D digital imaging and a portable terrestrial
laser scanner for detecting seasonal change within
broad-leaved forest.
[73] Development of a canopy leaf area density profile.
[48] 3D-imaging techniques for monitoring the spatio-temporaleffects of herbicides on plants.
Monitoring of stress to
woody plants [27]
Reactions to stress factors like drought, can often only be
observed years later in the form of biochemical, physiological
or geometrical changes to woody plant traits (e.g., in tree rings
observed from cross cuttings).
Age and development stages of the trees are a limiting factor
(often only trees up to the age of 5 years can be recorded).
2.2. Close-Range RS Approaches—Towers
In order to upscale findings from experimental laboratory settings, which are often for leaf
scale analysis, close-range RS can be implemented using mobile towers [20] or permanently installed
towers [49] equipped with RS sensors to analyze spectral effects on the canopy level. (Table 4, Figure 2).
Such systems may include automated digital time-lapse cameras (e.g., phenocams [51]), multispectral
and thermal sensors, and non-imaging spectral sensors to monitor the status and changes of FH
indicators over long periods of time at regular intervals. The US National Ecological Observatory
Network (NEON) and the European Union’s Integrated Carbon Observation System (ICOS) are
continental-scale ecological research networks that promote the international implementation, use,
standardization as well as analysis of phenocam data [51].
Flux towers generally include integrated sampling of different ecosystem variables such as carbon
dioxide, water vapour and energy fluxes. They are often coupled with sensor technologies such
Remote Sens. 2017, 9, 129 8 of 33
as spectrometers or soil sensors. Such combinations allow continuous measurements of vegetation
productivity and environmental variables such as fluxes, soil moisture, and evapotranspiration, which
can be assessed in relation to FH attributes. Permanently installed towers acquire individual point
information but are of particular importance in terms of long-term measurements for the calibration
and validation of airborne and space-borne RS data. Linking flux tower systems in an international
network (FLUXNET, [50]) supports a better understanding of ecological processes and changes in FH
using RS [74,75].
Table 4. Example studies illustrating the use of towers with different RS technology (mobile, permanently
installed) in common applications.
Application Example Studies Main Findings
Phenocam
networks
[51]
US National Ecological Observatory Network (NEON) and the
European Union’s Integrated Carbon Observation System (ICOS).
Fully-automated digital time-lapse cameras (phenocams) and other
cameras can be easily mounted on towers. They are crucial sensors
for recording, quantifying, monitoring and understanding of
phenological traits and the interactions of ST/STV relations to stress,
disturbances and resource limitations in forest ecosystems.
Individual
towers [52]
Fully automated spectral data recording system for phyto-pigments
(chlorophyll, carotenoids, anthocyanins) under different view and
sun angles. Used to assess diurnal and seasonal variations of plant
physiological processes under different illumination and weather
conditions. High spatial resolution allows measurement of spectral
response of individual tree crowns. Systematic recording of ST/STV
can be linked to eddy covariance gas exchange measurements.
Flux tower
networks [49,50]
Linking flux towers in an international network (FLUXNET, [76]).
Flux towers generally include integrated sampling of ecosystem
parameters such as carbon dioxide, water vapour and energy fluxes,
as they cycle through the atmosphere, vegetation and soil. FLUX
towers are often coupled with the sensor technologies such as
spectrometers or soil sensors.
Spectral
networks [77]
Spectral network (SpecNet, [78]). Multi-scale spectral RS from
satellites, aircraft, UAVs, mobile tram systems, portable
spectrometers over same area as flux measurements. The goals of
SpecNet are: (1) Monitoring surface–atmosphere fluxes of water,
carbon and vapor; (2) Understanding and assessing the impacts of
disturbance and dynamic events (e.g., fire, extreme weather events,
climate, land-use change).
2.3. Close-Range RS Approaches—Wireless Sensor Networks (WSN)
Wireless sensor networks (WSN) consist of a potentially large set of sensors that communicate with
each other based on wireless communication channels and standardized protocols. They can collect
data on various environmental variables which can be used to describe complex ecosystem forest
processes continually in a non-invasive, cost-effective, automated and real-time manner [53,79–81].
The design of the network, e.g., the spatial separation of the network nodes (sensor nodes), depends
on the spatial variability of the environmental variables and ranges from centimetres to a maximum of
about 1km depending on the wireless communication technology used.
In the context of FHM, WSNs are implemented for the detection and verification of forest fires in
real time [53,55]. Brum et al. [60] used WSNs to demonstrate the effects of the 2015 El-Niño extreme
drought on the sap flow of trees in eastern Amazonia. Oliveira et al. [58] used WSNs to record
important processes of soil-plant-atmosphere interactions in tropical montane cloud forests in Brazil.
They investigated how key forest ecosystem processes such as transpiration, carbon uptake and
storage, and water stripping from clouds are affected by climatic variation and the temporal and
Remote Sens. 2017, 9, 129 9 of 33
spatial forest structure. Moreover, Teodoro et al. [59] and Oliveira et al. [58] used WSNs to demonstrate
the interplay between hydraulic traits, growth performance and the stomata regulation capacity in
three shrub species in a tropical montane scrubland of Brazil under contrasting water availability.
The results showed that forest plant species employ different strategies in the regulation of hydraulic
and stomatal conductivity during drought stress and thus substantiate the need for setting up WSN
for different forest tree species and communities [59] (see Table 5). Innovative developments, ongoing
miniaturisation, the cost-effective production of different non-invasive environmental sensors [61,62]
as well as the latest data analysis and network technologies (e.g., semantics [82] and Linked Open
Data (LOD, [83]) are all promising factors that promote the implementation of WSN as part of
FHM programs.
Table 5. Example studies illustrating the use of wireless sensor networks (WSN) in applications.
Application Example Studies Main Findings
Forest fire
detection [53,55]
WSNs are implemented for the detection and verification of forest fires
in real time.
Drought stress [60] WSNs used to demonstrate the effects of the 2015 El-Niño extremedrought on the sap flow of trees in eastern Amazonia.
Understanding
physiological
and ecological
processes
[58]
Useful in recording and understanding important processes of
soil-plant-atmosphere interactions in tropical montane cloud forests in
Brazil, key forest ecosystem processes such as transpiration, carbon
uptake and storage, and water stripping from clouds that are affected
by climatic variation and the temporal and spatial forest structure.
[59]
Important in monitoring and understanding hydraulic traits, growth
performance and the stomata regulation capacity in three shrub species
in a tropical montane scrubland of Brazil under contrasting water
availability. The results showed that forest plant species employ
different strategies in the regulation of hydraulic and stomatal
conductivity during drought stress and thus substantiate the need for
setting up WSN for different forest tree species and communities.
3. Trends in Air-and Space-Borne RS for Assessing FH
RS has advanced rapidly in recent years, including widespread adoption of LiDAR, RADAR,
hyperspectral and integrated systems for a variety of forest applications. This section reviews these
sensors types and their applicability in analysis and monitoring.
3.1. Light Detection and Ranging (LiDAR)
LiDAR is an active RS technique, where short pulses of laser light are distributed from a scanning
device across a wide area and their reflections from different objects are subsequently recorded by the
sensor. The result is a set of 3D points, which represents the scanned surfaces from where the pulses
were reflected. More detailed descriptions of LiDAR technology can be found in [84,85].
Nowadays, laser-based instruments are used on all kinds of RS platforms, including terrestrial,
UAVs, the well established airborne LiDAR scanning and satellite-based laser-based instrument
(e.g., Geoscience Laser Altimeter System (GLAS), [86]).
The primary characteristic that makes LiDAR well suited for applications to monitor FH is the
potential to reconstruct 3D forest structures within and below the canopy, which cannot be provided
by passive remote-sensing techniques [87]. Therefore, LiDAR RS is an important component when
it comes to monitoring FH characteristics related to canopy and tree structure and it adds a further
dimension to the properties of optical RS.
Based on the capabilities of data recording, it is possible to distinguish between discrete return
systems and full waveform systems. At the onset of LiDAR development, the sensors were only able
to record either the first or the last reflection of the LiDAR beam, which in FES is generally the top
Remote Sens. 2017, 9, 129 10 of 33
of the trees (first reflection) and the terrain (last reflection). With the more recently developed full
waveform systems, the whole pathway of the LiDAR beam through the canopy can be detected and
recorded [88]. Moreover, additional echo attributes such as amplitude and intensity of the return
signal can be provided, which can support the classification process. As a result, the characterization
of the canopy structure is much more detailed with full waveform data [89], so that important FH
indicators such as forest height and understory cover can be estimated with a lower bias and higher
consistency [90].
For FHM applications, LiDAR systems are generally used that have a beam footprint of less than
one-meter diameter on the ground. These so-called small footprint systems are preferred, because
they provide a good link between the LiDAR beam and the structural vegetation attributes that subtly
change as a consequence of stress or damage, sometimes within individual trees. By comparison, large
footprint systems have beam diameters with up to several tens of meters on the ground; e.g., the GLAS
instrument mounted on the Ice, Cloud and Land Elevation Satellite ICESat Platform with a footprint
of 38 m [91]. Such systems can be used to model broad forest structural attributes but can only detect
significant and widespread changes in forest structural condition or health [92].
The most important environmental application of LiDAR is the precise mapping of terrain and
surface/canopy elevations. Such digital terrain models (DTMs) or digital surface models (DSMs)
can be useful in monitoring FH indicators such as changes in tree height resulting from damage or
deviations in growth due to stress [93–95]. Recent studies show that it is even possible to detect
objects located on the ground surface such as coarse woody debris [96,97]. Coarse woody debris is an
important indicator of FH, because it provides habitat to a multitude of endangered plant and animal
species and plays an important role in the forest carbon cycle [98,99].
Because of its characteristics, LiDAR is well suited for measuring forest biophysical parameters,
such as tree dimensions and canopy properties. Two main approaches have been developed over
recent years. The area-based approach is a straightforward methodology where the height distribution
of the LiDAR beam reflections is analyzed for a given area [100]. In a first step, different “LiDAR
metrics”, such as the maximum height or fractional cover are calculated for each area. In a second step,
these metrics are set in relation to conventional ground measurements of FH indicators such as Above
Ground Biomass (AGB) or stand density for model calibration. In a final step, the models are used to
predict the selected FH indicators for large areas using square grid cells. Such analysis is generally
conducted using a priori stratification of stand types and tree species. In the years that followed, this
methodology was proven to be able to determine key biophysical forest variables on a larger scale.
So far, this method has been shown to deliver a precision of 4%–8% for height, 6%–12% for mean stem
diameter, 9%–12% for basal area, 17%–22% for stem density and 11%–14% for volume estimations
of boreal forests [100–103]. Because of the high accuracy of estimation of important FH parameters
the area-based approach was further developed and adopted to operational forest inventories in
boreal forests of Scandinavia [104]. For the temperate zone similar accuracies have also been achieved,
although the presence of more complex forest structures, especially a higher number of tree species
and a higher amount of standing dead wood leads to less accurate estimations and requires more effort
in stratification and ground measurement to obtain species-specific results [105–107].
The second methodology is the individual tree approach, which has the objective of extracting
single trees and modeling their properties. The procedure consists of four steps: (1) delineation
of individual trees; (2) extraction of tree parameters such as height, species and crown parameters;
(3) model calibration of the biophysical parameters: Diameter at Breast Height (DBH), volume and
biomass with the help of reference trees measured on the ground; and finally (4) the application of
models to predict DBH, volume and biomass for all delineated LiDAR trees. Based on the crown
representations basic attributes reflecting tree health such as total volume, crown length, crown area
and crown base height can be derived [108,109]. The extracted individual trees also form the basis
for identifying the tree species. Within the 2D or 3D representation of the individual tree point cloud
and waveform features are calculated and with the help of classification techniques, tree species
Remote Sens. 2017, 9, 129 11 of 33
can be determined [90,110]. While the differentiation between deciduous and coniferous trees can
be performed with a high accuracy (>80%, up to ~97%) the differentiation within these classes is
more difficult and leads to a higher classification error [111]. Moreover, it is possible to distinguish
between living trees, standing dead trees and snags [112–114] or to map dead trees on the plot or stand
level [115,116]. However, 3D LiDAR has its limitations for differentiating between trees species and
dead trees unless it is not combined with multi- and hyperspectral optical data. One drawback of
the single tree approach is that the LiDAR beam loses some signal strength on its way through the
canopy. This results in an underestimation of individual trees in the understory [117]. To overcome
this problem, methods have been developed to predict stem diameter distributions of forest stands
based on the detectable trees in the upper canopy and LiDAR derived information about the vertical
forest structure and density [118].
Besides the traditional parameters related to forest inventory, a multitude of FH parameters
describing the ecological conditions of the forest can be estimated with LiDAR sensors. One key
element to assess FH is the canopy cover which is defined as the projection of the tree crowns onto the
ground divided by ground surface area [119]. This parameter can be easily translated to LiDAR data
by dividing the number of returns measured above a certain height threshold by the total number of
returns. Many studies have shown a strong (R2 > 0.7) relationship between this LiDAR-metric and
ground measurements [120]. By using hemispherical images or LAI-2000 sensor data for calibration,
LAI and solar radiation can also be derived from LiDAR data with a high precision over large
areas [108,121].
The vertical forest structure is of high relevance for the description of forest heterogeneity and
for biodiversity assessments [122,123]. A widely used LiDAR metric to represent vertical canopy
complexity is the coefficient of variation. Higher values correspond to more diverse multi-layer stands,
while low values represent single layer stands [124]. The coefficient of variation can be applied to the
point clouds, the digital crown model or individual trees. Zimble et al. [125] applied this principle
and classified forest stands according to stand structure with LiDAR data and achieved an overall
accuracy of 97%. Another approach is the partitioning of the vertical structure into different height
layers in relation to ecological importance. Latifi et al. [107] divided the canopy into height layers
according to phytosociological standards and found a strong relationship with different LiDAR metrics
using regression models. Similar approaches were used by Vogler et al. [126] and Ewald et al, [127] to
represent the understory relevant for birds and deer and to detect forest regeneration [128]. A more
recent study applied a 3D segmentation algorithm to estimate regeneration cover and achieved an
accuracy of 70% [129]. LiDAR-derived information about the vertical structure is also used for the
assessment of forest fuels and their vertical distribution, which are important input variables in forest
fire models supporting fire management [130,131].
As a result, LiDAR-RS is a powerful tool for monitoring FH indicators. It delivers detailed and
accurate information about forest properties down to the scale of the individual tree. Nowadays,
LiDAR is widely applied in RS research as a reference to test the accuracy of other methods as well
as in practical forest inventories in the boreal zone of Scandinavia [102]. The development of new
sensors will lead to multi- or hyperspectral LiDAR technology, which will combine the advantages
of today’s LiDAR and optical sensors. These systems will be able to collect accurate 3D information
and calibrated spectral information without facing the problems of varying illumination in the tree
crowns. Furthermore, the resolution of the data will increase, enabling parameter extraction at the
branch level [132].
Until now the application of LiDAR has been relatively costly because it is limited to airborne
missions covering local to regional areas; currently there are no active space-borne missions providing
data for mapping forest vegetation. The Global Ecosystem Dynamics Investigation (GEDI)–3D-LiDAR
of NASA, which is scheduled for 2019, will provide measurements of 3D vegetation structures which
can be used to model vertical stand structure, biomass, disturbance and recovery on a regional to a
global scale. Although a footprint size of 25 m will not enable the high resolution mapping that we are
Remote Sens. 2017, 9, 129 12 of 33
familiar with from small footprint LiDAR, the scientific objectives of the mission include modelling
finer scale structure [133]. A combination of Landsat, EnMAP, FLEX, HySPIRI and GEDI LiDAR will
combine structural and spectral information and improve the modelling, prediction and understanding
of FH [134–139]. Examples of the main findings from LiDAR studies in forest analysis are given in
Table 6.
Table 6. Example studies illustrating the use of LiDAR in common FH applications.
Application Example Studies Main Findings
Terrain
determination (DTM) [140–143]
Best remote sensing technique to produce high accuracy DTM´s
under dense forest canopy with an RMSE of 0.15–0.35 m.
Forest height
measurement (DCM) [93–95]
Best remote sensing technique to produce high accuracy forest
height measurements.
Area based approach [100–107]
Statistical approach where measurements of field plots were set in
relation to LiDAR-metrics. Precision of ~6% for height, ~10% for
mean diameter, ~10% for basal area, ~20% for stem density and
~12% for AGB estimations.
Individual tree
approach [112,141,144–146]
In a first step, single trees were automatically delineated from the
point cloud and in a second step tree parameters were estimated.
With these methods about 80% to 97% of the trees in the upper
canopy can be detected and height, crown parameters, DBH,
volume, species and health conditions estimated.
Coarse woody debris [96,97,113,114] Detection of standing and laying coarse woody debris on a singletree basis from LiDAR and in combination with optical data
Leafe area index,
canopy cover [108,119,120]
For high quality results, calibration with hemispherical
photography or LAI-2000 measurements are needed. But even
without calibration, fairly reliable results can be obtained for
fractional cover.
Vertical forest
structure [125–129,144]
LiDAR data is widely used to represent vertical forest canopy
complexity and forest regeneration. These variables have great
potential to predict species diversity.
3.2. RADAR
Synthetic Aperture RADAR (SAR) uses an antenna to transmit microwave pulses in a specific
waveband (or frequency) at an oblique (incident) angle to the target area [147]. RADAR data can be
acquired in a variety of modes, including standard polarizations (e.g., HH, VV, HV, VH), polarimetric
(phase information is preserved [148]), and interferometric (InSAR; phase processing/analysis of two
signals at slightly different incident angles [149]). These data can be analyzed for: (1) backscatter
intensity response, which increases with surface moisture and roughness, but also varies with incident
angle (steeper angle signals penetrate further into the canopy, with associated increases in volume
and perhaps surface scattering), wavelength [150] (longer wavelengths penetrate deeper), polarization
(cross polarizations (HV; VH) penetrate deeper), and topography (slopes perpendicular to the incident
signal generally produce the highest backscatter [151]); (2) polarimetric response, including phase
related variables, degree of depolarization, and decomposition analysis [152–155] to determine relative
contributions of scattering mechanisms (e.g., surface, dihedral, volume scattering); and (3) InSAR phase
differences, which are strongly related to the vertical position of the height of the scattering phase center
(hspc) in vegetation or to surface elevation. For the latter, in dense forest and for shorter wavelengths
(X, C-bands), hspc is closer to the top of the canopy but it is also dependent on system parameters
(e.g., wavelength, baseline length and attitude between the two signals, polarization, incident angle,
phase noise [156] and slope [157]). An InSAR generated DSM can be differenced from an accurate
DTM (e.g., from long wavelength InSAR [158,159] or LiDAR) to estimate tree height [157,160,161] and
potentially height/cover loss due to disturbance. Integration of polarimetric response with InSAR
(i.e., PolInSAR) has also been successful [159]. In addition, the coherence or correlation between the
Remote Sens. 2017, 9, 129 13 of 33
two phase components declines if changes in the forest have occurred. Lastly, the RADAR signal is
also subject to constructive and destructive interference due to multiple scattering elements within a
resolution cell, producing an essentially random cell to cell backscatter variation (speckle) [162] that
can be reduced using multiple looks or window-based filtering [163,164].
SAR RS for estimation and mapping of forest structure parameters is well developed and includes
several previous literature reviews [165–168]. However, literature on assessment and monitoring
of FH related attributes is more limited for applications such as changes in forest structure, leaf
quantity, distribution or size, or canopy moisture content from deforestation, canopy degradation
(i.e., reduced biomass or LAI) or mortality caused by insect infestation, disease, drought, flooding, or
other environmental factors. Studies in Table 7 are drawn from both groups of literature to provide
illustrative examples of the major FH applications applicable to SAR: deforestation, degradation, fire,
and inundation.
Detection of deforestation has been conducted with backscatter data, often using classification
techniques; examples from earlier studies to recent more advanced studies are given in Table 7.
In contrast to classification, biophysical modelling of forest structure parameters is a major application
that has seen intensive research. Many studies have been conducted in tropical areas where persistent
cloud cover requires RADAR data, or in northern boreal areas where low sun angle effects can
reduce the quality of optical model estimates. The most common parameter modelled is above
ground biomass (AGB). In a similar manner to visible-near infrared reflectance-based indices, the SAR
backscatter-AGB relationship is curvilinear and many studies have found that it typically saturates
in the range of 100–150 t/ha (from the reviews of [165,167,169]), whereas others have found lower
thresholds of saturation [170,171]. Thus, biomass and biomass loss are difficult to model and estimate
using backscatter or backscatter-based indices in dense forests. Studies in forests below the saturation
threshold have been successful, however. Examples in Table 7 include explicit estimation of AGB
loss as well as studies that only estimated biomass for a single date (i.e., where the context was not
forest degradation), but that have potential for temporal or spatial application to estimate biomass
loss. Other examples are given for leaf area index (LAI), canopy height and forest density that have
potential for canopy degradation mapping. With the BIOMASS mission of ESA planning a polarimetric
and interferometric P-band satellite sensor for 2020 [172], the combined use of these three SAR features
(polarimetry, InSAR, long wavelength) will offer better means for temporal monitoring of forest
structural condition and degradation.
The other two FH applications of RADAR presented in Table 7 are fire impacts and forest
inundation. Both have been addressed using backscatter and polarimetric analysis, mostly in
classification to detect areas impacted by either disturbance type.
It is clear that forest removal, disturbance and degradation analysis and monitoring using RADAR
depends on structural forest change that is manifested as a consequence of a given source of stress or
degradation. A direct analysis of backscatter response in specific polarizations and wavelengths, or
combinations of these, has been the most common approach and has shown some success. However,
more recently, improved potential has been shown using multi temporal data, InSAR, polarimetric and
decomposition techniques to detect and estimate changes in signal polarization, phase, and volume,
surface and double bounce scattering contributions that are related to changes in canopy structure.
It is in these areas that FH change research using RADAR data needs continued efforts.
Remote Sens. 2017, 9, 129 14 of 33
Table 7. Example studies illustrating the use of RADAR in common applications: deforestation,
degradation, fire, and inundation.
Application Example Studies Main Findings
Deforestation
[173] Shuttle Imaging Radar (SIR) L-band better than C-band. Multiplepolarizations required to detect intermediate Amazon deforestation classes.
[174]
Airborne L- and P-bands better than X- and C- bands in classifying selective
logging classes. Image texture also beneficial. Identified a need for
multi-temporal data to discriminate disturbed/logged areas from temporally
stable forest classes.
[175]
PALSAR L-band object-based classification of forest and deforestation classes
in Indonesia for 4 successive years. Markov chain analysis applied in future
scenarios modelling of potential deforestation areas to aid in reducing rates
of deforestation.
Forest degradation:
Estimation of forest
structure parameters
[176]
Above ground biomass (AGB) (avg. 78 t/ha) for 3 classes (non-forest, shrub
and forest) in disturbed forests of Laos using PALSAR L-band. High RMSE
(42%–52%) but similar to a model using AVNIR optical data, the latter being
more difficult to obtain due to the persistent cloud cover.
[171,177–179]
Improved AGB modelling with higher saturation threshold using:
(1) Cross-polarized (HV) data; (2) P-band that penetrates deeper into the
canopy; (3) Shorter-to-longer wavelength backscatter ratio (e.g., C/L);
(4) Averages of multi-temporal images to reduce moisture/ rain effects.
[180,181]
AGB loss estimation: (1) Relation between 2007 PALSAR HV L-band
backscatter and AGB in a forest-savannah of Africa applied to 1996 JERS-1
HH L-band data to estimate biomass and detect areas of biomass loss.
Concluded that better consistency and calibration between data types is
needed to conduct such temporal analysis; (2) Multi-temporal canopy height
derived from SRTM C-band InSAR and IceSAT GLAS LiDAR used in
allometric equations to estimate biomass and biomass loss in mangrove forest
due to hypersaline soils caused by anthropogenic hydrological changes.
[182]
Leaf Area Index (LAI): VV/HH ratio (and additional VV terms for dry forests
with stronger trunk response) derived from Envisat ASAR C-band images to
estimate LAI in boreal and subarctic forests.
[183]
Forest density: Six incoherent decomposition techniques compared using
RADARSAT-2 data in forests in India. Support Vector Machine (SVM)
classification of the decomposition parameters produced 70%–90% accuracy
for 3 forest density classes.
Fire impacts
[184] Greater backscatter in burned areas.
[185]
Burn severity and surface roughness in an Alaska forest were the strongest
factors affecting ERS-1 C-band VV backscatter. Relationship was strongest
when surface moisture variations were minimal.
[186]
Ratio of pre- to post-fire backscatter in a given polarization related to a
temperate forest burn intensity index. Dry period cross-polarized data
performed best.
[187]
Individual PALSAR polarimetric, phase, and decomposition parameters
distinguished Amazon burn classes if area burned 3 or more times. More
subtle burn discrimination achieved with multiple variable models,
particularly those with phase and power metrics.
Inundation
[173]
Flooded forest class mapped based on dead trees producing high double
bounce scattering and a large (120◦) phase difference between HH- and
VV-polarized returns at both C- and L-band.
[188]
Effects of incident angle on RADARSAT-1 C-band backscatter evaluated in
flooded forests of North Carolina. Moderate angles detected inundation best
during both leaf-off and leaf-on periods.
[189]
InSAR data generated from PALSAR L-band and RADARSAT-1 C-band data
to determine flooding levels in Louisiana swamp forests. HH best single
polarization. Swamp forest had a relatively high HH/HV ratio (0.4–1.0)
indicating significant double-bounce backscatter that helped distinguish it
from upland forest.
Remote Sens. 2017, 9, 129 15 of 33
3.3. Multi-Sensor Approaches
Various studies have shown that the implementation of multi-sensor RS approaches
improves the discrimination of ST/STV over time and thus the accuracy of estimation of FH
indicators [14,134–137,190]. Depending on their sensor characteristics (spatial, radiometric, spectral,
temporal or angular resolution), RS sensors can record specific ST/STV and thus discriminate between
certain species, populations, communities, habitats and biomes of FES [16,136]. Reviews, advantages
and limitations using multi- and hyperspectral RS data for monitoring FH and FES diversity have
been conducted [16,136]. Therefore, in this section, we focus on innovative present and future optical
and multi-sensor approaches to monitor FH.
By implementing (i) Optical 3D–RS approaches, as well as (ii) the combination and/or merging
of multi-RS sensors (optical, thermal, RADAR, LiDAR), additional 2/3D ST and STVs can be
recorded, which considerably improve the discrimination, monitoring and understanding of FH [139].
Hyperspectral sensors are increasingly being implemented for 3D imaging spectroscopy to create
hyperspectral 3D plant models [191,192]. Hyperspectral RS datasets can be combined or merged with
3D point clouds from LiDAR [192], enabling the assessment of tree and canopy spectral traits in three
dimensions. Furthermore, they allow to study the effects of plant geometry and sensor configurations,
which is key for physically-based reflectance models [192]. 3D thermal tree and canopy models can
contribute to the understanding of temperature distribution [193,194]. Thermal infrared (TIR) and 3D
simulations have been used to quantify and assess the spectral water stress traits of vegetation [195,196].
Multi-sensor RS techniques such as the Airborne Oblique System with 4 thermal cameras and 4 RGB
cameras have been developed with the purpose of recording 3D-temperature distributions in trees.
The latest development in UAV technologies also integrates several sensors to quantify 3D architecture
and spectral traits.
Use of multiple platforms for a given sensor type (e.g., the forthcoming Radarsat Constellation of
three platforms [197]) provides potential for more frequent data acquisition and a more dense temporal
dataset , particularly with optical sensors when cloud cover is a hindrance. For example, combining
data from sensors with similar radiometric and spatial characteristics such as Landsat 7 and 8, ASTER
or Sentinel 2, each of which has repetition periods of multiple days or weeks, can provide a dense data
set for analysis of high frequency or steep temporal gradients in FH traits. However, between-sensor
calibration is required in order to match datasets. Alternatively, multi-sensor systems on single
platforms are promising, as they: (a) enable the simultaneous acquisition of information related to
different ST; and (b) ensure the same illumination conditions, weather conditions and flight parameters
for all mounted sensors. Multi-sensor approaches on airborne platforms can include hyperspectral,
multispectral, TIR, RGB, LiDAR, active RADAR and passive microwave sensors to quantify spectral
trait indicators of FH [198]. For example, the Goddard LiDAR hyperspectral and thermal (G-LiHT)
airborne imaging system integrates a combination of lightweight and portable multi-sensors for
measuring vegetation structure, foliar spectra and surface temperatures at a high spatial resolution
(1 m) to map plant species composition, plant functional types, biodiversity, biomass as well as plant
growth [16,136].
Space-borne RS of the future will increasingly focus on the developments of multi-sensor RS
platforms. The HyspIRI mission (HyspIRI, [199], launch in 2020) includes two instruments, namely
an imaging spectrometer (380 nm–2500 nm, 214 spectral channels) and an 8-band multispectral
imager, including one band in the mid-wave-infrared region (MIR) at 4 µm and seven bands in the
long-wave infrared (LWIR) region between 7 and 13 µm [200]. Based on this sensor combination,
HyspIRI is expected to be useful in quantification of forest canopy biochemistry and components in the
biogeochemical cycles of FES [21,201–203], the health status of forests based on biochemical-biophysical
ST, or the pattern and spatial distribution and diversity of forest plant species and communities [204].
Furthermore, it will be able to discriminate and monitor forest species [204,205], forest plant
functional types, [206,207] or invasive species [21] as well as natural disasters and disturbance
regimes, i.e., volcano eruptions, wildfires, beetle infestations [208,209], and the global carbon cycles.
Remote Sens. 2017, 9, 129 16 of 33
The Hyperspectral Imager Suite (HISUI) is a coming space-borne multi-sensor platform combining
a hyperspectral imager (185 spectral channels) and a 5-band multispectral imager, which will be
launched on the Advanced Land Observation Satellite 3 (ALOS-3) [200]. Table 8 lists example findings
using multi-sensor approaches in common FH applications.
Table 8. Example studies illustrating the use of the multi-sensor approach in common FH applications.
Application Example Studies Main Findings
Estimation of ST/STV
of FH
[14,16,138,139]
Implementation of multi-sensor RS in modelling approaches
improves the discrimination, quantification and accuracy
when estimating ST/STV.
[197]
Multiple platforms for a given sensor type, e.g., the
forthcoming Radarsat Constellation of three platforms
provides potential for more frequent data acquisition.
[198]
Multi-sensor systems on single platforms are promising, as
they: (1) Enable data related to different ST to be recorded at
the same time; (2) Ensure the same illumination conditions,
weather conditions and flight parameters.
Forest diversity indicators
(taxonomic, structural and
functional diversity)
[210,211]
Sensor characteristics (spatial, radiometric, spectral, temporal
or angular or directional resolution) determine the
discrimination and classification capabilities of forest
diversity indicators.
3D structural ST/STV of
trees and canopies
[191,192] 3D-imaging spectroscopy is crucial to create hyperspectral 3Dplant models.
[192] Multi-hyperspectral RS datasets combined or merged with 3Dpoint clouds from LiDAR.
Canopy temperature
distribution [193,194]
3D thermal tree models and 3D thermal canopy models are
available for a better understanding of tree and canopy
temperature distribution.
FH indicators on the plot
and field scale [212–217]
UAVs can carry different sensor types and thus contribute to a
more comprehensive, rapid, cost-effective, comparable and
repetitive recording of FH indicators at very high resolution.
LiDAR, thermal infrared, multispectral, RGB, hyperspectral,
real time video.
Fusion of multi-sensor
RS Data
[218,219]
RS data can be merged to integrate the advantages of high
spatial, spectral and temporal resolution within one kind of
sensor. Pan-sharpening processes to combine a spatially
higher resolution pan-chromatic band with synchronously
recorded lower resolution multi-spectral bands.
[219]
Fusion of a large number of spectral bands with high spatial
resolution data can be achieved using multi-resolution
methods (e.g., Wavelets) and high-frequency injection
methods.
[220,221]
Spatial and Temporal Adaptive Reflectance Fusion Model
(STARFM,) that combines the spatial accuracy of Landsat data
with the temporal resolution of MODIS RS data.
[219] Data fusions approaches that merge spatial, spectral as well astemporally high resolution RS sensors into one data set.
4. Physical vs. Empirical Models
The estimation of ST and trait variations of forest stands from RS data generally relies on two
different approaches: (1) empirical modelling and (2) physical modelling. A third group of approaches
usually combines the two methodologies. Empirical models build on a statistical relationship between
one or more biotic traits (dependent variable) and one or more spectral (independent) variables
that are known to physically respond to the given biotic trait. In most cases, spectral reflectance in
the visible, near infrared, or vegetation indices, e.g., the Normalized Difference Vegetation Index
Remote Sens. 2017, 9, 129 17 of 33
(NDVI), serve as input for such models. The spectral information can further be complemented
by spatial information, e.g., image texture [222–224], semivariogram [225] and radiometric fraction
analysis [196]. The statistical relation is interpreted and applied by means of linear or non-linear models
created through regression and related techniques or through classification. Table 9 summarizes the
types of models that are frequently used for such purposes, ranging from different adaptations
of linear regression (e.g., ordinary least squares regression, OLS; reduced major axis regression,
RMA), non-linear regression (e.g., exponential, logarithmic), multivariate techniques such as canonical
correlation analysis or redundancy analysis to non-parametric regression models. In classification,
parametric techniques such as the Maximum Likelihood classification were traditionally used but more
recently, non-parametric classifiers such as k-Nearest Neighbours (kNN), Classification and Regression
Trees (CART), and the related ensemble classifier, Random Forests (RF), and Support Vector Machine
(SVM) algorithms have been widely adopted. In particular, RF [226] has gained popularity because it:
(i) can take input parameters that are continuous, interval or class-based; (ii) can process large numbers
of input variables with complex interactions, although variable correlations and spatial autocorrelation
have been shown to affect output accuracy [227]; (iii) incorporates bootstrapping for training and
validation (out-of-bag) sample selection from a set of reference data in each of many iterations and
their associated assessment of internal (out-of-bag) error; and (iv) can be used as a variable selection
tool to identify the most important variables that discriminate a set of classes. Although little has been
reported on RF use in FH applications, there is significant potential for its use in classification of FH
classes, or in RF regression to model and estimate FH traits.
The aim of empirical models in RS is to use relationships between image and field data for the
spatially explicit estimation of biophysical attributes. Thus, empirical models are most frequently
used in regional case studies. They provide the most straightforward way of estimating biotic traits
from RS. On the other hand, they are limited in transferability because the nature of the statistical
relation between biotic traits and RS indices changes with the modification of the underlying RS
data, calibration processes, date of acquisition and phenological state of the vegetation. Moreover,
any empirical model is considerably influenced by other external conditions such as the atmosphere,
sun angle, geometric resolution, and the accuracy of both the RS and field data. It is also evident
that the relation varies between different species (even within a forest stand) and that it is sensitive
to the canopy’s horizontal and vertical structure. In addition, many studies have shown that some
empirical relations reach a level of saturation that is often caused by the nature of the underlying
vegetation indices, especially in stands with a high biomass [224,228]. There is only slight agreement
about which algorithm, and which set of independent variables is best suited for the prediction of
forest biophysical attributes and their changes over time that can be related to FH conditions. In spite
of these limitations, empirical models are the foundation for an era of civilian earth RS studies based
on medium resolution satellite images like Landsat, SPOT and others. They are the appropriate choice
if no detailed knowledge about the geometric and optical properties of the vegetation exist, and the
information about the radiative path from the object to the sensor is incomplete, or if the direct physical
relationship between the biotic trait (e.g., biomass) and the reflected spectral energy is not known [224].
Canopy reflectance (CR) models, as part of the more generalized approach of radiative transfer
(RT) modelling build on the comprehensive description of the nature of interactions between incoming
radiation, vegetation parameters and environmental factors [229]. They assume a logical and direct
connection between biotic traits (e.g., LAI), the geometric properties of a canopy or leaf and the
directional reflected energy. Typically, these models are designed to predict the directional spectral
reflectance of a canopy based on the known spectral behaviour of the geometric-optical and biophysical
properties or parameters of the vegetation and the environment. Hence, for RS applications, CR models
must be inverted to estimate one or more canopy parameters as a function of the canopy reflectance.
All approaches to modelling the radiation in a canopy are based on the geometrical optics, the
radiative transfer or the canopy transmittance theory. They can be summarized by four categories after
Goel [229]. (1) Geometrical models; (2) Turbid medium models; (3) Hybrid models; and (4) Computer
Remote Sens. 2017, 9, 129 18 of 33
simulation models. Geometrical models consider the canopy to be a compilation of geometrical objects
with known shapes and dimensions that have distinct optical properties in terms of transmission,
absorption and reflectance [230,231]. By contrast, turbid medium models regard the canopy as a
collection of absorbing and scattering particles, with given optical properties that are distributed
randomly in horizontal layers [229]. Hybrid models combine the first two approaches and account for
geometrical-optical properties as well as the absorption and scattering of plant elements (e.g., leaves
or trunks). Finally, computer-simulation models completely model the arrangement and orientation
of plant elements and the interference of spectral radiative energy by one of these elements based on
computer vision. The advantage of computer models is that they enable a systematic configuration
and analysis of radiation regimes and canopy structure (e.g., size, orientation, position of leaves).
The application of a certain model is mostly determined by the availability of parameters. With regard
to saturation, geometrical models are more appropriate for sparse canopies, whereas turbid medium
models more adequately represent dense and homogenous forest stands. The main limitation of CR
models is the complexity of the model due to the complex physical nature of the relations between
spectral reflectance and vegetation parameters. The inversion of CR models to target variables requires
knowledge of all auxiliary model parameters.
The choice between physical and empirical models depends on the target application. Physical
models are widely used for the quantitative modelling of biophysical vegetation parameters on a
global scale (e.g., MODIS LAI, see [232]; Landsat LAI, see [233]). They rely on the physical description
of the energy-matter interaction. However, every physical model strongly depends on empirical data
for calibration and validation and also faces problems of saturation in regions with high biomass
(see, e.g., [234] for MODIS GPP modelling). Semi-empirical approaches combine both empirical and
physical modelling, e.g., by using the output from CR models to train neural networks to estimate
biophysical parameters [235].
Table 9. Synthesis of modelling approaches for the RS based assessment of ST for FH assessment.
Input RS Information ModellingApproach Model Type Algorithm
(Target Variable)
[Reference Application]
Spectral indices (e.g., SR,
NDVI, tasseled cap), Raw DN,
Spectral reflectance, Principal
components, PAR, APAR,
Multi-angular reflectance,
Image spatial or temporal
metrics, LiDAR waveform
metrics, SAR amplitude, SAR
coherence, SAR polarimetry
Empirical
Linear and non-linear
regression
Linear regression (AGB, carbon), [239];(AGB), [240,241]
Ordinary least
squares (height, density, DBH), [242]
Reduced major axis (AGB), [243]; (LAI), [244]
Canonical
Correlation
Analysis
(forest structural
conditions), [222]
Redundancy
Analysis
(forest structural
conditions), [245,246]
Trend analysis (growth), [247]
Non-parametric
regression
kNN (AGB, carbon), [248]
CART (tree cover), [249]; (basalarea, no. of trees) [250]
RF (AGB) [243,251]
SVM (height, density, DBH), [242]
Physical
Radiative
transfer/canopy
reflectance model
Geometric-Optical (LAI), [252]; (AGB), [253];(Chlorophyll), [254]
Turbid-medium (LAI), [255]
hybrid (allometry), [256]
Computer
simulation
Coherence-Amplitude
conversion
RVoG (height), [236,238]
WCM (AGB), [237]
Hybrid Neural Networks (LAI), [235]
Remote Sens. 2017, 9, 129 19 of 33
Similar modelling approaches are available based on SAR remote-sensing data. SAR-based
empirical modelling approaches are conceptually the same as for optical data and make use of the
backscatter, coherence or polarization information of the processed microwave signal as explained in
Section 3.2. The physically-based models rely on the coherence-amplitude conversion of the complex
SAR signal at different wavelengths and polarizations. In contrast to optical RS, SAR-based physical
models are more related to the vertical structure of vegetation and, hence, provide a more direct
approach for modelling parameters of the global carbon budget, e.g., AGB. The most widely used
microwave scattering models for forest stands are the Random Volume over Ground (RVoG, [236] and
the Water Cloud Model (WCM, [237]) and a number of adaptations [238].
5. Conclusions
Even though there is an increasing number of data sets available to the public from in-situ forest
inventories, experimental studies, and RS, there are only a handful of programs in operation that
integrate these sources into FHM programs. This paper reviewed the approaches of how close-range,
air and space-borne RS techniques can be used in combination with different platforms, sensor types,
and modelling approaches to assess indicators of FH.
Both RS as well as in-situ approaches have advantages and limitations and should therefore be
combined to provide a holistic view of forest conditions. In-situ forest monitoring approaches record
indicators of FH that are species-specific and in some cases use additional environmental ecosystem
factors that describe climate and soil characteristics. This kind of monitoring is accurate but limited
to sample points. Furthermore, in-situ forest inventory monitoring does not record indicators of FH
according to the ST/STV approach, making it difficult to link in-situ observations and RS approaches.
Here, new avenues for linking these approaches need to be explored and applied.
Compared to RS approaches, through long-term monitoring, field-based monitoring approaches
can draw conclusions about various processes and stress scenarios in FES. However, assessment,
monitoring and the capability to draw conclusions about short-term processes and stress scenarios are
often limited.
Why can RS measure ST/STV as indicators of stress, disturbance or resource limitations of FES
and therefore serve as a valuable means to assess FH?
• Forest plant traits are anatomical, morphological, biochemical, physiological, structural or
phenological features and characteristics that are influenced by taxonomic and phylogenetic
characteristics of forest species ([257,258].
• Stress, disturbances or resource limitations in FES can manifest in the molecular, genetic,
epigenetic, biochemical, biophysical or morphological-structural changes of traits and affect
trait variations [17,259–261] which can lead to irreversible changes in taxonomic, structural and
functional diversity in FES.
• RS is a physically-based system that can only directly or indirectly record the Spectral Traits (ST)
and spectral trait variations (STV)” [16] in FES [136,262].
Airborne- and space-borne RS approaches can monitor indicators of various processes and stress
scenarios taking place in FES via the ST/STV approach in an extensive, continuous, and repetitive
manner. Furthermore, insights on both short-term and long-term processes and stress scenarios in FES
can be gained. Hence, further knowledge is required about the phylogenetic reactions and adaptive
mechanisms of forest species on various stress factors.
This paper expands on existing close-range RS measurements to observe FH indicators and
illustrates their advantages and disadvantages. In a similar way to air-and space-borne RS approaches,
close-range RS measurements also record indicators of FH based on the ST/STV approach. By linking
in-situ, close-range and air- and space-borne RS approaches, key relations can be observed between
the genetic and phylogenetic characteristics and species-specific reactions to stress and their direct
relationship with the spectral response recorded by RS sensors. Moreover, close-range RS approaches
Remote Sens. 2017, 9, 129 20 of 33
offer crucial calibration and validation information for air-and space-borne RS data with a high
temporal resolution.
Furthermore, there is a lack of a standardized framework to link various information sources for
recording and assessing indicators of FH. The concept of ST and STV presented in [16,136] provides a
framework to develop a standardized monitoring approach for FH indicators. Ultimately, different
requirements will be needed to link the different approaches, sensors and platforms of RS with in-situ
forest inventories and experimental studies to better describe, explain, predict and understand FH.
Acknowledgments: We particularly thank the researchers for the Hyperspectral Equipment of the Helmholtz
Centre for Environmental Research—UFZ and TERENO funded by the Helmholtz Association and the Federal
Ministry of Education and Research. The authors also thank the reviewers for their very valuable comments
and recommendations.
Author Contributions: A.L., M.H., P.M., S.E. and D.J.K. were responsible for all parts of this review analysis,
writing and production of the figures. M.H. and P.M. provided extensive contributions on recording terrestrial
FH as well as substantial input for the LiDAR part and about methods used for implementing sensors in the
context of FH. D.J.K. wrote the RADAR section and S.E. added his knowledge to physical vs. empirical models
for understanding FH with remote-sensing techniques. All authors checked and contributed to the final text.
Conflicts of Interest: The authors declare no conflicts of interest.
Abbreviations
The following abbreviations are used in this manuscript:
AGB Above Ground Biomass
ALOS-3 Advanced Land Observation Satellite 3
AVHRR Advanced Very High Resolution Radiometer
BRDF Bidirectional Reflectance Distribution Function
CART Classification and Regression Trees
CR Canopy Reflectance
DBH Diameter at Breast Height
DSM Digital Surface Model
DTM Digital Terrain Model
EnMAP Environmental Mapping and Analysis Program
ESA European Space Agency
FAO Food and Agriculture Organization of the United Nations
FES Forest Ecosystems
FH Forest Health
FHM Forest Health Monitoring
FLEX Fluorescence Explorer
FRA Global Forest Resources Assessment
GCEF Global Change Experimental Facility
GEDI Global Ecosystem Dynamics Investigations
GLAS Geoscience Laser Altimeter System
GLCM Gray-Level Co-Occurence Matrix
GPP Gross Primary Productivity
GPS Global Positioning System
HISUI Hyperpsectral Imager Suite
HySPIRI Hyperspectral Infrared Imager
ICESat Ice, Cloud and Land Elevation Satellite
ICOS Integrated Carbon Observation System
ICP International Co-operative Programme on Assessment and Monitoring of Air Pollution on Forests
INS Internal Navigation System
InSAR Interferometric Synthetic Aperture RADAR
IPPN International Plant Phenotyping Network
JERS Japanese Earth Resources Satellite
Knn K-Nearst Neighbour
LAI Leaf Area Index
LiDAR Light Detection and Ranging
LOD Linked Open Data
Remote Sens. 2017, 9, 129 21 of 33
LWIR Long-wave infrared
MIR Mid-wave-infrared
MODIS Moderate Resolution Imaging Spectroradiometer
NASA National Aeronautics and Space Administration
NDVI Normalized Difference Vegetation Index
NEON National Ecological Observatory Network
NOAA National Oceanic and Atmospheric Administration
PALSAR Phased Array type L-band Synthetic Aperture RADAR
PRI Photochemical Reflectance Index
RADAR Radio Detection And Ranging
RF Random Forests
RMA Reduced Major Axis Regression
RMSE Root Mean Square Error
RS Remote Sensing
RT Radiative Transfer
RVoG Random Volume over Ground
SAR Synthetic Aperture RADAR
SIR Shuttle Imaging RADAR
ST Spectral Traits
STV Spectral Trait Variation
SVM Support Vector Machine
TIR Thermal Infrared, Thermal Infrared
TRGM Thermal Radiosity Graphics Model
UAV Unmanned Aerial Vehicle
UNECE United Nations Economic Commission for Europe
USDA United States Department of Agriculture
WCM Water Cloud Model
WSN Wireless sensor networks
References
1. Roy, B.A.; Alexander, H.M.; Davidson, J.; Campbell, F.T.; Burdon, J.J.; Sniezko, R.; Brasier, C. Increasing forest
loss worldwide from invasive pests requires new trade regulations. Front. Ecol. Environ. 2014, 12, 457–465.
[CrossRef]
2. Food and Agriculture Organization of the United Nations. State of Europe’s forests 2015. In Proceedings of
the Ministerial Conference on the Protection of Forests in Europe, Madrid, Spain, 20–21 October 2015.
3. Potter, K.M.; Conkling, B.L. Forest Health Monitoring: National Status, Trends, and Analysis; U.S. Department of
Agriculture: Asheville, NC, USA, 2015.
4. Yang, J.; Dai, G.; Wang, S. China’s National Monitoring Program on Ecological Functions of Forests:
An Analysis of the Protocol and Initial Results. Forests 2015, 6, 809–826. [CrossRef]
5. Tomppo, E.; Gschwantner, T.; Lawrence, M.; McRoberts, E.R. National Forest Inventories. Pathways for Common
Reporting; Springer: Heidelberg, Germany; Dordrecht, The Netherlands; London, UK; New York, NY,
USA, 2010.
6. Johann Heinrich von Thünen-Institute, Forest Condition Monitoring (FCM) Level-I-Monitoring. Available
online: https://www.thuenen.de/de/wo/arbeitsbereiche/waldmonitoring/ (accessed on 3 January 2017).
7. Federal Ministry for Food and Agriculture, National Forest Inventory Level-III-Monitoring. Available online:
https://bwi.info/ (accessed on 3 January 2017).
8. United States Department of Agriculture(USDA), Forest Service, FH Monitoring. Available online: https:
//www.fs.fed.us/foresthealth/monitoring/index.shtml (accessed on 3 January 2017).
9. Canadian Forest Service (CFS), National FHM Network. Available online: http://www.cfs.nrcan.gc.ca/
publications/?id=4105 (accessed on 3 January 2017).
10. National Forest Inventory (NFI), Canada, National Forest Inventory. Available online: https://nfi.nfis.org/
en/ (accessed on 3 January 2017).
11. United Nations Economic Commission for Europe (UNECE), ICP. Available online: http://icp-forests.net/
page/icp-forests-executive-report (accessed on 3 January 2017).
Remote Sens. 2017, 9, 129 22 of 33
12. Food and Agriculture Organization of the United Nations (FAO), Forest Resources Assessment. Available
online: http://www.fao.org/forest-resources-assessment/en/ (accessed on 3 January 2017).
13. Woodall, C.W.; Amacher, M.C.; Bechtold, W.A.; Coulston, J.W.; Jovan, S.; Perry, C.H.; Randolph, K.C.;
Schulz, B.K.; Smith, G.C.; Tkacz, B.; et al. Status and future of the forest health indicators program of the
USA. Environ. Monit. Assess. 2011, 177, 419–436. [CrossRef] [PubMed]
14. Pause, M.; Schweitzer, C.; Rosenthal, M.; Keuck, V.; Bumberger, J.; Dietrich, P.; Heurich, M.; Jung, A.;
Lausch, A. In Situ/Remote Sensing Integration to Assess Forest Health—A Review. Remote Sens. 2016, 8, 471.
[CrossRef]
15. Trumbore, S.; Brando, P.; Hartmann, H. Forest health and global change. Science 2015, 349, 814–818.
[CrossRef] [PubMed]
16. Lausch, A.; Erasmi, S.; King, D.J.; Magdon, P.; Heurich, M. Understanding forest health by remote sensing
Part I—A review of spectral traits, processes and remote sensing characteristics. Remote Sens. 2016, 8.
[CrossRef]
17. Garnier, E.; Lavorel, S.; Ansquer, P.; Castro, H.; Cruz, P.; Dolezal, J.; Eriksson, O.; Fortunel, C.; Freitas, H.;
Golodets, C.; et al. Assessing the effects of land-use change on plant traits, communities and ecosystem
functioning in grasslands: A standardized methodology and lessons from an application to 11 European
sites. Ann. Bot. 2007, 99, 967–985. [CrossRef] [PubMed]
18. Brosinsky, A.; Lausch, A.; Doktor, D.; Salbach, C.; Merbach, I.; Gwillym-Margianto, S.; Pause, M. Analysis of
Spectral Vegetation Signal Characteristics as a Function of Soil Moisture Conditions Using Hyperspectral
Remote Sensing. J. Indian Soc. Remote Sens. 2013, 42, 311–324. [CrossRef]
19. Wollschla’ger, U.; Attinger, S.; Borchardt, D.; Brauns, M.; Cuntz, M.; Dietrich, P.; Fleckenstein, J.H.;
Friese, K.; Friesen, J.; Harpke, A.; et al. The Bode Hydrological Observatory: A platform for
integrated, interdisciplinary hydro- ecological research within the TERENO Harz/Central German Lowland
Observatorytle. Environ. Earth Sci. 2010, 76, 29. [CrossRef]
20. Clasen, A.; Somers, B.; Pipkins, K.; Tits, L.; Segl, K.; Brell, M.; Kleinschmit, B.; Spengler, D.; Lausch, A.;
Förster, M. Spectral unmixing of forest crown components at close range, airborne and simulated Sentinel-2
and EnMAP spectral imaging scale. Remote Sens. 2015, 7, 15361–15387. [CrossRef]
21. Asner, G.P.; Martin, R.E. Airborne spectranomics: Mapping canopy chemical and taxonomic diversity in
tropical forests. Front. Ecol. Environ. 2009, 7, 269–276. [CrossRef]
22. Asner, G.P.; Anderson, C.B.; Martin, R.E.; Tupayachi, R.; Knapp, D.E.; Sinca, F. Landscape biogeochemistry
reflected in shifting distributions of chemical traits in the Amazon forest canopy. Nat. Geosci. 2015, 8, 567–573.
[CrossRef]
23. Asner, G.P.; Martin, R.E. Spectral and chemical analysis of tropical forests: Scaling from leaf to canopy levels.
Remote Sens. Environ. 2008, 112, 3958–3970. [CrossRef]
24. Asner, G.P.; Martin, R.E. Canopy phylogenetic, chemical and spectral assembly in a lowland Amazonian
forest. New Phytol. 2011, 189, 999–1012. [CrossRef] [PubMed]
25. Nink, S.; Hill, J.; Buddenbaum, H.; Stoffels, J.; Sachtleber, T.; Langshausen, J. Assessing the suitability of
future multi- and hyperspectral satellite systems for mapping the spatial distribution of norway spruce
timber volume. Remote Sens. 2015, 7, 12009–12040. [CrossRef]
26. Buddenbaum, H.; Rock, G.; Hill, J.; Werner, W. European Journal of Remote Sensing Measuring Stress
Reactions of Beech Seedlings with PRI, Fluorescence, Temperatures and Emissivity from VNIR and Thermal
Field Imaging Spectroscopy. Eur. J. Remote Sens. Eur. J. Remote Sens. 2015, 48, 263–282. [CrossRef]
27. Buddenbaum, H.; Stern, O.; Paschmionka, B.; Hass, E.; Gattung, T.; Stoffels, J.; Hill, J.; Werner, W. Using VNIR
and SWIR field imaging spectroscopy for drought stress monitoring of beech seedlings. Int. J. Remote Sens.
2015, 36, 4590–4605. [CrossRef]
28. Lausch, A.; Pause, M.; Schmidt, A.; Salbach, C.; Gwillym-Margianto, S.; Merbach, I. Temporal hyperspectral
monitoring of chlorophyll, LAI, and water content of barley during a growing season. Can. J. Remote Sens.
2013, 39, 191–207. [CrossRef]
29. Doktor, D.; Lausch, A.; Spengler, D.; Thurner, M. Extraction of plant physiological status from hyperspectral
signatures using machine learning methods. Remote Sens. 2014, 6, 12247–12274. [CrossRef]
30. Furbank, R.T. Plant phenomics: From gene to form and function. Funct. Plant Biol. 2009, 36, 5–6.
31. Ehrhardt, D.W.; Frommer, W.B.; Ehrhardt, D.W. New technologies for 21st century plant science. Plant Cell
2012, 24, 374–394. [CrossRef] [PubMed]
Remote Sens. 2017, 9, 129 23 of 33
32. Fiorani, F.; Schurr, U. Future scenarios for plant phenotyping. Annu. Rev. Plant Biol. 2013, 64, 267–291.
[CrossRef] [PubMed]
33. Großkinsky, D.K.; Svensgaard, J.; Christensen, S.R.T. Plant phenomics and the need for physiological
phenotyping across scales to narrow the genotype-to-phenotype knowledge gap. J. Exp. Bot. 2015, 66,
5429–5440. [CrossRef] [PubMed]
34. Großkinsky, D.K.; Pieruschka, R.; Svensgaard, J.; Rascher, U.; Christensen, S.; Schurr, U.; Roitsch, T.
Phenotyping in the fields: Dissecting the genetics of quantitative traits and digital farming. New Phytol. 2015,
207, 950–952. [CrossRef] [PubMed]
35. Pieruschka, R.; Lawson, T. Preface. J. Exp. Bot. 2015, 66, 5385–5387. [CrossRef] [PubMed]
36. Virlet, N.; Costes, E.; Martinez, S.; Kelner, J.J.; Regnard, J.L. Multispectral airborne imagery in the field
reveals genetic determinisms of morphological and transpiration traits of an apple tree hybrid population in
response to water deficit. J. Exp. Bot. 2015, 66, 5453–5465. [CrossRef] [PubMed]
37. Li, L.; Zhang, Q.; Huang, D. A review of imaging techniques for plant phenotyping. Sensors (Switzerland)
2014, 14, 20078–20111. [CrossRef] [PubMed]
38. Lawton, J.H.; Naeem, S.; Woodfin, R.M.; Brown, V.K.; Gange, A.; Godfray, H.C.J.; Heads, P.A.; Lawler, S.;
Magda, D.; Thomas, C.D.; et al. The Ecotron: A controlled environmental facility for the investigation of
population and ecosystem processes. Philos. Trans. Biol. Sci. 1993, 341, 181–194. [CrossRef]
39. Eisenhauer, N.; Barnes, A.D.; Cesarz, S.; Craven, D.; Ferlian, O.; Gottschall, F.; Hines, J.; Sendek, A.; Siebert, J.;
Thakur, M.P.; et al. Biodiversity-ecosystem function experiments reveal the mechanisms underlying the
consequences of biodiversity change in real world ecosystems. J. Veg. Sci. 2016, 27, 1061–1070. [CrossRef]
40. Bradford, M.A.; Tordoff, G.M.; Eggers, T.; Jones, T.H.; Newington, J.E. Microbiota, fauna, and mesh size
interactions in litter decomposition. Oikos 2002, 99, 317–323. [CrossRef]
41. Kiirats, O.; Cruz, J.A.; Edwards, G.E.; Kramer, D.M. Feedback limitation of photosynthesis at high CO2 acts
by modulating the activity of the chloroplast ATP synthase. Funct. Plant Biol. 2009, 36, 893. [CrossRef]
42. Busch, F.; Hüner, N.P.A.; Ensminger, I. Biochemical constrains limit the potential of the photochemical
reflectance index as a predictor of effective quantum efficiency of photosynthesis during the winter spring
transition in Jack pine seedlings. Funct. Plant Biol. 2009, 36, 1016. [CrossRef]
43. Acˇ, A.; Malenovský, Z.; Hanuš, J.; Tomášková, I.; Urban, O.; Marek, M.V. Near-distance imaging
spectroscopy investigating chlorophyll fluorescence and photosynthetic activity of grassland in the daily
course. Funct. Plant Biol. 2009, 36, 1006–1015. [CrossRef]
44. Siebke, K.; Ball, M.C. Non-destructive measurement of chlorophyll b: A ratios and identification of
photosynthetic pathways in grasses by reflectance spectroscopy. Funct. Plant Biol. 2009, 36, 857–866.
[CrossRef]
45. Rascher, U.; Alonso, L.; Burkart, A.; Cilia, C.; Cogliati, S.; Colombo, R.; Damm, A.; Drusch, M.; Guanter, L.;
Hanus, J.; et al. Sun-induced fluorescence—A new probe of photosynthesis: First maps from the imaging
spectrometer HyPlant. Glob. Chang. Biol. 2015, 21, 4673–4684. [CrossRef] [PubMed]
46. Rascher, U. FLEX—Fluorescence Explorer: A remote sensing approach to quatify spatio-temporal variations
of photosynthetic efficiency from space. Photosynth. Res. 2007, 91, 1387–1390.
47. Jansen, M.; Gilmer, F.; Biskup, B.; Nagel, K.A.; Rascher, U.; Fischbach, A.; Briem, S.; Dreissen, G.; Tittmann, S.;
Braun, S.; et al. Simultaneous phenotyping of leaf growth and chlorophyll fluorescence via Growscreen
Fluoro allows detection of stress tolerance in Arabidopsis thaliana and other rosette plants. Funct. Plant Biol.
2009, 36, 902–914. [CrossRef]
48. Konishi, A.; Eguchi, A.; Hosoi, F.; Omasa, K. 3D monitoring spatio–temporal effects of herbicide on a whole
plant using combined range and chlorophyll a fluorescence imaging. Funct. Plant Biol. 2009, 36, 874–879.
[CrossRef]
49. Schimel, D.; Pavlick, R.; Fisher, J.B.; Asner, G.P.; Saatchi, S.; Townsend, P.; Miller, C.; Frankenberg, C.;
Hibbard, K.; Cox, P. Observing terrestrial ecosystems and the carbon cycle from space. Glob. Chang. Biol.
2015, 21, 1762–1776. [CrossRef] [PubMed]
50. Baldocchi, D.; Falge, E.; Gu, L.H.; Olson, R.; Hollinger, D.; Running, S.; Anthoni, P.; Bernhofer, C.; Davis, K.;
Evans, R.; et al. FLUXNET: A New Tool to Study the Temporal and Spatial Variability of Ecosystem-Scale
Carbon Dioxide, Water Vapor, and Energy Flux Densities. Bull. Am. Meteorol. Soc. 2001, 82, 2415–2434.
[CrossRef]
Remote Sens. 2017, 9, 129 24 of 33
51. Brown, T.B.; Hultine, K.R.; Steltzer, H.; Denny, E.G.; Denslow, M.W.; Granados, J.; Henderson, S.; Moore, D.;
Nagai, S.; Sanclements, M.; et al. Using phenocams to monitor our changing earth: Toward a global
phenocam network. Front. Ecol. Environ. 2016, 14, 84–93. [CrossRef]
52. Hilker, T.; Coops, N.C.; Nesic, Z.; Wulder, M.A.; Black, A.T. Instrumentation and approach for unattended
year round tower based measurements of spectral reflectance. Comput. Electron. Agric. 2007, 56, 72–84.
[CrossRef]
53. Yu, L.; Wang, N.; Meng, X. Real-time forest fire detection with wireless sensor networks. In Proceedings
of the 2005 International Conference on Wireless Communications, Networking and Mobile Computing,
Shanghai, China, 21–23 September 2005.
54. Ruiz-Garcia, L.; Lunadei, L.; Barreiro, P.; Robla, J.I. A review of wireless sensor technologies and applications
in agriculture and food industry: State of the art and current trends. Sensors 2009, 9, 4728–4750. [CrossRef]
[PubMed]
55. Lloret, J.; Garcia, M.; Bri, D.; Sendra, S. A wireless sensor network deployment for rural and forest fire
detection and verification. Sensors 2009, 9, 8722–8747. [CrossRef] [PubMed]
56. Hwang, J.; Shin, C.; Yoe, H. Study on an agricultural environment monitoring server system using wireless
sensor networks. Sensors 2010, 10, 11189–11211. [CrossRef] [PubMed]
57. Mafuta, M.; Zennaro, M.; Bagula, A.; Ault, G.; Gombachika, H.; Chadza, T. Successful Deployment of a
Wireless Sensor Network for Precision Agriculture in Malawi. Int. J. Distrib. Sens. Netw. 2013, 2013, 1–13.
[CrossRef]
58. Oliveira, R.S.; Eller, C.B.; Burgess, S.; Barros, F.V.; Muller, C.; Bittencourt, P. Soil-plant-atmosphere interactions
in a tropical montane cloud forest. In Proceedings of the Soil-Plant-Atmosphere Interactions in a Tropical
Montane Cloud Forest, Emerging Issues in Tropical Ecohydrology, AGU CHAPMAN Conderence, Cuenca,
Ecuador, 5–9 June 2016.
59. Teodoro, G.S.; Eller, C.B.; Pereira, L.; Brum, M., Jr.; Mauro; Oliveira, R.S. Interplay between stomatal
regulation capacity, hydraulic traits and growth performance in three shrub species in a tropical montane
scrubland under contrasting water availability. In Proceedings of the Soil-Plant-Atmosphere Interactions in
a Tropical Montane Cloud Forest. Emerging Issues in Tropical Ecohydrology, AGU CHAPMAN Conderence,
Cuenca, Ecuador, 5–9 June 2016.
60. Mauro, B., Jr.; Oliveira, R.S.; Gutierrez, J.; Licata, J.; Pypker, T.G.; Asbjornsen, H. Effects of the 2015 El-Niño
extreme drought on the sapflow of trees in eastern Amazonia. In Proceedings of the Soil-Plant-Atmosphere
Interactions in a Tropical Montane Cloud Forest. Emerging Issues in Tropical Ecohydrology, AGU
CHAPMAN Conderence, Cuenca, Ecuador, 5–9 June 2016.
61. Mollenhauer, H.; Remmler, P.; Schuhmann, G.; Lausch, A.; Merbach, I.; Assing, M.M.; Olaf, D.;
Peter Bumberger, J. Adaptive Multichannel Radiation Sensors for Plant Parameter Monitoring.
In Proceedings of the Geophysical Research Abstracts, EGU (European Geosciences Union General
Assemply), EGU General Assembly 2016, Austria, Vienna, 17–22 April 2016.
62. Mollenhauer, H.; Schima, R.; Assing, M.; Mollenhauer, O.; Dietrich, P.; Bumberger, J. Development of
Innovative and Inexpensive Optical Sensors in Wireless Ad-hoc Sensor Networks for Environmental
Monitoring. In Proceedings of the 12th EGU General Assembly, Wien, Austria, 12–17 April 2015.
63. Müller, J. Forestry and water budget of the lowlands in northeast Germany—Consequences for the choice of
tree species and for forest management. J. Water L. Dev. 2009, 13A, 133–148. [CrossRef]
64. Beck, W.; Müller, J.; Eichhorn, J. Impact of heat and drought on tree and stand vitality—Dendroecological
methods and first results from level 2-plots in southern Germany. Schriftenr. Forstl. Fak Univ. Göttingen ud
Nordwestdtsch. Forstl. Versuchsanst 2007, 142, 120–128.
65. Riggs, G.A.; Running, S.W. Detection of canopy water stress in conifers using the Airborne Imaging
Spectrometer. Remote Sens. Environ. 1991, 35, 51–68. [CrossRef]
66. Eitel, J.U.H.; Vierling, L.A.; Litvak, M.E.; Long, D.S.; Schulthess, U.; Ager, A.A.; Krofcheck, D.J.; Stoscheck, L.
Broadband, red-edge information from satellites improves early stress detection in a New Mexico conifer
woodland. Remote Sens. Environ. 2011, 115, 3640–3646. [CrossRef]
67. Mittler, R.; Blumwald, E. Genetic engineering for modern agriculture: Challenges and perspectives.
Annu. Rev. Plant Biol. 2010, 61, 443–462. [CrossRef] [PubMed]
Remote Sens. 2017, 9, 129 25 of 33
68. York, L.M.; Lynch, J.P. Intensive field phenotyping of maize (Zea mays L.) root crowns identifies phenes and
phene integration associated with plant growth and nitrogen acquisition. J. Exp. Bot. 2015, 66, 5493–5505.
[CrossRef] [PubMed]
69. Lausch, A.; Pause, M.; Merbach, I.; Zacharias, S.; Doktor, D.; Volk, M.; Seppelt, R. A new multiscale
approach for monitoring vegetation using remote sensing-based indicators in laboratory, field, and landscape.
Environ. Monit. Assess. 2013, 185, 1215–1235. [CrossRef] [PubMed]
70. Kraft, S.; Del Bello, U.; Bouvet, M.; Drusch, M.; Moreno, J. FLEX: ESA’s Earth Explorer 8 candidate mission.
In Proceedings of the IEEE International Geoscience and Remote Sensing Symposium, Munich, Germany,
22–27 July 2012.
71. Krajewski, P.; Chen, D.; C´wiek, H.; Van Dijk, A.D.J.; Fiorani, F.; Kersey, P.; Klukas, C.; Lange, M.;
Markiewicz, A.; Nap, J.P.; et al. Towards recommendations for metadata and data handling in plant
phenotyping. J. Exp. Bot. 2015, 66, 5417–5427. [CrossRef] [PubMed]
72. International Plant Phenotyping Network. Available online: http://www.plant-phenotyping.org/ (accessed
on 3 January 2017).
73. Hosoi, F.; Omasa, K. Detecting seasonal change of broad-leaved woody canopy leaf area density profile
using 3D portable LIDAR imaging. Funct. Plant Biol. 2009, 36, 998–1005. [CrossRef]
74. Chen, X. A case study using remote sensing data to compare biophysical properties of a forest and an urban
area in Northern Alabama, USA. J. Sustain. For. 2016, 35, 261–279. [CrossRef]
75. Yang, Y.; Guan, H.; Batelaan, O.; McVicar, T.R.; Long, D.; Piao, S.; Liang, W.; Liu, B.; Jin, Z.; Simmons, C.T.
Contrasting responses of water use efficiency to drought across global terrestrial ecosystems. Sci. Rep. 2016, 6.
[CrossRef] [PubMed]
76. FLUXNET. Available online: http://www.fluxnet.ornl.gov/ (accessed on 3 January 2017).
77. Gamon, J.A.; Rahman, A.F.; Dungan, J.L.; Schildhauer, M.; Huemmrich, K.F. Spectral Network
(SpecNet)-What is it and why do we need it? Remote Sens. Environ. 2006, 103, 227–235. [CrossRef]
78. Spectral network (SpecNet). Available online: http://specnet.info (accessed on 3 January 2017).
79. Hart, J.K.; Martinez, K. Environmental Sensor Networks: A revolution in the earth system science?
Earth-Sci. Rev. 2006, 78, 177–191. [CrossRef]
80. Collins, S.L.; Bettencourt, L.M.; Hagberg, A.; Brown, R.F.; Moore, D.I.; Bonito, G.; Delin, K.A.; Jackson, S.P.;
Johnson, D.W.; Burleigh, S.C.; et al. New opportunities in ecological sensing using wireless sensor networks.
Front. Ecol. Environ. 2006, 4, 402–407. [CrossRef]
81. Clark, J.S.; Agarwal, P.; Bell, D.M.; Flikkema, P.G.; Gelfand, A.; Nguyen, X.L.; Ward, E.; Yang, J. Inferential
ecosystem models, from network data to prediction. Ecol. Appl. 2011, 21, 1523–1536. [CrossRef] [PubMed]
82. Wang, Z.; Chen, N.; Zhang, W.; Pu, F.; Du, C. Semantic integration of wireless sensor networks into open
geospatial consortium sensor observation service to access and share environmental monitoring systems.
IET Softw. 2016, 10, 45–53.
83. Lausch, A.; Schmidt, A.; Tischendorf, L. Data mining and linked open data—New perspectives for data
analysis in environmental research. Ecol. Model. 2015, 295, 5–17. [CrossRef]
84. Wehr, A.; Lohr, U. Airborne laser scanning—An introduction and overview. ISPRS J. Photogramm. Remote Sens.
1999, 54, 68–82. [CrossRef]
85. Popescu, S.C. LiDAR Remote Sensing. Advances in Environmental Remote Sensing: Sensors, Algorithms, and
Applications; CRC Press: Boca Raton, FL, USA, 2011.
86. Geoscience Laser Altimeter System (GLAS). Available online: http://attic.gsfc.nasa.gov/glas (accessed on 3
January 2017).
87. Koch, B.; Kattenborn, T.; Straub, C.; Vauhkonen, J. Segmentation of forest to tree objects. In Forestry
Applications of Airborne Laser Scanning: Concepts and Case Studies; Maltamo, M., Næsset, E., Vauhkonen, J.,
Eds.; Springer Science+Business Media: Dordrecht, The Netherlands, 2014.
88. Wagner, W.; Ullrich, A.; Melzer, T. From Single-Pulse to Full-Waveform Airborne Laser Scanners: Potential
and Practical Challenges. Available online: http://www.isprs.org/proceedings/XXXV/congress/comm3/
papers/267.pdf (accessed on 2 February 2017).
89. Anderson, K.; Hancock, S.; Disney, M.; Gaston, K.J. Is waveform worth it? A comparison of LiDAR
approaches for vegetation and landscape characterization. Remote Sens. Ecol. Conserv. 2016, 2, 5–15.
[CrossRef]
Remote Sens. 2017, 9, 129 26 of 33
90. Reitberger, J.; Krzystek, P.; Stilla, U. Analysis of full waveform LIDAR data for the classification of deciduous
and coniferous trees. Int. J. Remote Sens. 2008, 29, 1407–1431. [CrossRef]
91. Schutz, B.E.; Zwally, H.J.; Shuman, C.A.; Hancock, D.; DiMarzio, J.P. Overview of the ICESat mission.
Geophys. Res. Lett. 2005, 32, 1–4. [CrossRef]
92. Lefsky, M.A.; Harding, D.J.; Keller, M.; Cohen, W.B.; Carabajal, C.C.; Del Bom Espirito-Santo, F.; Hunter, M.O.;
de Oliveira, R. Estimates of forest canopy height and aboveground biomass using ICESat. Geophys. Res. Lett.
2005, 32, 1–4. [CrossRef]
93. Heurich, M.; Kennel, E. Evaluation of tree height determination from airborne laser scanner data over the
richly structured natural forests of the Bavarian Forest National Park. Photogramm. Fernerkungung Geoinf.
2008, 1, 253–263.
94. Magnussen, S.; Boudewyn, P. Derivations of stand heights from airborne laser scanner data with
canopy-based quantile estimators. Can. J. For. Res. 1998, 28, 1016–1031. [CrossRef]
95. Nilsson, M. Estimation of tree heights and stand volume using an airborne LiDAR system. Remote Sens.
Environ. 1996, 56, 1–7. [CrossRef]
96. Polewski, P.; Yao, W.; Heurich, M.; Krzystek, P.; Stilla, U. Detection of fallen trees in ALS point clouds using
a Normalized Cut approach trained by simulation. ISPRS J. Photogramm. Remote Sens. 2015, 105, 252–271.
[CrossRef]
97. Mücke, W.; Deák, B.; Schroiff, A.; Hollaus, M.; Pfeifer, N. Detection of fallen trees in forested areas using
small footprint airborne laser scanning data. Can. J. Remote Sens. 2013, 39, S32–S40. [CrossRef]
98. Jonsson, B.G.; Kruys, N.; Ranius, T. Ecology of species living on dead wood—Lessons for dead wood
management. Silva Fenn. 2005, 39, 289–309. [CrossRef]
99. Woodall, C.W.; Heath, L.S.; Smith, J.E. National inventories of down and dead woody material forest carbon
stocks in the United States: Challenges and opportunities. For. Ecol. Manag. 2008, 256, 221–228. [CrossRef]
100. Næsset, E. Predicting forest stand characteristics with airborne scanning laser using a practical two-stage
procedure and field data. Remote Sens. Environ. 2002, 80, 88–99. [CrossRef]
101. Maltamo, M.; Eerikäinen, K.; Packalén, P.; Hyyppä, J. Estimation of stem volume using laser scanning-based
canopy height metrics. Forestry 2006, 79, 217–229. [CrossRef]
102. Næsset, E. Airborne laser scanning as a method in operational forest inventory: Status of accuracy
assessments accomplished in Scandinavia. Scand. J. For. Res. 2007, 22, 433–442. [CrossRef]
103. Holmgren, J. Prediction of tree height, basal area and stem volume in forest stands using airborne laser
scanning. Scand. J. For. Res. 2004, 19, 543–553. [CrossRef]
104. Maltamo, M.; Packalen, P. Species-specific management inventory in Finland. In Forestry Applications of
Airborne Laser Scanning: Concepts and Case Studies; Maltamo, M., Næsset, E., Vauhkonen, J., Eds.; Springer
Science+Business Media: Dordrecht, The Netherlands, 2014.
105. Heurich, M.; Thoma, F. Estimation of forestry stand parameters using laser scanning data in temperate,
structurally rich natural European beech (Fagus sylvatica) and Norway spruce (Picea abies) forests. Forestry
2008, 81, 645–661. [CrossRef]
106. Latifi, H.; Nothdurft, A.; Koch, B. Non-parametric prediction and mapping of standing timber volume and
biomass in a temperate forest: Application of multiple optical/LiDAR-derived predictors. Forestry 2010, 83,
395–407. [CrossRef]
107. Latifi, H.; Fassnacht, F.E.; Müller, J.; Tharani, A.; Dech, S.; Heurich, M. Forest inventories by LiDAR data:
A comparison of single tree segmentation and metric-based methods for inventories of a heterogeneous
temperate forest. Int. J. Appl. Earth Obs. Geoinf. 2015, 42, 162–174. [CrossRef]
108. Morsdorf, F.; Kötz, B.; Meier, E.; Itten, K.I.; Allgöwer, B. Estimation of LAI and fractional cover from small
footprint airborne laser scanning data based on gap fraction. Remote Sens. Environ. 2006, 104, 50–61.
[CrossRef]
109. Yao, W.; Krzystek, P.; Heurich, M. Tree species classification and estimation of stem volume and DBH based
on single tree extraction by exploiting airborne full-waveform LiDAR data. Remote Sens. Environ. 2012, 123,
368–380. [CrossRef]
110. Heinzel, J.; Koch, B. Investigating multiple data sources for tree species classification in temperate forest and
use for single tree delineation. Int. J. Appl. Earth Obs. Geoinf. 2012, 18, 101–110. [CrossRef]
111. Koenig, K.; Höfle, B. Full-Waveform airborne laser scanning in vegetation studies—A review of point cloud
and waveform features for tree species classification. Forests 2016, 7. [CrossRef]
Remote Sens. 2017, 9, 129 27 of 33
112. Yao, W.; Krzystek, P.; Heurich, M.; Forest, B.; Park, N. Identifying standing dead trees in forest areas based
on 3D Single tree detection from full waveform LiDAR data. In Proceedings of the ISPRS Annals of the
Photogrammetry, Remote Sensing and Spatial Information Sciences, Melbourne, Australia, 5 August–1
September 2012.
113. Polewski, P.; Yao, W.; Heurich, M.; Krzystek, P.; Stilla, U. Combining Active and Semisupervised Learning of
Remote Sensing Data Within a Renyi Entropy Regularization Framework. IEEE J. Sel. Top. Appl. Earth Obs.
Remote Sens. 2016, 9, 2910–2922. [CrossRef]
114. Pesonen, A.; Maltamo, M.; Eerikäinen, K.; Packalèn, P. Airborne laser scanning-based prediction of coarse
woody debris volumes in a conservation area. For. Ecol. Manag. 2008, 255, 3288–3296. [CrossRef]
115. Kim, Y.; Yang, Z.; Cohen, W.B.; Pflugmacher, D.; Lauver, C.L.; Vankat, J.L. Distinguishing between live and
dead standing tree biomass on the North Rim of Grand Canyon National Park, USA using small-footprint
LIDAR data. Remote Sens. Environ. 2009, 113, 2499–2510. [CrossRef]
116. Bright, B.C.; Hudak, A.T.; Kennedy, R.E.; Meddens, A.J.H. Landsat time series and LiDAR as predictors of
live and dead basal area across five bark beetle-affected forests. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
2014, 7, 3440–3452. [CrossRef]
117. Amiri, N.; Yao, W.; Heurich, M.; Krzystek, P.; Skidmore, A.K. Estimation of regeneration coverage in a
temperate forest by 3D segmentation using airborne laser scanning data. Int. J. Appl. Earth Obs. Geoinf. 2016,
52, 252–262. [CrossRef]
118. Maltamo, M.; Gobakken, T. Predicting tree diameter distributions. In Forestry Applications of Airborne Laser
Scanning: Concepts and Case Studies; Maltamo, M., Næsset, E., Vauhkonen, J., Eds.; Springer science+Business
Media: Dordrecht, The Netherlands, 2014.
119. Van Leeuwen, M.; Nieuwenhuis, M. Retrieval of forest structural parameters using LiDAR remote sensing.
Eur. J. For. Res. 2010, 129, 749–770. [CrossRef]
120. Korhonen, L.; Korpela, I.; Heiskanen, J.; Maltamo, M. Airborne discrete-return LiDAR data in the estimation
of vertical canopy cover, angular canopy closure and leaf area index. Remote Sens. Environ. 2011, 115,
1065–1080. [CrossRef]
121. Solberg, S.; Brunner, A.; Hanssen, K.H.; Lange, H.; Næsset, E.; Rautiainen, M.; Stenberg, P. Mapping LAI in a
Norway spruce forest using airborne laser scanning. Remote Sens. Environ. 2009, 113, 2317–2327. [CrossRef]
122. MacArthur, R.H. Environmental factors affecting bird species diversity. Am. Nat. 1964, 98, 387–397.
[CrossRef]
123. Gao, T.; Hedblom, M.; Emilsson, T.; Nielsen, A.B. The role of forest stand structure as biodiversity indicator.
For. Ecol. Manag. 2014, 330, 82–93. [CrossRef]
124. Müller, J.; Moning, C.; Bässler, C.; Heurich, M.; Brandl, R. Using airborne laser scanning to model potential
abundance and assemblages of forest passerines. Basic Appl. Ecol. 2009, 10, 671–681. [CrossRef]
125. Zimble, D.A.; Evans, D.L.; Carlson, G.C.; Parker, R.C.; Grado, S.C.; Gerard, P.D. Characterizing vertical forest
structure using small-footprint airborne LiDAR. Remote Sens. Environ. 2003, 87, 171–182. [CrossRef]
126. Vogeler, J.C.; Hudak, A.T.; Vierling, L.A.; Evans, J.; Green, P.; Vierling, K.T. Terrain and vegetation structural
influences on local avian species richness in two mixed-conifer forests. Remote Sens. Environ. 2014, 147, 13–22.
[CrossRef]
127. Ewald, M.; Dupke, C.; Heurich, M.; Müller, J.; Reineking, B. LiDAR remote sensing of forest structure
and GPS telemetry data provide insights on winter habitat selection of european roe deer. Forests 2014, 5,
1374–1390. [CrossRef]
128. Ole Ørka, H.; Gobakken, T.; Næsset, E. Predicting Attributes of Regeneration Forests Using Airborne Laser
Scanning. Can. J. Remote Sens. 2016, 42, 541–553. [CrossRef]
129. Amiri, N.; Yao, W.; Heurich, M.; Krzystek, P. Regeneration detection by 3D segmentation in a temperate
forest using airborne full waveform LiDAR data. In Proceedings of the SilviLaser 2015: 14th Conference
on LiDAR Applications for Assessing and Managing Forest Ecosystems, La Grande Motte, France, 28–30
September 2015.
130. Andersen, H.E.; McGaughey, R.J.; Reutebuch, S.E. Estimating forest canopy fuel parameters using LIDAR
data. Remote Sens. Environ. 2005, 94, 441–449. [CrossRef]
131. Skowronski, N.S.; Clark, K.L.; Duveneck, M.; Hom, J. Three-dimensional canopy fuel loading predicted
using upward and downward sensing LiDAR systems. Remote Sens. Environ. 2011, 115, 703–714. [CrossRef]
Remote Sens. 2017, 9, 129 28 of 33
132. Eitel, J.U.H.; Höfle, B.; Vierling, L.A.; Abellán, A.; Asner, G.P.; Deems, J.S.; Glennie, C.L.; Joerg, P.C.;
LeWinter, A.L.; Magney, T.S.; et al. Beyond 3-D: The new spectrum of LiDAR applications for earth and
ecological sciences. Remote Sens. Environ. 2016, 186, 372–392. [CrossRef]
133. Stysley, P.R.; Coyle, D.B.; Kay, R.B.; Frederickson, R.; Poulios, D.; Cory, K.; Clarke, G. Long term performance
of the High Output Maximum Efficiency Resonator (HOMER) laser for NASA’s Global Ecosystem Dynamics
Investigation (GEDI) LiDAR. Opt. Laser Technol. 2015, 68, 67–72. [CrossRef]
134. Thenkabail, P.S. Hyperspectral Remote Sensing of Vegetation; CRC Press Inc.: Boca Raton, FL, USA, 2011.
135. Thenkabail, P.S.; Enclona, E.A.; Ashton, M.S.; Legg, C.; De Dieu, M.J. Hyperion, IKONOS, ALI, and ETM+
sensors in the study of African rainforests. Remote Sens. Environ. 2004, 90, 23–43. [CrossRef]
136. Lausch, A.; Bannehr, L.; Beckmann, M.; Boehm, C.; Feilhauer, H.; Hacker, J.M.; Heurich, M.; Jung, A.;
Klenke, R.; Neumann, C.; et al. Linking Earth Observation and taxonomic, structural and functional
biodiversity: Local to ecosystem perspectives. Indic. Ecol. 2016, 70, 317–339. [CrossRef]
137. Kuenzer, C.; Bluemel, A.; Gebhardt, S.; Quoc, T.V.; Dech, S. Remote Sensing of Mangrove Ecosystems:
A Review. Remote Sens. 2011, 3, 878–928. [CrossRef]
138. Joshi, N.; Mitchard, E.T.A.; Woo, N.; Torres, J.; Moll-rocek, J.; Ehammer, A. Mapping dynamics of
deforestation and forest degradation in tropical forests using radar satellite data. Environ. Res. Lett.
2015, 10. [CrossRef]
139. Joshi, N.; Baumann, M.; Ehammer, A.; Fensholt, R.; Grogan, K.; Hostert, P.; Jepsen, M.R.; Kuemmerle, T.;
Meyfroidt, P.; Mitchard, E.T.A.; et al. A review of the application of optical and radar remote sensing data
fusion to land use mapping and monitoring. Remote Sens. 2016, 8, 70. [CrossRef]
140. Andersen, H.-E.; Reutebuch, S.E.; McGaughey, R.J. Accuracy of an {IFSAR-derived} digital terrain model
under a conifer forest canopy. Can. J. Remote Sens. 2005, 31, 283–288. [CrossRef]
141. Heurich, M. Automatic recognition and measurement of single trees based on data from airborne laser
scanning over the richly structured natural forests of the Bavarian Forest National Park. For. Ecol. Manag.
2008, 255, 2416–2433. [CrossRef]
142. Meng, X.; Currit, N.; Zhao, K. Ground filtering algorithms for airborne LiDAR data: A review of critical
issues. Remote Sens. 2010, 2, 833–860. [CrossRef]
143. Kraus, K.; Pfeifer, N. Determination of terrain models in wooded areas with airborne laser scanner data.
ISPRS J. Photogramm. Remote Sens. 1998, 53, 193–203. [CrossRef]
144. Latifi, H.; Fassnacht, F.E.; Hartig, F.; Berger, C.; Hernández, J.; Corvalán, P.; Koch, B. Stratified aboveground
forest biomass estimation by remote sensing data. Int. J. Appl. Earth Obs. Geoinf. 2015, 38, 229–241. [CrossRef]
145. Persson, A.; Holmgren, J.; Söderman, U. Detecting and measuring individual trees using an airborne laser
scanner. Photogramm. Eng. Remote Sens. 2002, 68, 925–932.
146. Morsdorf, F.; Meier, E.; Kötz, B.; Itten, K.I.; Dobbertin, M.; Allgöwer, B. LIDAR-based geometric
reconstruction of boreal type forest stands at single tree level for forest and wildland fire management.
Remote Sens. Environ. 2004, 92, 353–362. [CrossRef]
147. Henderson, F.M.; Lewis, A.J. Principles and Applications of Imaging Radar—Manual of Remote Sensing; Wiley:
Hoboken, NJ, USA, 1998.
148. Ulaby, F.; Held, D.; Donson, M.; McDonald, K.; Senior, T.A. Relating Polaization Phase Difference of SAR
Signals to Scene Properties. IEEE Trans. Geosci. Remote Sens. 1987, GE-25, 83–92. [CrossRef]
149. Balzter, H. Forest mapping and monitoring with interferometric synthetic aperture radar (InSAR).
Prog. Phys. Geogr. 2009, 25, 159–177. [CrossRef]
150. Hamlyn, G.; Jones, R.A.V. Remote Sensing of Vegetation; Oxford University Press: Oxford, UK, 2010.
151. Peake, W.; Oliver, T. The Response of Terrestrial Surface at Microwave Frequencies; Defense Technical Information
Center: Fort Belvoir, VA, USA, 1971.
152. Van Zyl, J.J. Unsupervised classification of scattering behavior using radar polarimetry data. IEEE Trans.
Geosci. Remote Sens. 1989, 27, 36–45. [CrossRef]
153. Claude, S.R.; Pottier, E. A Review of Target Decomposition Theorems in Radar Polarimetry. IEEE Trans.
Geosci. Remote Sens. 1996, 34, 498–518. [CrossRef]
154. Freeman, A.; Durden, S.L. A three-component scattering model for polarimetric SAR data. IEEE Trans. Geosci.
Remote Sens. 1998, 36, 963–973. [CrossRef]
155. Touzi, R. Target scattering decomposition in terms of roll-invariant target parameters. IEEE Trans. Geosci.
Remote Sens. 2007, 45, 73–84. [CrossRef]
Remote Sens. 2017, 9, 129 29 of 33
156. Kellndorfer, J.M.; Walker, W.S.; Dobson, M.C.; Vona, M.; Clutter, M. Vegetation height derivation from
Shuttle Radar Topography Mission data in southeast Georgia, USA. In Proceedings of the IEEE International
Geoscience and Remote Sensing Symposium, Anchorage, AK, USA, 20–24 September 2004.
157. Tighe, M. Empirical Assessment of Multi-Wavelength Synthetic Aperture Radar for Land Cover and Canopy
Height Estimation. Ph.D. Thesis, Carleton University, Ottawa, ON, Canada, 2012.
158. Neeff, T.; Dutra, L.V.; Dos Santos, J.R.; Da Costa Freitas, C.; Araujo, L.S. Tropical forest measurement by
interferometric height modeling and P-band radar backscatter. For. Sci. 2005, 51, 585–594.
159. Balzter, H.; Rowland, C.S.; Saich, P. Forest canopy height and carbon estimation at Monks Wood National
Nature Reserve, UK, using dual-wavelength SAR interferometry. Remote Sens. Environ. 2007, 108, 224–239.
[CrossRef]
160. Andersen, H.-E.; McGaughey, R.J.; Reutebuch, S.E. Assessing the influence of flight parameters,
interferometric processing, slope and canopy density on the accuracy of X-band IFSAR-derived forest
canopy height models. Int. J. Remote Sens. 2008, 29, 1495–1510. [CrossRef]
161. Tighe, M.L.; Balzter, H.; McNairn, H. Comparison of X/C-HH InSAR and L-PolInSAR for canopy height
estimation in a lodgepole pine forest. In Proceedings of the 4th International Workshop on Science and
Applications of SAR Polarimetry and Polarimetric Interferometry, Frascati, Italy, 26–30 January 2009.
162. Woodhouse, I.H. Introduction to Microwave Remote Sensing; Taylor and Francis Group: Boca Raton, FL,
USA, 2006.
163. Touzi, R.; Lopes, A. The principle of speckle filtering in polarimetric SAR imagery. IEEE Trans. Geosci.
Remote Sens. 1994, 32, 1110–1114. [CrossRef]
164. Lee, J.S.; Grunes, M.R.; De Grandi, G. Polarimetric SAR speckle filtering and its impact on classification.
In Proceedings of the IEEE International Geoscience and Remote Sensing Symposium, Singapore,
3–8 August 1997.
165. Lu, D. The potential and challenge of remote sensing-based biomass estimation. Int. J. Remote Sens. 2006, 27,
1297–1328. [CrossRef]
166. Lutz, D.A.; Allen, R.A.W.; Shugart, H.H. Remote sensing of boreal forest biophysical and inventory
parameters: A review. Can. J. Remote Sens. 2008, 34, S286–S313. [CrossRef]
167. Koch, B. Status and future of laser scanning, synthetic aperture radar and hyperspectral remote sensing data
for forest biomass assessment. ISPRS J. Photogramm. Remote Sens. 2010, 65, 581–590. [CrossRef]
168. Timothy, D.; Onisimo, M.; Cletah, S.; Adelabu, S.; Tsitsi, B. Remote sensing of aboveground forest biomass:
A review. Trop. Ecol. 2016, 57, 125–132.
169. Assal, T.J.; Anderson, P.J.; Sibold, J. Spatial and temporal trends of drought effects in a heterogeneous
semi-arid forest ecosystem. For. Ecol. Manag. 2016, 365, 137–151. [CrossRef]
170. Luckman, A.; Baker, J.; Kuplich, T.M.; Corina da Costa, F.Y.; Alejandro, C.F. A study of the relationship
between radar backscatter and regenerating tropical forest biomass for spaceborne SAR instruments. Remote
Sens. Environ. 1997, 60, 1–13. [CrossRef]
171. Lucas, R.M.; Cronin, N.; Lee, A.; Moghaddam, M.; Witte, C.; Tickle, P. Empirical relationships between
AIRSAR backscatter and LiDAR-derived forest biomass, Queensland, Australia. Remote Sens. Environ. 2006,
100, 407–425. [CrossRef]
172. Le Toan, T.; Quegan, S.; Davidson, M.W.J.; Balzter, H.; Paillou, P.; Papathanassiou, K.; Plummer, S.; Rocca, F.;
Saatchi, S.; Shugart, H.; et al. The BIOMASS mission: Mapping global forest biomass to better understand
the terrestrial carbon cycle. Remote Sens. Environ. 2011, 115, 2850–2860. [CrossRef]
173. Rignot, E.; Salas, W.A.; Skole, D.L. Mapping deforestation and secondary growth in Rondonia, Brazil, using
imaging radar and thematic mapper data. Remote Sens. Environ. 1997, 59, 167–179. [CrossRef]
174. Van der Sanden, J.J.; Hoekman, D.H. Potential of Airborne Radar To Support the Assessment of Land Cover
in a Tropical Rain Forest Environment. Remote Sens. Environ. 1999, 68, 26–40. [CrossRef]
175. Thapa, R.B.; Shimada, M.; Watanabe, M.; Motohka, T.; Shiraishi, T. The tropical forest in south east Asia:
Monitoring and scenario modeling using synthetic aperture radar data. Appl. Geogr. 2013, 41, 168–178.
[CrossRef]
176. Häme, T.; Rauste, Y.; Antropov, O.; Ahola, H.A.; Kilpi, J. Improved Mapping of Tropical Forests With Optical
Estimation. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2013, 6, 92–101. [CrossRef]
Remote Sens. 2017, 9, 129 30 of 33
177. Santos, J.R.; Freitas, C.C.; Araujo, L.S.; Dutra, L.V.; Mura, J.C.; Gama, F.F.; Soler, L.S.; Sant, A.; Anna, S.J.S.
Airborne P-band SAR applied to the aboveground biomass studies in the Brazilian tropical rainforest. Remote
Sens. Environ. 2003, 87, 482–493. [CrossRef]
178. Foody, G.M.; Green, R.M.; Lucas, R.M.; Curran, P.J.; Honzak, M.; Do Amaral, I. Observations on the
relationship between SIR-C radar backscatter and the biomass of regenerating tropical forests. Int. J.
Remote Sens. 1997, 18, 687–694. [CrossRef]
179. Englhart, S.; Keuck, V.; Siegert, F. Aboveground biomass retrieval in tropical forests—The potential of
combined X- and L-band SAR data use. Remote Sens. Environ. 2011, 115, 1260–1271. [CrossRef]
180. Mitchard, E.T.A.; Saatchi, S.S.; Lewis, S.L.; Feldpausch, T.R.; Woodhouse, I.H.; Sonké, B.; Rowland, C.; Meir, P.
Measuring biomass changes due to woody encroachment and deforestation/degradation in a forest-savanna
boundary region of central Africa using multi-temporal L-band radar backscatter. Remote Sens. Environ.
2011, 115, 2861–2873. [CrossRef]
181. Simard, M.; Rivera-Monroy, V.H.; Mancera-Pineda, J.E.; Castañeda-Moya, E.; Twilley, R.R. A systematic
method for 3D mapping of mangrove forests based on Shuttle Radar Topography Mission elevation data,
ICEsat/GLAS waveforms and field data: Application to Ciénaga Grande de Santa Marta, Colombia.
Remote Sens. Environ. 2008, 112, 2131–2144. [CrossRef]
182. Manninen, T.; Stenberg, P.; Rautiainen, M.; Voipio, P. Leaf area index estimation of boreal and subarctic
forests Using VV/HH ENVISAT/ASAR data of various swaths. IEEE Trans. Geosci. Remote Sens. 2013, 51,
3899–3909. [CrossRef]
183. Varghese, A.O.; Suryavanshi, A.; Joshi, A.K. Analysis of different polarimetric target decompositionmethods
in forest density classification using C band SAR data. Int. J. Remote Sens. 2016, 37, 694–709. [CrossRef]
184. Kasischke, E.S.; Bourgeauchavez, L.L.; French, N.H.F.; Harrell, P.; Christensen, N.L. Initial Observations on
Using Sar to Monitor Wildfire Scars in Boreal Forests. Int. J. Remote Sens. 1992, 13, 3495–3501. [CrossRef]
185. Chavez, L.L.B.; Kasischke, E.S.; French, N.H.F.; Szeto, L.H.; Kherkher, C.M. Using ERS-1 SAR imagery to
monitor variations in burn severity in \nan Alaskan fire-disturbed boreal forest ecosystem. In Proceedings of
the International Geoscience and Remote Sensing Symposium—Surface and Atmospheric Remote Sensing:
Technologies, Data Analysis, and Interpretation, Pasadena, CA, USA, 8–12 August 1994.
186. Tanase, M.A.; Kennedy, R.; Aponte, C. Radar Burn Ratio for fire severity estimation at canopy level:
An example for temperate forests. Remote Sens. Environ. 2015, 170, 14–31. [CrossRef]
187. Martins, F.; da, S.R.V.; dos Santos, J.R.; Galvão, L.S.; Xaud, H.A.M. Sensitivity of ALOS/PALSAR imagery to
forest degradation by fire in northern Amazon. Int. J. Appl. Earth Obs. Geoinf. 2016, 49, 163–174. [CrossRef]
188. Lang, M.W.; Townsend, P.A.; Kasischke, E.S. Influence of incidence angle on detecting flooded forests using
C-HH synthetic aperture radar data. Remote Sens. Environ. 2008, 112, 3898–3907. [CrossRef]
189. Kim, J.W.; Lu, Z.; Lee, H.; Shum, C.K.; Swarzenski, C.M.; Doyle, T.W.; Baek, S.H. Integrated analysis of
PALSAR/Radarsat-1 InSAR and ENVISAT altimeter data for mapping of absolute water level changes in
Louisiana wetlands. Remote Sens. Environ. 2009, 113, 2356–2365. [CrossRef]
190. Kuenzer, C.; Ottinger, M.; Wegmann, M.; Guo, H. Earth observation satellite sensors for biodiversity
monitoring: Potentials and bottlenecks. Int. J. Remote Sens. 2014, 35, 6599–6647. [CrossRef]
191. Kim, M.H.; Rushmeier, H.; Dorsey, J.; Harvey, T.A.; Prum, R.O.; Kittle, D.S.; Brady, D.J. 3D imaging
spectroscopy for measuring hyperspectral patterns on solid objects. ACM Trans. Graph. 2012, 31, 1–11.
[CrossRef]
192. Kim, M.H. Foundations and Applications of 3D Imaging. Theory and Applications of Smart Cameras; Springer:
Berlin, Germany, 2016.
193. Lagouarde, J.P.; Ballans, H.; Moreau, P.; Guyon, D.; Coraboeuf, D. Experimental study of brightness surface
temperature angular variations of maritime pine (Pinus pinaster) stands. Remote Sens. Environ. 2000, 72,
17–34. [CrossRef]
194. Duffour, C.; Lagouarde, J.P.; Olioso, A.; Demarty, J.; Roujean, J.L. Driving factors of the directional variability
of thermal infrared signal in temperate regions. Remote Sens. Environ. 2016, 177, 248–264. [CrossRef]
195. Giuliani, R.; Magnanini, E.; Flore, J.A. Potential Use of Infrared Thermometry For The Detection Of Water
Deficit In Apple And Peach Orchards. Acta Hortic. 2001, 557, 399–406. [CrossRef]
196. Luquet, D.; Vidal, A.; Dauzat, J.; Bégué, A.; Olioso, A.; Clouvel, P. Using directional TIR measurements and
3D simulations to assess the limitations and opportunities of water stress indices. Remote Sens. Environ. 2004,
90, 53–62. [CrossRef]
Remote Sens. 2017, 9, 129 31 of 33
197. Guy Séguin, D.G. RADARSAT constellation mission overview. In Proceedings of the ASTRO 2010, 15th
CASI (Canadian Aeronautics and Space Institute) Conference, Toronto, ON, Canada, 4–6 May 2010.
198. Pang, Y.; Li, Z.; Ju, H.; Lu, H.; Jia, W.; Si, L.; Guo, Y.; Liu, Q.; Li, S.; Liu, L.; et al. LiCHy: The CAF’s LiDAR,
CCD and hyperspectral integrated airborne observation system. Remote Sens. 2016, 8. [CrossRef]
199. HyspIRI Mission (HyspIRI). Available online: http://hyspiri.jpl.nasa.gov/ (accessed on 3 January 2017).
200. Lee, C.M.; Cable, M.L.; Hook, S.J.; Green, R.O.; Ustin, S.L.; Mandl, D.J.; Middleton, E.M. An introduction
to the NASA Hyperspectral InfraRed Imager (HyspIRI) mission and preparatory activities. Remote Sens.
Environ. 2015, 167, 6–19. [CrossRef]
201. Asner, G.P.; Martin, R.E.; Suhaili, A. Bin sources of canopy chemical and spectral diversity in lowland
Bornean forest. Ecosystems 2012, 15, 504–517. [CrossRef]
202. Asner, G.P.; Martin, R.E.; Tupayachi, R.; Anderson, C.B.; Sinca, F.; Jiménez, L.C.; Martinez, P. Amazonian
functional diversity from forest canopy chemical assembly. Proc. Natl. Acad. Sci. USA 2014, 111, 5604–5609.
[CrossRef] [PubMed]
203. Ustin, S.L. Remote sensing of canopy chemistry. Proc. Natl. Acad. Sci. USA 2013, 110, 804–805. [CrossRef]
[PubMed]
204. Feret, J.B.; Asner, P.G. Tree species discrimination in tropical forests using airborne imaging spectroscopy.
IEEE Trans. Geosci. Remote Sens. 2013, 51, 73–84. [CrossRef]
205. Richter, R.; Reu, B.; Wirth, C.; Doktor, D. The use of airborne hyperspectral data for tree species classification
in a species-rich Central European forest area. Int. J. Appl. Earth Obs. Geoinf. 2016, 52, 464–474. [CrossRef]
206. Möckel, T.; Löfgren, O.; Prentice, H.C.; Eklundh, L.; Hall, K. Airborne hyperspectral data predict Ellenberg
indicator values for nutrient and moisture availability in dry grazed grasslands within a local agricultural
landscape. Ecol. Indic. 2016, 66, 503–516. [CrossRef]
207. Schmidtlein, S.; Feilhauer, H.; Bruelheide, H. Mapping plant strategy types using remote sensing. J. Veg. Sci.
2012, 23, 395–405. [CrossRef]
208. Fassnacht, F.E.; Latifi, H.; Ghosh, A.; Joshi, P.K.; Koch, B. Assessing the potential of hyperspectral imagery to
map bark beetle-induced tree mortality. Remote Sens. Environ. 2014, 140, 533–548. [CrossRef]
209. Wulder, M.A.; Dymond, C.C.; White, J.C.; Leckie, D.G.; Carroll, A.L. Surveying mountain pine beetle damage
of forests: A review of remote sensing opportunities. For. Ecol. Manag. 2006, 221, 27–41. [CrossRef]
210. Fassnacht, F.E.; Latifi, H.; Steren´czak, K.; Modzelewska, A.; Lefsky, M.; Waser, L.T.; Straub, C.; Ghosh, A.
Review of studies on tree species classification from remotely sensed data. Remote Sens. Environ. 2016, 186,
64–87. [CrossRef]
211. Immitzer, M.; Atzberger, C.; Koukal, T. Tree species classification with Random forest using very high spatial
resolution 8-band WorldView-2 satellite data. Remote Sens. 2012, 4, 2661–2693. [CrossRef]
212. Wallace, L.; Lucieer, A.; Watson, C.; Turner, D. Development of a UAV-LiDAR system with application to
forest inventory. Remote Sens. 2012, 4, 1519–1543. [CrossRef]
213. Zhou, G.; Li, C.; Cheng, P. Unmanned aerial vehicle (UAV) real-time video registration for forest fire
monitoring. In Proceedings of the IEEE International Geoscience and Remote Sensing Symposium, Seoul,
South Korea, 25–29 July 2005.
214. Sheng, H.; Chao, H.; Coopmans, C.; Han, J.; McKee, M.; Chen, Y. Low-cost UAV-based thermal infrared
remote sensing: Platform, calibration and applications. In Proceedings of the 2010 IEEE/ASME International
Conference on Mechatronic and Embedded Systems and Applications, Qingdao, China, 15–17 July 2010.
215. Nebiker, S.; Annen, A.; Scherrer, M.; Oesch, D. A light-weight multispectral sensor for micro
UAV—Opportunities for very high resolution airborne remote sensing. Int. Arch. Photogramm. Remote Sens.
Spat. Inf. Sci. 2008, XXXVII, 1193–1200.
216. Getzin, S.; Nuske, R.S.; Wiegand, K. Using unmanned aerial vehicles (UAV) to quantify spatial gap patterns
in forests. Remote Sens. 2014, 6, 6988–7004. [CrossRef]
217. Mäkynen, J.; Saari, H.; Holmlund, C.; Mannila, R.; Antila, T. Multi- and hyperspectral UAV imaging system
for forest and agriculture applications. In Proceedings of the SPIE Defense, Security, and Sensing, Baltimore,
MD, USA, 23 April 2012.
218. Gamba, P.; Chanussot, J. Foreword to the Special Issue on Data Fusion. IEEE Trans. Geosci. Remote Sens. 2008,
46, 1283–1288. [CrossRef]
219. Ghassemian, H. A review of remote sensing image fusion methods. Inf. Fusion 2016, 32, 75–89. [CrossRef]
Remote Sens. 2017, 9, 129 32 of 33
220. Gao, F.; Masek, J.; Schwaller, M.; Hall, F. On the blending of the landsat and MODIS surface reflectance:
Predicting daily landsat surface reflectance. IEEE Trans. Geosci. Remote Sens. 2006, 44, 2207–2218.
221. Hilker, T.; Wulder, M.A.; Coops, N.C.; Seitz, N.; White, J.C.; Gao, F.; Masek, J.G.; Stenhouse, G. Generation of
dense time series synthetic Landsat data through data blending with MODIS using a spatial and temporal
adaptive reflectance fusion model. Remote Sens. Environ. 2009, 113, 1988–1999. [CrossRef]
222. Cosmopoulos, P.; King, D.J. Temporal analysis of forest structural condition at an acid mine site using
multispectral digital camera imagery. Int. J. Remote Sens. 2004, 25, 2259–2275. [CrossRef]
223. King, D.J.; Olthof, I.; Pellikka, P.K.E.; Seed, E.D.; Butson, C. Modelling and mapping damage to forests from
an ice storm using remote sensing and environmental data. Nat. Hazards 2005, 35, 321–342. [CrossRef]
224. Song, C. Optical remote sensing of forest leaf area index and biomass. Prog. Phys. Geogr. 2013, 37, 98–113.
[CrossRef]
225. Lévesque, J.; King, D.J. Airborne digital camera image semivariance for evaluation of forest structural
damage at an acid mine site. Remote Sens. Environ. 1999, 68, 112–124. [CrossRef]
226. Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [CrossRef]
227. Millard, K.; Richardson, M.C. On the importance of training data sample selection in RF classification: A case
study in peatland ecosystem mapping. Remote Sens. 2015, 7, 1–26. [CrossRef]
228. Zhu, X.; Liu, D. Accurate mapping of forest types using dense seasonal landsat time-series. ISPRS J.
Photogramm. Remote Sens. 2014, 96, 1–11. [CrossRef]
229. Goel, N.S. Models of vegetation canopy reflectance and their use in estimation of biophysical parameters
from reflectance data. Remote Sens. Rev. 1988, 4, 1–212. [CrossRef]
230. Li, X.; Strahler, A. Geometric-Optical Modeling of a Conifer Forest Canopy. IEEE Trans. Geosci. Remote Sens.
1985, GE-23, 705–721. [CrossRef]
231. Omari, K.; White, H.P.; Staenz, K. King Retrieval of Forest Canopy Parameters by Inversion of the PROFLAIR
Leaf-Canopy Reflectance Model Using the LUT Approach. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
2013, 6, 715–723. [CrossRef]
232. Knyazikhin, Y.; Martonchik, J.V.; Myneni, R.B.; Diner, D.J.; Running, S.W. Synergistic algorithm for estimating
vegetation canopy leaf area index and fraction of absorbed photosynthetically active radiation from MODIS
and MISR data. J. Geophys. Res. Atmos. 1998, 103, 32257–32275. [CrossRef]
233. Ganguly, S.; Nemani, R.R.; Zhang, G.; Hashimoto, H.; Milesi, C.; Michaelis, A.; Wang, W.; Votava, P.;
Samanta, A.; Melton, F.; et al. Generating global Leaf Area Index from Landsat: Algorithm formulation and
demonstration. Remote Sens. Environ. 2012, 122, 185–202. [CrossRef]
234. Propastin, P.; Ibrom, A.; Knohl, A.; Erasmi, S. Effects of canopy photosynthesis saturation on the estimation
of gross primary productivity from MODIS data in a tropical forest. Remote Sens. Environ. 2012, 121, 252–260.
[CrossRef]
235. Fang, H.; Liang, S. A hybrid inversion method for mapping leaf area index from MODIS data: Experiments
and application to broadleaf and needleleaf canopies. Remote Sens. Environ. 2005, 94, 405–424. [CrossRef]
236. Garestier, F.; Dubois-Fernandez, P.C.; Papathanassiou, K.P. Pine Forest Height Inversion Using Single-Pass
X-Band PolInSAR Data. IEEE Trans. Geosci. Remote Sens. 2008, 46, 59–68. [CrossRef]
237. Behera, M.D.; Tripathi, P.; Mishra, B.; Kumar, S.; Chitale, V.S.; Behera, S.K. Above-ground biomass and carbon
estimates of Shorea robusta and Tectona grandis forests using QuadPOL ALOS PALSAR data. Adv. Space
Res. 2016, 57, 552–561. [CrossRef]
238. Wang, C.; Wang, L.; Fu, H.; Xie, Q.; Zhu, J. The impact of forest density on forest height inversion modeling
from polarimetric InSAR data. Remote Sens. 2016, 8. [CrossRef]
239. Rahman, M.M.; Csaplovics, E.; Koch, B. Satellite estimation of forest carbon using regression models. Int. J.
Remote Sens. 2008, 29, 6917–6936. [CrossRef]
240. Foody, G.M.; Boyd, D.S.; Cutler, M.E.J. Predictive relations of tropical forest biomass from Landsat TM data
and their transferability between regions. Remote Sens. Environ. 2003, 85, 463–474. [CrossRef]
241. Schlund, M.; von Poncet, F.; Kuntz, S.; Schmullius, C.; Hoekman, D.H. TanDEM-X data for aboveground
biomass retrieval in a tropical peat swamp forest. Remote Sens. Environ. 2015, 158, 255–266. [CrossRef]
242. Monnet, J.-M.; Chanussot, J.; Berger, F. Support Vector Regression for the Estimation of Forest Stand
Parameters Using Airborne Laser Scanning. IEEE Geosci. Remote Sens. Lett. 2011, 8, 580–584. [CrossRef]
243. Powell, J.; Kennedy, B.; White, P.; Bentz, B.; Logan, J.; Roberts, D. Mathematical elements of attack risk
analysis for mountain pine beetles. J. Theor. Biol. 2000, 204, 601–620. [CrossRef] [PubMed]
Remote Sens. 2017, 9, 129 33 of 33
244. Cohen, W.B.; Maiersperger, T.K.; Gower, S.T.; Turner, D.P. An improved strategy for regression of biophysical
variables and Landsat ETM+ data. Remote Sens. Environ. 2003, 84, 561–571. [CrossRef]
245. Pasher, J.; King, D.J. Multivariate forest structure modelling and mapping using high resolution airborne
imagery and topographic information. Remote Sens. Environ. 2010, 114, 1718–1732. [CrossRef]
246. Pasher, J.; King, D.J. Development of a forest structure complexity index based on multispectral airborne
remote sensing and topographic data. Can. J. For. Res. Can. Rech. For. 2011, 41, 44–58. [CrossRef]
247. Czerwinski, C.J.; King, D.J.; Mitchell, S.W. Mapping forest growth and decline in a temperate mixed forest
using temporal trend analysis of Landsat imagery, 1987–2010. Remote Sens. Environ. 2014, 141, 188–200.
[CrossRef]
248. Fuchs, H.; Magdon, P.; Kleinn, C.; Flessa, H. Estimating aboveground carbon in a catchment of the Siberian
forest tundra: Combining satellite imagery and field inventory. Remote Sens. Environ. 2009, 113, 518–531.
[CrossRef]
249. Donmez, C.; Berberoglu, S.; Erdogan, M.A.; Tanriover, A.A.; Cilek, A. Response of the regression tree model
to high resolution remote sensing data for predicting percent tree cover in a Mediterranean ecosystem.
Environ. Monit. Assess. 2015, 187. [CrossRef] [PubMed]
250. Gómez, C.; Wulder, M.A.; Montes, F.; Delgado, J.A. Modeling forest structural parameters in the
Mediterranean pines of Central Spain using QuickBird-2 imagery and classification and regression tree
analysis (CART). Remote Sens. 2012, 4, 135–159. [CrossRef]
251. Knorn, M.M.; Moisen, G.G.; Healey, S.P.; Keeton, W.S.; Freeman, E.A.; Hostert, P. Evaluating the remote
sensing and inventory-based estimation of biomass in the Western Carpathians. Remote Sens. 2011, 3,
1427–1446. [CrossRef]
252. Li, C.; Song, J.; Wang, J. Modifying geometric-optical bidirectional reflectance model for direct inversion of
forest canopy leaf area index. Remote Sens. 2015, 7, 11083–11104. [CrossRef]
253. Soenen, S.A.; Peddle, D.R.; Hall, R.J.; Coburn, C.A.; Hall, F.G. Estimating aboveground forest biomass from
canopy reflectance model inversion in mountainous terrain. Remote Sens. Environ. 2010, 114, 1325–1337.
[CrossRef]
254. Croft, H.; Chen, J.M.; Zhang, Y.; Simic, A.; Noland, T.L.; Nesbitt, N.; Arabian, J. Evaluating leaf chlorophyll
content prediction from multispectral remote sensing data within a physically-based modelling framework.
ISPRS J. Photogramm. Remote Sens. 2015, 102, 85–95. [CrossRef]
255. Rautiainen, M.; Stenberg, P.; Nilson, T.; Kuusk, A.; Smolander, H. Application of a forest reflectance model in
estimating leaf area index of Scots pine stands using Landsat-7 ETM reflectance data. Can. J. Remote Sens.
2003, 29, 314–323. [CrossRef]
256. Vauhkonen, J.; Holopainen, M.; Kankare, V.; Vastaranta, M.; Viitala, R. Geometrically explicit description of
forest canopy based on 3D triangulations of airborne laser scanning data. Remote Sens. Environ. 2016, 173,
248–257. [CrossRef]
257. Kattge, J.; Díaz, S.; Lavorel, S.; Prentice, I.C.; Leadley, P.; Bönisch, G.; Garnier, E.; Westoby, M.; Reich, P.B.;
Wright, I.J.; et al. TRY - a global database of plant traits. Glob. Chang. Biol. 2011, 17, 2905–2935. [CrossRef]
258. Kühn, I.; Durka, W.; Klotz, S.; Diversity, S.; Issue, S.; Invasion, P.; Sep, E. BiolFlor: A new plant-trait database
as a tool for plant invasion ecology linked references are available on JSTOR for this article: database as a
tool for plant invasion ecology. Divers. Distrib. 2004, 10, 363–365.
259. De Vries, F.T.; Manning, P.; Tallowin, J.R.B.; Mortimer, S.R.; Pilgrim, E.S.; Harrison, K.A.; Hobbs, P.J.;
Quirk, H.; Shipley, B.; Cornelissen, J.H.C.; et al. Abiotic drivers and plant traits explain landscape-scale
patterns in soil microbial communities. Ecol. Lett. 2012, 15, 1230–1239. [CrossRef] [PubMed]
260. Knapp, S.; Kühn, I.; Schweiger, O.; Klotz, S. Challenging urban species diversity: Contrasting phylogenetic
patterns across plant functional groups in Germany. Ecol. Lett. 2008, 11, 1054–1064. [CrossRef] [PubMed]
261. Knapp, S.; Kühn, I.; Bakker, J.P.; Kleyer, M.; Klotz, S.; Ozinga, W.A.; Poschlod, P.; Thompson, K.; Thuiller, W.;
Römermann, C. How species traits and affinity to urban land use control large-scale species frequency.
Divers. Distrib. 2009, 15, 533–546. [CrossRef]
262. Ustin, S.L.; Gamon, J.A. Remote sensing of plant functional types. New Phytol. 2010, 186, 795–816. [CrossRef]
[PubMed]
© 2017 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).