IEEE Geoscience and Remote Sensing Magazine - March 2020 - 11

RESEARCH
Analyzing publications is one way of measuring the importance of a topic [3]; another is to see what projects are
actually done. Here, we determine the proportion of SARand InSAR-related research projects in the Dragon framework over the years. Dragon is a strongly SAR-friendly
project, as the ESA and Chinese research community are
both very involved in SAR and InSAR research, as discussed previously.
In our analysis, shown in Figure 6, we consider projects
during the four Dragon periods. We included reports and
extracted projects that involve SAR or InSAR and a third
category of projects without any SAR-related work. Because of the change in the project structure of Dragon-4,
we consider each subproject individually. We did not assess the extent of SAR or InSAR usage within a project,
only the mention of the use of these methods or the use
of SAR data.
As we can see from Figure 6, the proportion of SARand non-SAR-related projects stayed relatively constant
over time, with approximately 50% of projects having no
SAR contribution. This also shows the relatively strong
focus of Dragon on SAR and InSAR. Based on our bibliometric analysis, the norm is approximately 25% SARrelated projects.
The big changes are in the InSAR-related projects. Throughout
the Dragon program, we see a comparative overrepresentation of InSAR-related projects, with approximately 20% of
all projects having an InSAR component in Dragon-1 and
Dragon-3. We see a relative decline in Dragon-2, which
reflects the decrease in publications before the year 2008.
However, we find a strong increase in InSAR-related projects in the most recent Dragon-4 phase, during which the
majority of SAR-related projects also include at least an InSAR component in their research proposals.
In analyzing the results, we must understand that the
projects and their topics are determined when the proposal
is written. The content of the proposals for Dragon-1 was
decided in 2003, for Dragon-2 in 2007, for Dragon-3 in
2011, and for Dragon-4 in 2015. Although some projects
may have included or excluded SAR- or InSAR-related projects during the research periods, this may not be reflected
in our assessment.
Based on this, we see similarities to our bibliometric
analysis. In Dragon-1, we have a relatively strong InSAR component, which is reduced in Dragon-2. The absolute number
was identical, but no new teams joined the InSAR research.
This is, at first, counterintuitive to our hypothesis about an
increase in SAR- and InSAR-related research during the dataavailability phase that began in 2008. However, as the proposals for Dragon-2 were written before 2008, the proposed
research content in Dragon-2 belongs to the first phase. The
relative decrease in InSAR-related research is also similar to
the relative decrease in InSAR-related papers before 2008.
In Dragon-3, we see an increase in InSAR-related research, driven by increased research and applications
MARCH 2020

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

from the data-availability phase. Finally, we see a large
increase in InSAR-related projects in Dragon-4. Although
the overall proportion of SAR- and non-SAR-related projects remained at 50%, the majority of SAR-related projects
now also include an InSAR
component. We believe this
is related to the increase in
SAR DATA WILL BECOME
InSAR processing knowledge
UBIQUITOUS, GIVEN THE
and expertise.
VARIETY OF NEW AND

UPCOMING SENSORS AND

RESEARCH OUTLOOK
THE ENTRY OF COMMERCIAL
Most of today's standard methSTART-UPS INTO THE SAR
odologies and techniques for
DATA MARKET.
surface-motion estimation from
SAR were created in the first
phase, whereas the development of applications and their wide use occurred during the
second phase, mainly because data availability is a prerequisite for widespread application of the methods. We can
only speculate about the third phase, but the large amount
of data available and the current development of algorithms in machine learning and their need for such large
volumes of data indicate a trend in this direction.

1,500
1,000

China (Including Hong Kong)
United States and Canada
EU and Switzerland
Others

500
0

2004-2008

2009-2013

2014-2018

FIGURE 5. InSAR-related papers in the Web of Science
-database, with attributed funding information sorted by the
origin of the funding.

80
70
60
50
40
30
20
10
0

27%
20%

19%
31%
50%

11%
31%
58%

21%

25%
55%

52%

Dragon-1
Dragon-2
Dragon-3
Dragon-4
(2004-2008) (2008-2012) (2012-2016) (2016-2020)

FIGURE 6. The number of SAR-, InSAR-, and non-SAR-related
projects in the Dragon program. Blue indicates non-SAR related
projects, orange shows SAR-related projects (excluding InSAR),
and gray represents InSAR-related projects.

IEEE Geoscience and Remote Sensing Magazine - March 2020

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - March 2020