ArticlePDF Available

Comparison of Field-And Satellite-Based vegetation cover estimation methods

Authors:

Abstract and Figures

Background Monitoring terrestrial vegetation cover condition is important to evaluate its current condition and to identify potential vulnerabilities. Due to simplicity and low cost, point intercept method has been widely used in evaluating grassland surface and quantifying cover conditions. Field-based digital photography method is gaining popularity for the purpose of cover estimate, as it can reduce field time and enable additional analysis in the future. However, the caveats and uncertainty among field-based vegetation cover estimation methods is not well known, especially across a wide range of cover conditions. We compared cover estimates from point intercept and digital photography methods with varying sampling intensities (25, 49, and 100 points within an image), across 61 transects in typical steppe, forest steppe, and desert steppe in central Mongolia. We classified three photosynthetic groups of cover important to grassland ecosystem functioning: photosynthetic vegetation, non-photosynthetic vegetation, and bare soil. We also acquired normalized difference vegetation index from satellite image comparison with the field-based cover. ResultsPhotosynthetic vegetation estimates by point intercept method were correlated with normalized difference vegetation index, with improvement when non-photosynthetic vegetation was combined. For digital photography method, photosynthetic and non-photosynthetic vegetation estimates showed no correlation with normalized difference vegetation index, but combining of both showed moderate and significant correlation, which slightly increased with greater sampling intensity. Conclusions Results imply that varying greenness is playing an important role in classification accuracy confusion. We suggest adopting measures to reduce observer bias and better distinguishing greenness levels in combination with multispectral indices to improve estimates on dry matter.
Content may be subject to copyright.
R E S E A R C H Open Access
Comparison of field- and satellite-based
vegetation cover estimation methods
Dongwook W. Ko
1*
, Dasom Kim
1
, Amartuvshin Narantsetseg
2
and Sinkyu Kang
3
Abstract
Background: Monitoring terrestrial vegetation cover condition is important to evaluate its current condition and to
identify potential vulnerabilities. Due to simplicity and low cost, point intercept method has been widely used in
evaluating grassland surface and quantifying cover conditions. Field-based digital photography method is gaining
popularity for the purpose of cover estimate, as it can reduce field time and enable additional analysis in the future.
However, the caveats and uncertainty among field-based vegetation cover estimation methods is not well known,
especially across a wide range of cover conditions. We compared cover estimates from point intercept and digital
photography methods with varying sampling intensities (25, 49, and 100 points within an image), across 61
transects in typical steppe, forest steppe, and desert steppe in central Mongolia. We classified three photosynthetic
groups of cover important to grassland ecosystem functioning: photosynthetic vegetation, non-photosynthetic
vegetation, and bare soil. We also acquired normalized difference vegetation index from satellite image comparison
with the field-based cover.
Results: Photosynthetic vegetation estimates by point intercept method were correlated with normalized
difference vegetation index, with improvement when non-photosynthetic vegetation was combined. For digital
photography method, photosynthetic and non-photosynthetic vegetation estimates showed no correlation with
normalized difference vegetation index, but combining of both showed moderate and significant correlation, which
slightly increased with greater sampling intensity.
Conclusions: Results imply that varying greenness is playing an important role in classification accuracy confusion.
We suggest adopting measures to reduce observer bias and better distinguishing greenness levels in combination
with multispectral indices to improve estimates on dry matter.
Keywords: Point intercept, Digital photography, Land cover estimate, NDVI, Photosynthetic vegetation, Greenness
Background
Defined as land covered with herbaceous plants with less
than 10% trees and shrub cover(White et al. 2000), grass-
land is an important ecosystem which dominates much of
the global terrestrial ecosystem. Grasslands provide a wide
range of ecological services across the world, including
critical resources for nomadic livelihood, biodiversity, car-
bon storage, water and nutrient cycling, and soil erosion
protection (Mosier et al. 1991, White et al. 2000).
Unfortunately, grassland has been subject to large-
scale degradation, causing serious ecological and social
problems in various geographical regions. In northeast
Asia, for example, serious social and ecological damages
are caused by intensive yellow dust phenomena, which
frequently originates from the degraded grassland in
Mongolia (Phadnis and Carmichael 2000, In and Park
2002). In Mongolia, overgrazing and drought conditions
have played a significant role in grassland degradation
(McCarthy 2001). Such conditions make the grassland
ecosystem more vulnerable to dzud, which occurs in the
severe winter following a suboptimal growing season,
resulting in high level of livestock mortality and social
instability (Fernández-Giménez et al. 2015).
Considering the importance of grassland ecosystem to
a wide range of human societies and ecological systems,
accurate monitoring of vegetation status in grasslands is
important. In representing grassland condition, quantify-
ing land cover type is one of the most widely used
* Correspondence: dwko@kookmin.ac.kr
1
Department of Forest Environmental System, Kookmin University, Seoul,
Republic of Korea
Full list of author information is available at the end of the article
Journal of Ecolog
y
a
n
d
Envir
o
nm
e
n
t
© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Ko et al. Journal of Ecology and Environment (2017) 41:5
DOI 10.1186/s41610-016-0022-z
indicators, through classifying and proportioning the
land cover in several categories such as vegetation, bare
ground, and water (Meyer and Turner 1994). One of the
most popular field methods to quantify land cover type
is the point interception method, due to its simplicity,
unbiasedness, and low cost (Canfield 1941, Ramsey
1979, Sutherland 2006). Point interception method is an
extremely efficient method, but the method is known to
underestimate cover types with uneven or patchy distri-
bution (Buckland et al. 2007). Another alternative is the
quadrats method, which estimates percent ground cover
within a quadrat divided into several cover classes, or
cover abundance scores, such as a Braun-Blanquet
method (Daubenmire 1959). However, the estimated land
cover is often known to be dependent upon the method
utilized, because of their sensitivity to plant size, growth
form, and crown density (Floyd and Anderson 1987).
Analyzing digital images acquired from the field can
be advantageous since the production of permanent im-
ages enables the researcher to reanalyze the data later on
with more advanced methods and softwares (Boyd and
Svejcar 2005). This method can be particularly helpful
since it can drastically reduce time spent in the field and
control surveyor-bias (Booth et al. 2005). A study based
on turf-grass dominated sites indicated that compared
with line intersect, digital photography analysis was able
to generate accurate results in much less time (Richardson
et al. 2001). Another study which compared digital photo
analysis and point intercept method also suggested
that the results between the two methods were simi-
lar when sufficient number of plots were combined
together (Booth et al. 2005).
Remote sensing technology is highly useful for system-
atic and long-term vegetation cover monitoring (Iverson
et al. 1989, Gemmell 1999, Turner et al. 2007, Yim et al.
2010). Through the use of various vegetation indices
(e.g., normalized difference vegetation index (NDVI)), it
can represent vegetation condition based on analyzing
the spectral characteristics of the grassland (Cui et al.
2011). However, the method still requires ground truth
data to validate its results. Moreover, studies have shown
that establishing consistent guidelines and understand-
ing the properties of field survey methods is critical in
improving proper integration of remotely sensed and
field-based data (Reinke et al. 2006). Considering sam-
pling density for the spatial scale of interest, and spatial
heterogeneity of the target vegetation type are some of
the important aspects to consider in deciding which
survey method to use.
Meanwhile, greenness (photosynthetically active com-
ponent) is not the only important factor in grassland
ecosystems. Due to the wide range of seasonal condi-
tions of temperature, precipitation, wind, fire, grazing,
and human management of rangeland, the amount of
dry matter on grasslands can provide valuable information
(Bradley and Mustard 2005, Guerschman et al. 2009). For
example, carbon and nutrient cycling, surface reflection,
soil erosion, land degradation, and phonology assessment
can all benefit from information on non-photosynthetic
biomass, which makes dry matter estimation a very im-
portant factor in evaluating the condition of grassland
ecosystem (Byambakhuu et al. 2010, Stoner et al. 2016).
In light of the information above, what should
researchers expect from a variety of field methods and
survey target components in grasslands under various
conditions? To answer this question, we compared per-
formances of three different grassland cover estimation
methods: field point intercept, digital photography analysis
with varying sampling densities, and NDVI acquired from
a remote sensing platform (MODIS) were compared.
We compared results for three ecosystem functional
components in grassland: photosynthetic vegetation,
non-photosynthetic vegetation, and bare soil. We also
explored the effect of various grassland conditions, by
comparing results across typical steppe, desert steppe,
and forest steppe in central Mongolia.
Methods
Study area
In this study, sites are distributed across the central part
of Mongolia, between the latitude of 48° 480.72to 45°
413.12N and longitude of 96° 5056.88to 105° 50
45.06E, within Tov, Arkhangai, Zavkhan, Bayankhon-
gor, and Ovorkhangai aimag (Fig. 1, Table 1). Central
Mongolian steppe zone can be roughly grouped into
typical steppe, forest steppe, desertified steppe, and des-
ert steppe based on the botanico-geographical groups
(Karamysheva and Khramtsov 1995). In this study, based
on the location and environment of the sites, three dif-
ferent steppe types were covered by the site locations:
forest steppe, typical steppe, and desert steppe. Central
Mongolian steppe is characterized by high elevation
(10431350 m) and sandy loam with abundant gravels,
with higher number of livestock compared to the eastern
Mongolian steppe (Hirobe and Kondo 2012). The central
Mongolian steppe covers a range of vegetation character-
istics and environmental conditions, dominated by various
grass and sedges (Stipa spp. and Achnatherum spp.), pea-
shrubs (Caragana spp.), and sages (Artemisia spp.).
Field survey
Survey for field data collection was conducted from July
16 to 25, 2013. A total of 61 transects across 28 sites
were surveyed (Table 1). All sites were chosen so that
none were close to a major road and at least several
hundred meters away from minor paths. Each site in-
cluded one to three line transects. Two methods for field
survey were adopted: (1) point intercept and (2) digital
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 2 of 11
photography. For both field surveys, one to three 30-m
parallel line transects were established, at least 30 m
away from each other. Five survey crew carried out the
survey, who were ecologists and botanists who had at
least graduate-level training.
For point intercept cover estimate, cover type was re-
corded at points at 1-m interval (30 points per transect
whenever terrain allows). For digital photography estimate,
digital photos of the surface was taken at 3-m intervals
(Samsung ES95, 16.1MP), total of ten photographs per
transect. Photos were taken at 1.2 m height, with the
photograph plane parallel to ground surface. Later, digital
images were cropped to cover approximately 2 m
2
(164 ×
123 cm) of the surface area with the final resolution of
2304 × 1728 pixels. A total of 1529 points and 549 digital
photographs were collected.
Fig. 1 Map of Mongolia and the location of field survey sites
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 3 of 11
For both point intercept cover estimate, six major cat-
egories were initially used in the field to record plant
functional types as cover types at each point: grass, forb,
shrub, litter, bare soil, and rock. For the final analysis,
cover types were reclassified into functional groups
based on photosynthetic properties considering their im-
portance in representing a wide range of grassland con-
ditions (sensu Guerschmann et al. 2009). For comparing
cover estimates among field-based methods, photosyn-
thetic vegetation (PV) cover was estimated by combining
grass, forb, and half of shrub, and non-photosynthetic
vegetation (NPV) was estimated by combining litter and
half of shrub. Shrub cover was equally assigned to PV
and NPV, considering that it is a mix of photosynthetic
leaf part and non-photosynthetic woody part. Bare soil
and rock were combined to bare soil (BS). Potential
photosynthetic vegetation (PPV) was also calculated as
the sum of PV and NPV. Since remainder of PPV is BS,
we only estimated and analyzed the cover of PV, NPV,
and PPV based on the total point frequency assigned to
the corresponding cover types.
For the digital photography cover estimate, photo-
graphs were analyzed using the SamplePointsoftware,
which assists classifying individual pixels within a photo-
graph (Booth et al. 2006). We generated regularly
distributed crosshairs over each photographs to classify
the overlapping single pixels (Fig. 2). To consider the ef-
fects of sampling intensity in digital photograph method,
we used a variety of sampling intensities, by generating
25, 49, and 100 regularly spaced crosshairs in each
photography (5 × 5, 7 × 7, and 10 × 10 sampling points,
hereafter mentioned as SP25, SP49, and SP100, respect-
ively). Each pixel under the crosshair was classified into
six cover types following the same classification scheme
used for the field point intercept method and then was
reclassified into PV, NPV, and BS. PPV was also
Table 1 Site description and location of plots
Site Aimag Number of transects Latitude Longitude Eco-region
TO01 Tov 1 47° 52' 40" 105° 50' 45" Steppe
AR00 Arkhangai 2 47° 15' 43" 103° 19' 46" Forest steppe
OV01 Arkhangai 2 47° 15' 47" 103° 33' 47" Steppe
OV02A Ovorkhangai 3 46° 15' 17" 102° 47' 40" Steppe
OV02B Ovorkhangai 3 46° 15' 00" 102° 47' 00" Steppe
OV03A Ovorkhangai 3 46° 14' 15" 102° 49' 06" Steppe
OV04 Ovorkhangai 2 45° 48' 05" 101° 53' 40" Desert steppe
OV05 Ovorkhangai 2 45° 41' 42" 101° 40' 05" Desert steppe
OV06 Ovorkhangai 3 45° 41' 03" 101° 36' 37" Desert steppe
BA01 Bayankhongor 1 46° 13' 40" 100° 36' 37" Steppe
BA03A Bayankhongor 3 46°43'28" 99°49'21" Steppe
BA03B Bayankhongor 2 46° 43' 00" 99° 49' 00" Steppe
ZA03 Zavkhan 2 47° 10' 26" 97° 16' 59" Steppe
ZA04 Zavkhan 2 47° 19' 37" 96° 58' 20" Steppe
ZA05 Zavkhan 1 47° 22' 38" 96° 55' 37" Forest steppe
UL01 Zavkhan 2 47° 43' 03" 96° 50' 56" Forest steppe
TE01 Zavkhan 3 48° 48' 00" 97° 30' 32" Forest steppe
AR01A Arkhangai 3 47° 56' 25" 100° 38' 29" Forest steppe
AR01B Arkhangai 3 47° 56' 00" 100° 38' 00" Forest steppe
AR02 Arkhangai 2 47° 37' 50" 101° 04' 36" Forest steppe
AR03A Arkhangai 3 47°33'36" 101°00'10" Forest steppe
AR03B Arkhangai 3 47° 33' 00" 101° 00' 00" Forest steppe
AR04A Arkhangai 1 47° 27' 58" 101° 28' 53" Forest steppe
AR04B Arkhangai 2 47° 27' 00" 101° 29' 00" Forest steppe
AR06 Arkhangai 2 47° 29' 27" 102° 09' 05" Steppe
AR07 Arkhangai 2 47° 32' 12" 102° 14' 27" Steppe
AR08A Arkhangai 3 47° 49' 50" 102° 55' 37" Steppe
AR08B Arkhangai 3 47° 49' 00" 102° 55' 00" Steppe
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 4 of 11
calculated as the sum of PV and NPV. To ensure
consistency in classification, we adopted the following
measures: (1) all classifications were made on the same
computer-monitor set to maintain the visual characteris-
tics of the images and (2) before actual classification, ob-
servers spent 2 h together training on the same images.
Satellite imagery and NDVI estimates
In arid or semi-arid region, normalized difference
vegetation index (NDVI) is often used for the estima-
tion of green vegetation cover (Pickup et al. 1993,
Chen et al. 2006) and productivity (Chen et al. 2004,
Wang et al. 2004). NDVI is based on the spectral
properties generated by photosynthetic process: it
compares the ratio between visible red light, which is
strongly absorbed, and near-infrared, which is strongly
reflected by green vegetation. A variety of satellite
platforms provides spectral information to calculate
NDVI.Inthisstudy,weusedMODIS(moderatereso-
lution imaging spectroradiometer) products for its
reliability of image acquirement in Mongolia espe-
cially considering the non-optimal sky conditions that
frequents the growing season (Jang et al. 2010). Spe-
cifically, the 16-day composite products (MOD13Q1,
250 m resolution) based on MODIS Level-2G (daily)
surface reflectance data with the acquisition date of
July 1227 (2013) were acquired for the NDVI value.
NDVI from the corresponding pixel(s) for the study
sites was extracted from the images and then was
compared with point intercept and digital photog-
raphy cover estimates. For comparing NDVI and
field-based surveys, the field cover estimates of PV,
NPV, and PPV were used to evaluate how different
surface components, especially the dry matter, influ-
enced the fit against the spectral properties of NDVI
(Asner 1998, Booth et al. 2006).
Fig. 2 Example of crosshairs generated by SamplePoint software for estimating cover by field-based digital photography method
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 5 of 11
Statistical analysis
Cover estimates from point intercept and digital photog-
raphy method with varying intensity (SP25, SP49, and
SP100) were compared by calculating summary statistics
(mean, median, minimum, maximum, range, standard
deviation (SD), and coefficient of variation (CV)) and
conducting correlation analysis. Analysis of variance and
Tukeys HSD test was also conducted to analyze differ-
ences between cover estimates of each method. For fur-
ther insight, field estimates were grouped by steppe
types for comparisons as well. NDVI estimates were
compared among steppe types by calculating summary
statistics, analysis of variance, and Tukeys HSD test. For
comparison of field-based estimates against MODIS-
NDVI, correlation analysis was conducted. All statistical
analyses were conducted with R (version 3.2.1).
Results and discussion
Comparison of field-based cover estimatespoint
intercept and digital photography
Both field-based mean and median estimates of PV,
NPV, and BS did not differ among field-based methods,
even when steppe types were considered (α= 0.01, Fig. 3,
Table 2). Considering PV estimates, differences were
only noticeable in desert steppe, but not significant
(α= 0.01, Table 2, Fig. 3a). Median of PV cover estimate
in desert steppe was 17.8, 29.6, 26.1, and 23.9% for point
intercept, SP25, SP49, and SP100, respectively, and range
Fig. 3 Box plots of acPV cover estimates in each steppe type based on field-based methods, and dePV, NPV, and bare soil cover estimates of all
sites combined. Thick black line within the rectangle indicates median, rectangles are the interquartile range (IQR), and whiskers indicate 1.5 times IQR
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 6 of 11
was smaller in point intercept method (15.0%) compared
to digital photograph methods (32.233.3%) (Table 2).
This difference likely resulted from the sparse vegetation
conditions in desert steppe, which can penalize cover esti-
mates based on smaller sampling density (Milberg et al.
2008). This is a common issue, particularly for estimating
abundance of rare components in environments with high
level of spatial variability (Bergstedt et al. 2009, Burg et al.
2015). Overall, digital photography method seems to
retain its consistency regardless of sampling density across
most statistics in PV cover estimates.
Mean NPV cover estimates did not differ significantly
among field-based methods (α= 0.01, Table 2, Fig. 3b).
While most statistics were similar to each other in
typical steppe and forest steppe of NPV, the wider range
of point intercept method in forest-steppe sites with
extremely patchy or heterogeneous patterns of cover
conditions is noticeable. For example, site UL01 had very
Table 2 Summary statistics of PV, NPV, and BS cover estimates of steppe types and field-based methods
Cover type Steppe type Method Mean Median Min Max Range SD CV
PV Typical steppe PI 35.0 35.6 11.1 63.3 52.2 15.4 0.440
SP25 34.4 34.6 9.5 58.0 48.5 13.5 0.392
SP49 33.9 31.2 7.3 58.6 51.3 13.6 0.401
SP100 33.4 30.6 7.7 57.3 49.6 13.7 0.410
Forest steppe PI 40.3 43.3 15.6 73.3 57.8 18.1 0.449
SP25 39.9 41.6 21.6 56.8 35.2 10.1 0.253
SP49 38.5 40.0 18.4 57.6 39.2 11.6 0.301
SP100 37.5 37.7 18.1 62.9 44.8 12.9 0.344
Desert steppe PI 22.0 17.8 16.7 31.7 15.0 8.4 0.382
SP25 36.6 29.6 23.5 56.8 33.3 17.7 0.484
SP49 34.7 26.1 22.5 55.5 33.0 18.1 0.522
SP100 34.0 23.9 22.9 55.1 32.2 18.3 0.538
NPV Typical steppe PI 28.5 23.4 10.6 80.0 69.4 18.8 0.660
SP25 23.9 21.0 2.3 66.8 64.5 18.2 0.759
SP49 23.7 22.8 1.7 60.8 59.1 17.1 0.723
SP100 24.5 23.1 2.0 60.9 58.9 17.2 0.705
Forest steppe PI 28.2 30.0 0.0 70.0 70.0 23.5 0.833
SP25 21.9 20.3 7.2 39.2 32.0 11.4 0.521
SP49 23.1 19.9 5.1 44.7 39.6 14.0 0.607
SP100 24.5 24.6 4.4 45.4 41.0 13.8 0.563
Desert steppe PI 12.0 14.4 6.7 15.0 8.3 4.6 0.385
SP25 5.2 4.4 4.0 7.2 3.2 1.7 0.335
SP49 5.4 4.5 2.2 9.6 7.4 3.8 0.698
SP100 6.6 3.5 2.3 14.0 11.7 6.4 0.976
BS Typical Steppe PI 36.5 34.4 3.3 60.0 56.7 16.0 0.439
SP25 41.7 38.6 11.2 69.3 58.1 17.6 0.421
SP49 42.3 38.9 13.5 70.0 56.5 16.7 0.394
SP100 42.1 39.0 14.0 69.8 55.8 16.3 0.388
Forest steppe PI 31.5 27.7 8.9 80.0 71.1 19.9 0.632
SP25 38.2 39.2 21.5 59.6 38.1 11.6 0.303
SP49 38.4 37.3 22.1 56.7 34.6 11.0 0.286
SP100 38.0 35.1 23.1 54.6 31.5 10.6 0.280
Desert steppe PI 65.9 67.8 53.3 76.6 23.3 11.8 0.179
SP25 58.2 63.2 38.8 72.5 33.7 17.4 0.299
SP49 59.9 64.3 42.3 73.0 30.7 15.8 0.264
SP100 59.4 62.1 42.6 73.6 31.0 15.7 0.264
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 7 of 11
high bare soil cover but moderate PV cover, and none of
the points was able to capture the NPV present at site.
Meanwhile, in site AR04B, dominated by Alium spp.
with extremely fine leaves in both green and desiccated
state, field observations performed poorly in capturing
the fine-textured green cover and classified most of
them as NPV (70%), while detailed observation of digital
photograph method was capable of capturing the nu-
anced greenness and classifying them as PV (5763%)
rather than NPV (414%).
In classifying BS, most estimates showed high agree-
ment, and mean cover estimates were not significantly
different from each other (α= 0.01, Table 2, Fig. 3c). It
seems that point intercept method had a relatively high
level of confusion in forest-steppe types as suggested by
its large range (71%), which occurred in sites with very
low or high level of NPV with finely textured vegetation
(UL01 and AR01A).
When all sites were combined and compared, mean
cover estimates of PV, NPV, and BS did not show signifi-
cant differences among methods (Fig. 3df). Ranges from
point intercept method was relatively larger compared to
digital photography methods, suggesting that digital pho-
tography methods may ensure better consistency in cover
estimates. A number of studies pointed out the issue of
over- or underestimation of grassland cover from field
methods (Dethier et al. 1993, Fensholt et al. 2004). Dethier
et al. (1993) pointed out that point intercept method is
subject to over-estimating cover compared to photo
analysis, which is inherently constrained by the smaller
number of sample points at field. However, this did
not consistently apply to PV or NPV estimates in our
study in any of the steppe types. Moreover, PV esti-
mates from point intercept method in desert-steppe
sites was slightly higher than estimates from digital
photography method (Fig. 3).
Correlation analysis of PV, NPV, and PPV cover estimates
showed that digital photography methods were highly and
significantly correlated with each other (Fig. 4ac). How-
ever, PV estimates of point intercept method were not cor-
related with estimates of any of the digital photography
methods (Fig. 4a). Interestingly, NPV estimates of point
intercept method were significantly correlated, although
moderately, with estimates of all digital photography
methods (Fig. 4b). These suggest that compared to NPV
classification, there is a higher level of disagreement in PV
classification between point intercept and digital photog-
raphy methods. This mismatch was dramatically reduced
Fig. 4 Scatterplot matrix and correlation coefficients of estimated
covers between cover types (a-cPV, NPV, and PPV) and field-based
methods. Lower left panels show scatterplots, and upper right panels
show correlation coefficients and significance levels (*p= 0.01;
**p< 0.001; ***p=0)
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 8 of 11
when correlation of PPV among field methods are consid-
ered, since estimates of PPV cover from all field methods
showed significantly high level of correlation (Fig. 4c). Such
confusion of grassland classification may be due to the leaf
angle distribution of dominant grass and sedge species, ac-
cumulating dry matter, or the overwhelming bare soil in
the background (Beck et al. 1990, Guerschman et al. 2009).
Comparison of field-based cover estimates and NDVI
NDVI estimates ranged from 0.14 to 0.50, with most
sites showing NDVI values between 0.2 and 0.4 (Table 3).
Forest- and typical-steppe NDVI estimates were signifi-
cantly higher than the desert-steppe NDVI (p< 0.01,
Table 3). Correlation analysis results indicate that PV
estimates of point intercept method had moderate and
significant correlation with NDVI (Fig. 4a). However,
none of the PV estimates from digital photography
methods were correlated with NDVI, and none of the
field-based NPV estimates were correlated with NDVI
(Fig. 4b).
When cover estimates of PPV (combination of PV and
NPV) were considered, all methods had moderate and
significant correlation with NDVI (p< 0.01, Fig. 4c). In
addition, for the point intercept method, correlation of
PPV estimates with NDVI slightly improved, compared
to the correlation of PV estimate with NDVI (Fig. 4a, c).
Results suggest that part of the NDVI-related spectral
signal can be traced back to PPV component identified
in the field-based methods.
Correlation between field methods and NDVI suggest
some important insights. General consensus among ob-
servers was that classifications of point intercept method
in the field were subject to greater confusion, because of
various field conditions experienced, such as time of day,
weather, livestock trampling and droppings, and obser-
ver conditions. However, results show that detailed
classification of digital photography method may have
underestimated PV component by limiting it to distinct-
ively green vegetation, while excluding partly green vege-
tation, and NPV included vegetation with a wide range
of greenness, from slightly desiccated plant materials to
very dry materials. In contrast, it is possible that point
intercept method included widely varying green compo-
nents, which was likely the reason why PV estimate
alone had significant and moderately high level of correl-
ation with NDVI. Theoretically, NDVI is intended to
represent the photosynthetically active component in
the image by considering its characteristics in differential
absorption of red and near-infrared spectrum (Beck
et al. 1990). However, grassland ecosystems are subject
to a widely varying condition of greenness as influenced
by phenology and inter- and intra-annual variability
(Bradley and Mustard 2005). Therefore, identifying the
varying condition and abundance of green and dry mat-
ter is very important for evaluating the functional condi-
tion of grasslands which often cannot be decisively
classified as green or dry. For this reason, researchers
suggest that the use of cellulose absorption index (CAI)
in addition to vegetation indices (e.g., NDVI) for improv-
ing and enriching how surface vegetation components is
represented (Guerschman et al. 2009). Our study con-
firms that point intercept method at field alone can be
prone to errors since PV and NPV could not be easily
distinguished, and varying levels of greenness were likely
classified as PV. Therefore, evaluating how accurately
any of the field methods can distinguish dry matter is
subject to further studies by utilizing multispectral im-
ages to compare with the estimated CAI and ideally with
field method that can differentiate a more detailed grade
of greenness.
Moreover, scale and resolution of the data can be im-
portant factors when characterizing surface properties
(Turner et al. 1989). Although we carefully considered
homogeneity of surface properties as site selection
criteria, a single pixel from MODIS imagery (250 m
resolution) covered a spatial extent that is significantly
larger compared to field survey samples, including
greater heterogeneity to be represented in the NDVI
than we expected. This probably contributed to weaker
fit with the field-based cover estimates (Moody and
Woodcock 1995). Time scale can also be important,
considering the short growing season and the rapidly
shifting phenology of Mongolian grasslands (Boone et al.
2005). Since the 16-day composite NDVI was selected
for this study, such a time frame may include time-
driven phonological difference that may have influenced
the results.
Another important factor that may have confounded
the results is the observers bias. Two types of observers
bias are acknowledged: (1) bias caused by less-trained ob-
server, which can amplify within-observer variation and
(2) bias caused by inter-observer differences (Dethier et al.
1993, Bergstedt et al. 2009). For the field-based point
intercept method, in particular, we noted that mis-
recognition could be a significant factor, even with train-
ing sessions before each survey. This is especially true for
spatially heterogeneous surface or sites with very sparse
green vegetation against extensive bare soil background
during mid-day. Ultimately, bias-control is another factor
that a researcher must carefully examine as trade-offs
Table 3 Summary statistics and comparisons of NDVI of each
steppe types (α= 0.01)
Steppe type Mean Median Min Max Range SD CV
Typical steppe 0.33 (a) 0.30 0.24 0.50 0.26 0.08 0.24
Forest steppe 0.35 (a) 0.34 0.30 0.44 0.14 0.05 0.14
Desert steppe 0.15 (b) 0.15 0.14 0.16 0.02 0.01 0.07
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 9 of 11
among survey methods, in terms of time and labor cost,
scale-dependency, and vegetation characteristics. We sug-
gest a variety of measures to control the bias of observers
such as field training and consistency check among ob-
servers before the actual survey begins and for digital pho-
tography methods, ensuring the quality and consistency of
color and texture representation of visual devices (use
identical graphic card and monitor, with color specifica-
tions and resolution matched), co-training session for ob-
servers to ensure classification consistency, and diversified
greenness-levels for classification.
In our study, we suspect that the mismatch between
point intercept and digital photography method is a re-
sult of how varying range of greenness was treated, as
PPV of both methods showed meaningful relationship
with NDVI. Aside from this issue, it is notable that
greater sampling density in digital photography methods
of PPV estimates showed a slightly better fit against
NDVI (Fig. 4c). Fit of SP100 and SP49 was slightly better
compared to SP25, and there was no difference between
SP100 and SP49. Considering the trade-offs between
effort and performance, we propose that SP49 has the
potential advantage among all methods, given that the
confusion of greenness is alleviated, because of the
additional virtues of digital photography method:
quicker application in the field compared to point
intercept method, possibility of re-analysis in the fu-
ture when the need arises, and control over observers
bias (Floyd and Anderson 1987, Boyd and Svejcar
2005, Booth et al. 2006).
Our study suggests the potential of digital photography
method to estimate vegetation cover. Digital photog-
raphy method showed potential for application to other
terrestrial ecosystems, such as forest or wetlands, and to
evaluate rapid changes due to disturbances such as
drought, grazing, or fire. This is especially relevant with
the advent of unmanned aerial vehicle (UAV)-based
digital photography for surveying and monitoring terres-
trial ecosystems that can further reduce field cost and time
(Rango et al. 2009, Cunliffe et al. 2016). While there are
numerous sophisticated methods to analyze massive
amount of images acquired from UAVs (Hervouet et al.
2011, Bollard-Breen et al. 2015), simple analysis of UAV-
acquired images that share common grounds with trad-
itional field survey methods, such as point intercept
method, proves to be useful in reconstructing and evaluat-
ing landscape change in the past and future.
Conclusions
Our study showed that when both PV and NPV were
combined, estimates from point intercept method (simul-
taneous field data and classification) and digital photog-
raphy method (photo taken from field and classification
later in the lab) showed moderate agreement against the
satellite-derived NDVI (R
2
=0.43to0.48, p= <0.01). Point
intercept method was more inclusive for a wider range of
greenness compared to digital photography method.
Greater sampling intensity of digital photography method
slightly increased its agreement with NDVI; therefore,
considering the efforts required, we suggest 49 points over
2m
2
is sufficient. Our study confirms the merits of point
intercept method, the simplicity and low cost, but also
suggests the potential of the digital photography
method because of the possibility of future re-analysis.
Wesuggestthatincorporatingamoreexplicitclassifi-
cation scheme to differentiate greenness may improve
cover estimation results for both point intercept and
digital photography methods.
Abbreviations
BS: Bare soil; CAI: Cellulose absorption index; MODIS: Moderate resolution
imaging spectroradiometer; NDVI: Normalized difference vegetation index;
NPV: Non-photosynthetic vegetation; PPV: Potential photosynthetic
vegetation; PV: Photosynthetic vegetation; UAV: Unmanned aerial vehicle
Acknowledgements
This work was supported by the research grants from the Korea Forest
Service (S121414L090110) and the National Research Foundation of Korea
(NRF-201100009423). The authors wish to thank Dowon Lee, Reverend
Sungil, Jaebum Kim, and Wanhyuk Park for their assistance in field survey
and laboratory work. The field survey was completed under the permission
of the Mongolian Academy of Sciences.
Funding
This work was supported by the research grants from the Korea Forest
Service (S121414L090110) and the National Research Foundation of Korea
(NRF-201100009423). The role of the fund from Korea Forest Service was in
remote sensing data analysis and manuscript preparation. The role of the
fund from the National Research Foundation of Korea was in the field data
collection and initial analysis.
Availability of data materials
The data that support the findings of this study are available from the
corresponding author (DWK) upon reasonable request. The data are not
publicly available due to sensitive information regarding surface
information of the study area.
Authorscontributions
DWK conceived of the study, led its design and coordination, drafted the
manuscript, collected and analyzed data and prepared results. DK participated
in data collection, classification and remote sensing analysis. AN designed and
guided the field survey and analysis methods. SK took part in the analysis and
drafting the discussions. All authors read and approved the final manuscript.
Competing interests
The authors declare that they have no competing interests.
Consent for publication
Not applicable.
Ethics approval and consent to participate
Not applicable.
Author details
1
Department of Forest Environmental System, Kookmin University, Seoul,
Republic of Korea.
2
Institute of General and Experimental Biology, Mongolian
Academy of Sciences, Ulaanbaatar, Mongolia.
3
Department of Environmental
Science, Kangwon National University, Kangwon, Republic of Korea.
Received: 7 October 2016 Accepted: 6 December 2016
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 10 of 11
References
Asner, G. P. (1998). Biophysical and biochemical sources of variability in canopy
reflectance. Remote Sensing of Environment, 64, 234253.
Beck, L. R., Hutchinson, C. F., & Zauderer, J. (1990). A comparison of greenness
measures in two semi-arid grasslands. Climatic Change, 17, 287303.
Bergstedt, J., Westerberg, L., & Milberg, P. (2009). In the eye of the beholder: bias
and stochastic variation in cover estimates. Plant Ecology, 204, 271283.
Bollard-Breen, B., Brooks, J. D., Jones, M. R. L., Robertson, J., Betschart, S., Kung, O.,
Craig Cary, S., Lee, C. K., & Pointing, S. B. (2015). Application of an unmanned
aerial vehicle in spatial mapping of terrestrial biology and human disturbance in
the McMurdo Dry Valleys, East Antarctica. Polar Biology, 38,573578.
Boone, R. B., BurnSilver, S. B., Thornton, P. K., Worden, J. S., & Galvin, K. A. (2005).
Quantifying declines in livestock due to land subdivision. Rangeland Ecology
& Management, 58, 523532.
Booth, D. T., Cox, S. E., Fifield, C., Phillips, M., & Williamson, N. (2005). Image
analysis compared with other methods for measuring ground cover. Arid
Land Research Management, 19,91100.
Booth, D. T., Cox, S. E., & Berryman, R. D. (2006). Point sampling digital imagery
with Samplepoint..Environmental Monitoring and Assessment, 123,97108.
Boyd, C. S., & Svejcar, T. J. (2005). A visual obstruction technique for photo
monitoring of willow clumps. Rangeland Ecology & Management, 58,434438.
Bradley, B. A., & Mustard, J. F. (2005). Identifying land cover variability distinct
from land cover change: cheatgrass in the Great Basin. Remote Sensing of
Environment, 94, 204213.
Buckland, S. T., Borchers, D. L., Johnston, A., Henrys, P. A., & Marques, T. A. (2007).
Line transect methods for plant surveys. Biometrics, 63, 989998.
Burg, S., Rixen, C., Stöckli, V., & Wipf, S. (2015). Observation bias and its causes in
botanical surveys on high-alpine summits. Journal of Vegetation Science, 26,
191200.
Byambakhuu, I., Sugita, M., & Matsushima, D. (2010). Remote sensing of
environment spectral unmixing model to assess land cover fractions in
Mongolian steppe regions. Remote Sensing of Environment, 114, 23612372.
Canfield, R. H. (1941). Application of the line interception method in sampling
range vegetation. Journal of Forestry, 39, 388394.
Chen, Z. M., Babiker, I. S., Chen, Z. X., Komaki, K., Mohamed, M. A. A., & Kato, K. (2004).
Estimation of interannual variation in productivity of global vegetation using
NDVI data. International Journal of Remote Sensing, 25, 31393159.
Chen, X.-L., Zhao, H.-M., Li, P.-X., & Yin, Z.-Y. (2006). Remote sensing image-based
analysis of the relationship between urban heat island and land use/cover
changes. Remote Sensing of Environment, 104, 133146.
Cui, G., Lee, W.-K., Kwak, D.-A., Choi, S., Park, T., & Lee, J. (2011). Desertification
monitoring by LANDSAT TM satellite imagery. Forest Science and Technology,
7, 110116.
Cunliffe, A. M., Brazier, R. E., & Anderson, K. (2016). Ultra-fine grain landscape-scale
quantification of dryland vegetation structure with drone-acquired structure-
from-motion photogrammetry. Remote Sensing of Environment, 183, 129143.
Daubenmire, R. (1959). A canopy-coverage method of vegetational analysis.
Northwest Science, 33,4364.
Dethier, M. N., Graham, E. S., Cohen, S., & Tear, L. M. (1993). Visual versus random-
point percent cover estimations: objectiveis not always better. Marine
Ecology Progress Series, 96,93100.
Fensholt, R., Sandholt, I., & Rasmussen, M. S. (2004). Evaluation of MODIS LAI,
fAPAR and the relation between fAPAR and NDVI in a semi-arid
environment using in situ measurements. Remote Sensing of Environment,
91, 490507.
Fernández-Giménez, M. E., Batkhishig, B., Batbuyan, B., & Ulambayar, T. (2015).
Lessons from the dzud: community-based rangeland management increases
the adaptive capacity of Mongolian herders to winter disasters. World
Development, 68,4865.
Floyd, D. A., & Anderson, J. E. (1987). A comparison of three methods for
estimating plant cover. Journal of Ecology, 75, 221228.
Gemmell, F. (1999). Estimating conifer forest cover with Thematic Mapper data using
reflectance model inversion and two spectral indices in a site with variable
background characteristics. Remote Sensing of Environment, 69,105121.
Guerschman, J. P., Hill, M. J., Renzullo, L. J., Barrett, D. J., Marks, A. S., & Botha, E. J.
(2009). Estimating fractional cover of photosynthetic vegetation, non-
photosynthetic vegetation and bare soil in the Australian tropical savanna
region upscaling the EO-1 Hyperion and MODIS sensors. Remote Sensing of
Environment, 113, 928945.
Hervouet, A., Dunford, R., Piégay, H., Belletti, B., & Trémélo, M.-L. (2011). Analysis of
post-flood recruitment patterns in braided-channel rivers at multiple scales
based on an image series collected by unmanned aerial vehicles, ultra-light
aerial vehicles, and satellites. GIScience & Remote Sensing, 48,5073.
Hirobe, M., & Kondo, J. (2012). Effects of climate and grazing on surface soil in
grassland. In N. Yamamura, N. Fujita, & A. Maekawa (Eds.), The Mongolian
Ecosystem Network: Environmental Issues Under Climate and Social Changes
(pp. 105114). Japan: Springer.
In, H.-J., & Park, S.-U. (2002). A simulation of long-range transport of Yellow Sand
observed in April 1998 in Korea. Atmospheric Environment, 36, 41734187.
Iverson, L. R., Cook, E. A., & Graham, R. L. (1989). A technique for extrapolating
and validating forest cover across large regions calibrating AVHRR data with
TM data. International Journal of Remote Sensing, 10, 18051812.
Jang, K., Kang, S., Kim, J., Lee, C. B., Kim, T., Kim, J., Hirata, R., & Saigusa, N. (2010).
Mapping evapotranspiration using MODIS and MM5 four-dimensional data
assimilation. Remote Sensing of Environment, 114, 657673.
Karamysheva, Z. V., & Khramtsov, V. N. (1995). The steppes of Mongolia. Braun-
Blanquetia, 17,579.
McCarthy, J. J. (2001). Climate change 2001: impacts, adaptation, and
vulnerability: contribution of working group II to the Third Assessment
Report of the Intergovernmental Panel on Climate Change. Cambridge:
Cambridge University Press.
Meyer, W. B., Turner II, B. L. (1994). Changes in land use and land cover: a global
perspective. Cambridge: Cambridge University Press.
Milberg, P., Bergstedt, J., Fridman, J., Odell, G., & Westerberg, L. (2008). Observer
bias and random variation in vegetation monitoring data. Journal of
Vegetation Science, 19, 633644.
Moody, A., & Woodcock, C. E. (1995). The influence of scale and the spatial
characteristics of landscapes on land-cover mapping using remote sensing.
Landscape Ecology, 10, 363379.
Mosier, A., Schimel, D., Valentine, D., Bronson, K., & Parton, W. (1991). Methane
and nitrous oxide fluxes in native, fertilized and cultivated grasslands. Nature,
350, 330332.
Phadnis, M. J., & Carmichael, G. R. (2000). Numerical investigation of the influence
of mineral dust on the tropospheric chemistry of East Asia. Journal of
Atmospheric Chemistry, 36, 285323.
Pickup, G., Chewings, V. H., & Nelson, D. J. (1993). Estimating changes in
vegetation cover over time in arid rangelands using Landsat MSS data.
Remote Sensing of Environment, 43, 243263.
Ramsey, F. L. (1979). Parametric models for line transect surveys. Biometrika, 66,
505512.
Rango, A., Laliberte, A., Herrick, J. E., Winters, C., Havstad, K., Steele, C., & Browning,
D. (2009). Unmanned aerial vehicle-based remote sensing for rangeland
assessment, monitoring, and management. Journal of Applied Remote
Sensing, 3, 33542.
Reinke, K., Reinke, K., Jones, S., & Jones, S. (2006). Integrating vegetation field
surveys with remotely sensed data. Ecological Management and Restoration,
7, S18S23.
Richardson, M. D., Karcher, D. E., & Purcell, L. C. (2001). Quantifying turfgrass cover
using digital image analysis. Crop Science, 41, 18841888.
Stoner, D. C., Sexton, J. O., Nagol, J., Bernales, H. H., & Edwards, T. C. (2016).
Ungulate reproductive parameters track satellite observations of plant
phenology across latitude and climatological regimes. PloS One, 11,
e0148780.
Sutherland, W. J. (2006). Ecological census techniques: a handbook. Cambridge:
Cambridge University Press.
Turner, M. G., Dale, V. H., & Gardner, R. H. (1989). Predicitng across scales: theory
development and testing. Landscape Ecology, 3, 245252.
Turner, B. L., Lambin, E. F., & Reenberg, A. (2007). The emergence of land change
science for global environmental change and sustainability. Proceedings of
the National Academy of Sciences, 104, 2066620671.
Wang, J., Rich, P. M., Price, K. P., & Kettle, W. D. (2004). Relations between NDVI
and tree productivity in the central Great Plains. International Journal of
Remote Sensing, 25, 31273138.
White, R. P., Murray, S., Rohweder, M., Prince, S. D., & Thompson, K. M. (2000).
Grassland ecosystems. Washington DC: World Resources Institute.
Yim, J., Kleinn, C., Cho, H., & Shin, M. (2010). Integration of digital satellite data
and forest inventory data for forest cover mapping in Korea. Forest Science
and Technology, 6,8796.
Ko et al. Journal of Ecology and Environment (2017) 41:5 Page 11 of 11
... For decades, this technology has been developed and employed to collect data related to many types of biomass in various conditions (Franklin and Hiernaux, 1991;Jianya et al., 2008;Kumar et al., 2015;Pandapotan Situmorang and Sugianto, 2016;Roy and Ravan, 1996). Satellite data are easily collected and used for estimating carbon stocks derived from spatio-temporal geographical and global dimensions (Gonzalez et al., 2010;Ko et al., 2017;Myeong et al., 2006;Xiao et al., 2015;Yao et al., 2015). Experiments employing remote sensing are used to generate regression models focusing on the relationships between field observations and satellite picture vegetation indicators for estimating above-ground carbon (AGC) stocks (Yao et al., 2015). ...
... Based on the statistical evaluation, all VIs showed very close results with good R 2 , but the highest values with the least relative deviation of determination coefficient was to the NDVI followed by SAVI. This outcome concurred with the findings in the literature where the exponential model was the most reliable in the case of NDVI-AGC (Bright et al., 2012;Ko et al., 2017;Myeong et al., 2006;Yao et al., 2015). With the regression modeling, the best-fitting model was the exponential model of AGC-NDVI for PP and AP with R 2 (0.64, 0.87) and a p-value of less than 0.05% (0.005% for PP, and 0.001% respectively). ...
Article
Full-text available
Urban areas are significant sources of anthropogenic carbon dioxide (CO 2), which elevates air pollution. However, urban greenery has a positive effect on mitigating air pollution and the impact of CO 2 on the climate. Quantifying the benefits of greenery for urban environments involves complex calculations and requires significant resources. Such a quantifying exercise is not cost-effective. The satellite remote sensing method can analyze current and near-present carbon-stock dynamics through spectral band imaging. In this research study, field measurements determine above-ground carbon (AGC) stock. The field measurements are derived from three types of land use, comprising public parks and gardens, institutional parks, and street and avenue greenery in selected locations in Doha city in Qatar. These field measurements are then correlated with results from satellite images. Linear and non-linear regression models are established between AGC and five vegetative spectral indices (VIs) derived from the Landsat 8 Operational Land Imager (OLI). The AGC stock for the selected locations in Doha in 2014 is evaluated using the highest coefficient of determination with the highest accuracy expected. The results of the analysis reveal that both the normalized difference vegetation index (NDVI) (R 2 0.64) and the relative ratio vegetation index (R 2 0.71) significantly correlate with the AGC in public parks. In avenue vegetation, all the VIs exhibit high R 2 , but the best fit is NDVI (R 2 0.87). The CO 2 equivalent range evaluated from the AGC in the plots studied in Doha is measured as 650.6 tons for the period between 2014 and 2020, with an annual sequestration rate of 108.4 tons per year. This CO 2 equivalent storage amount has the social value of USD 42,286, which is the equivalent of QR 155,192. The AGC-VI correlation in land-use groups may be influenced by the turf grass and impervious surfaces in the background of the images. Further study of urban landscapes and vegetation with high biomass is likely to show its positive effects for cities and that it can improve carbon dioxide abatement, resulting in more sustainable societies. This improvement in CO 2 abatement in Qatar can be useful for various environmental estimations for the upcoming mega event of World Cup 2022.
... The field-based digital photography method is gaining popularity for the purpose of cover estimation, as it can reduce field time and enable additional analysis in the future (Ko et al. 2017 ) High resolution, nadir photography can serve as a realistic ground plot. It is information rich, understandable to a broad base of people, and the unanalysed information can be archived for future use. ...
Thesis
Natural rangelands are one of the significant pillars of support for the Libyan national economy. The total area of rangelands in Libya is c.13.3 million hectares. This resource plays an important role providing part of the food needs of the large numbers of grazing animals, in turn providing food for human consumption. In the eastern Libyan rangelands, vegetation cover has changed both qualitatively and quantitatively due to natural factors and human activity. This raises concerns about the sustainability of these resources. Observation methods at ground- based sites are widely used in studies assessing rangeland degradation in Libya. However, observations across the periods of time between the studies are often not integrated nor repeatable, making it difficult for rangeland managers to detect degradation consistently. The cost of such studies can be high in comparison to their accuracy and reliability, in terms of the time and resources required. These costs are not expected to encourage the local administrators of rangelands to make repeated or continuous observations in order to determine possible changes in managed areas. This has led to a lack of time-series data, and a lack of regularly updated information. The sustainability of rangelands requires effective management, which in turn is dependent upon accurate and timely monitoring data to support the assessment of rangeland deterioration. The aim of this research has been to develop a framework for monitoring and evaluating rangeland condition in the east of Libya with a prediction of the future condition based on a historical assessment. This approach was achieved through the utilisation of medium resolution satellite imagery to classify vegetation cover using vegetation indices. A number of vegetation indices applied in arid and semi- arid rangelands similar to the study area were assessed using ground-based colour vertical photography (GBVP) methods to identify the most appropriate index for classifying percentage vegetation cover. The vegetation cover data were integrated with climate data, topography and soil erosion assessment using the RUSLE system to form a Rangeland Assessment Management Information System (RAMIS). These data were used to assess the historical and predicted future rangeland condition. The MSAVI2 vegetation index was identified as the most appropriate index to map vegetation cover as this had good correlation with the ground data (R2 = 0.874). The RUSLE prediction identified that over 1,300,000 hectares were affected by soil loss over the time period from 1986 to 2010 representing nearly 97% of the study area. The RAMIS output indicated that most of the study area in 1986 was affected by a high risk of rangeland degradation, with less than 10 % of the area having a moderate and low-risk of degradation. The rangeland condition up to 2010 indicated a slight improvement in degradation distribution, with a slight decrease from 90% to 85% in the high risk of degradation area and the area having a low risk of degradation increasing from 2% to 8% in 2010. The result of the predictions made showed that the area of low cover class, which in 2017 reached about 1,280,000 hectares continues to increase through 2030 to 2050, to some 1,400,000 hectares, with a consequent increase in areas of high risk of rangeland degradation. The result of implementing the RAMIS framework over the historical period illustrates changes in the rangeland condition, reflecting the fluctuation in the effectiveness of rangeland management development projects linked to the financial resources available in the 1980s, with increasing numbers of grazing animals exceeding the rangeland capacity and the expansion of rangeland cultivation. Libyan rangeland managers need to focus more on expanding the fenced area, conducting soil survey, and implementing soil erosion studies that can be used in erosion model calibration at a large scale to better inform rangeland management planning. Otherwise, the future projections of change up to 2050 indicate a continuance of the deterioration of rangeland condition, increasing the areas of low vegetation. However, this projection is based only on the vegetation data as the lack of available climate data did not permit its incorporation into the prediction.
... The most utilized indices for vegetation and drought monitoring are Normalized Difference Vegetation Index (NDVI) and SPI, respectively. Currently, the products of MODIS sensor including vegetation indices (NDVI and EVI) have been effectively used for the assessment of vegetation dynamics monitoring and their responses to drought at various scales Dubovyk et al., 2015;Gulácsi and Kovács, 2015;Mu et al., 2016;Ko et al., 2017;Damavandi et al., 2016;Safari Shad et al., 2017). For example, Dubovyk et al. (2015) used MODIS-EVI time-series data to study the trend of vegetation changes in southern Africa. ...
... The most utilized indices for vegetation and drought monitoring are Normalized Difference Vegetation Index (NDVI) and SPI, respectively. Currently, the products of MODIS sensor including vegetation indices (NDVI and EVI) have been effectively used for the assessment of vegetation dynamics monitoring and their responses to drought at various scales Dubovyk et al., 2015;Gulácsi and Kovács, 2015;Mu et al., 2016;Ko et al., 2017;Damavandi et al., 2016;Safari Shad et al., 2017). For example, Dubovyk et al. (2015) used MODIS-EVI time-series data to study the trend of vegetation changes in southern Africa. ...
... The most utilized indices for vegetation and drought monitoring are Normalized Difference Vegetation Index (NDVI) and SPI, respectively. Currently, the products of MODIS sensor including vegetation indices (NDVI and EVI) have been effectively used for the assessment of vegetation dynamics monitoring and their responses to drought at various scales Dubovyk et al., 2015;Gulácsi and Kovács, 2015;Mu et al., 2016;Ko et al., 2017;Damavandi et al., 2016;Safari Shad et al., 2017). For example, Dubovyk et al. (2015) used MODIS-EVI time-series data to study the trend of vegetation changes in southern Africa. ...
Article
The assessment of relationships between satellite-derived vegetation indices and meteorological drought improves our understanding of how these indices respond to climatic changes. The combination of climate data and the Normalized Difference Vegetation Index (NDVI) product of Moderate Resolution Imaging Spectroradiometer (MODIS) imagery provided an opportunity to evaluate the impact of drought on land degradation over the growing seasons. The main goal of this study was to investigate the effect of drought on vegetation degradation in Meyghan plain, Arak, Iran. For this purpose, climatic and satellite data were used. The annual Standardized Precipitation Index (SPI) was calculated for 20 years (1998-2017). Then, the NDVI maps were classified into three classes according to the Tokunaga-Thug method. These classes are: Class 1) no vegetation; class 2) low-density or poor rangelands, and class 3) semi-dense and dense vegetation cover such as agricultural lands. The relationship between the percentage of vegetation cover classes (classes 2 and 3) and the drought index of the previous year was assessed using the Pearson correlation test. The results showed that the correlation between these variables was significantly dependent on vegetation degradation in the poor vegetation area (R=0.51; P-value<0.05). In contrast, there was a negative significant relationship between drought and the percentage of dense areas of vegetation (R=-0.46; P-value<0.06). Hence, it was concluded that the sensitivity of the low-density area (poor rangeland) to drought was more than dense vegetation covers (agricultural lands). Its reason is that the most important source of water supply for natural rangelands is the atmospheric precipitation that has been reduced due to the occurrence of droughts in recent years.
... Cover distribution provides a parametrised mathematical function that can be used to calculate the probability for an individual cover observation from a sample space and also allows estimation of the likelihood of having a cover observation equal to or less than a given value. Several field methods such as sampling strategy, line intercept, transect method, quadrat method and statistical analysis have been applied around the world at small scales in arid and semiarid regions for the estimation of groundcover variability (Zhou et al. 1998;Seefeldt and Booth 2006;Damgaard 2009;Ko et al. 2017). These techniques were also tested for Australian rangeland for an area of 10 000 m 2 with an assumption that the groundcover type is homogeneous throughout the area (Zhou et al. 1998). ...
Article
Full-text available
Due to the spatial and temporal variation of ground cover in grazing land, it is desirable to use a simple and robust model to represent the spatial variation in cover to quantify its effect on runoff and soil loss. The purpose of the study was to test whether a 2-parameter beta distribution could be used to adequately characterize cover variation in space at the sub-catchment scale. Twenty sub-catchments in the Burnett-Mary Region, Queensland, were randomly selected with the area varying from 35.8 to 231 km2. Thirty raster layers of ground cover at 30m resolution were prepared for these 20 sub-catchments with the average cover for the 30 layers ranging from 24% to 91%. Three methods (visual goodness-of-fit assessment and Kolmogorov-Smirnov (K-S) test, the fractional area with cover ≤ 53%, and estimated runoff amount for a given rainfall amount for the area with cover ≤ 53%) were used to test the appropriateness of the beta distribution to characterize the cover variation in space. The K-S test on 30 x 100 samples of ground cover showed that the hypothesis of beta distribution for ground cover could not be rejected at the 0.05 significance level for 97.5% of the cases. A comparison of the actual and beta distributions in terms of the fractional area with cover ≤ 53% showed that the discrepancy was no more than 8% for the 30 layers considered. A comparison in terms of the estimated runoff showed that results using the actual cover distribution and the beta distribution were highly correlated (R2 ranging from 0.91 to 0.98; the Nash-Sutcliffe efficiency measure ranging from 0.88 to 0.99). The mean absolute error of estimated runoff ranged from 0.98 to 8.10 mm and the error relative to the mean was 4 to 16%. The results indicated that the 2-parameter beta distribution can be adequately used to characterize the spatial variation in cover and to evaluate the effect of cover on runoff for these predominantly grazing catchments.
... Ignoring the NPV component often results in misleading declaration and incorrect conclusions as the findings do not correctly represent the true fraction of vegetation cover that protects soil from erosion (De Asis & Omasa, 2007;De Jong, 1994;Tanyaş, Kolat, & Süzen, 2015). It appears to be necessary to use an FVC index instead of NDVI as the earlier can better reflect the vegetation composition (Ko, Kim, Narantsetseg, & Kang, 2017;McKenzie et al., 2017). However, there is a significant knowledge gap in remote sensing of NPV and incorporating the information of NPV and ground litter into the C-factor estimation (Song et al., 2017). ...
Article
Full-text available
The cover‐management factor (C‐factor) is used in the revised universal soil loss equation (RUSLE) to represent the effect of vegetation cover and its management practices on hillslope erosion. Remote sensing has been widely used to estimate vegetation cover and the C‐factor, but most previous studies only used the photosynthetic vegetation (PV) or green vegetation indices (e.g. NDVI) for estimating the C‐factor and the important non‐PV (NPV) component was often ignored. In this study, we developed a new technique to estimate monthly time‐series C‐factor using the fractional vegetation cover (FVC) including both PV and NPV, and weighted by monthly rainfall erosivity ratio. The monthly FVC was derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) and Landsat data with field validation. We conducted the case study over China's Loess Plateau and analysed the spatio‐temporal variations of FVC and the C‐factor and their impacts on erosion over the Plateau. Our study reveals a significant increase in total vegetation cover from 56% to 76.8%, with a mean of 71.2%, resulting in about 20% decrease in the C‐factor and erosion risk during the 17‐year period. Our method has an advantage in estimating the C‐factor from total vegetation cover at a monthly scale providing a basis for continuously and consistently monitoring of vegetation cover, erosion risk and climate impacts. This article is protected by copyright. All rights reserved.
... For evaluating the accuracy of DCI-based FVC, a ground-truth FVC dataset was produced by manual classification of green pixels in digital camera images. The manual classification method has been applied in image-based vegetation cover estimation (Booth, Cox., and Berryman 2006;Ko et al. 2017), which imitates conventional field FVC observation such as quadrat-charting or ocular estimation technique (Cook and Bonham 1977;Gibbens and Beck 1988;Higgins et al. 1996). In this study, a photo with an area of 1.2 m 2 was divided into 157 × 118 sub-zones and the green vegetation pixel were counted by ocular decision. ...
Article
Full-text available
Fractional green vegetation cover (FVC) is a useful indicator for monitoring grassland status. Satellite imagery with coarse spatial but high temporal resolutions has been preferred to monitor seasonal and inter-annual FVC dynamics in wide geographic area such as Mongolian steppe. However, the coarse spatial resolution can cause a certain uncertainty in the satellite-based FVC estimation, which calls attention to develop a robust statistical test for the relationship between field FVC and satellite-derived vegetation indices. In the arid and semi-arid Mongolian steppe, nadir pointing digital camera images (DCI) were collected and used to produce a FVC dataset to support the evaluation of satellite-based FVC retrievals. An optimal DCI processing method was determined with respect to three color spaces (RGB, HIS, L*a*b*) and six green pixel classification algorithms, from which a country-wide dataset of DCI-FVC was produced and used for evaluating the accuracy of satellite-based FVC estimates from MODIS vegetation indices. We applied three empirical and three semi-empirical MODIS-FVC retrieval models. DCI data were collected from 96 sites across the Mongolian steppe from 2012 to 2014. The histogram algorithm using the hue (H) value of the HIS color space was the optimal DCI method (r2 = 0.94, percent root-mean-square-error (RMSE) = 7.1%). For MODIS-FVC retrievals, semi-empirical Baret model was the best-performing model with the highest r2 (0.69) and the lowest RMSE (49.7%), while the lowest MB (+1.1%) was found for the regression model with normalized difference vegetation index (NDVI). The high RMSE (>50% or so) is an issue requiring further enhancement of satellite-based FVC retrievals accounting for key plant and soil parameters relevant to the Mongolian steppe and for scale mismatch between sampling and MODIS data.
Article
Wildfires play a vital role in ecosystems and are one of the most significant natural processes that shape the northern Sierra Nevada and southern Cascade Mountain Ranges in northeastern California. At the same time, across the American West, a history of fire exclusion, timber salvage and harvest, and human-assisted reforestation has caused widespread changes in forest vegetation patterns over the past century. Federal agencies such as the National Park Service (NPS) and US Forest Service (USFS) have different missions and approaches to these land management activities, creating a mosaic of ecological conditions across managed landscapes both before and after wildfires. We studied vegetation recovery following a major fire that crossed jurisdictional boundaries to explore how management history (fire exclusion and timber harvest and salvage) and post-fire reforestation efforts may influence vegetation regeneration. Using both field-based data collection and remote sensing Normalized Difference Vegetation Index (NDVI) analyses, we assessed differences in vegetation communities found in the NPS and USFS portions of the footprint of the 2012 Reading Fire in northeastern California. We found that a legacy of timber harvest combined with fire exclusion was associated with denser shrub regeneration and minimal sapling recovery. In this context, areas that were reforested post-fire returned a relatively diverse and abundant suite of mixed-conifer tree species to the landscape. In areas with a history of fire exclusion that regenerated without human interference after the fire, saplings did establish on their own but were predominantly white fir (Abies concolor), a shade-tolerant species that is known to increase wildfire severity.
Article
Full-text available
The human habitation since the Palaeolithic time has been along with horticultural activity. In the earlier time horticulture was practiced for various needs and with indigenous methods, which is well documented in historical records. Later during the Neolithic period humans developed advanced techniques to grow and manage plants. Since then settlements evolved into towns and towns evolved into cities across the world in different civilizations. After industrial revolution in 1750s the cities swelled into Megalopolises and urban agglomerations. In the modern world every Mega city of the world needs to have healthy Green Index to maintain its ecological habitats. The quality of life of the populations of such places heavily depends upon it. Today there have been huge issues of Smog which has taken serious turn in the Mega cities of the world. Lahore is an example of Smog affected city. Karachi needs to be given special attention in this regard. Karachi being the largest city of Pakistan is also the most industrialized city as well. The natural environment of Karachi does not support vegetation because of the aridity. The fertility of soil is not so great and the sources of fresh water are located far away.
Article
Full-text available
Covering 40% of the terrestrial surface, dryland ecosystems characteristically have distinct vegetation structures that are strongly linked to their function. Existing survey approaches cannot provide sufficiently fine-resolution data at landscape-level extents to quantify this structure appropriately. Using a small, unpiloted aerial system (UAS) to acquire aerial photographs and processing theses using structure-from-motion (SfM) photogrammetry, three-dimensional models were produced describing the vegetation structure of semi-arid ecosystems at seven sites across a grass–to shrub transition zone. This approach yielded ultra-fine (< 1 cm2) spatial resolution canopy height models over landscape-levels (10 ha), which resolved individual grass tussocks just a few cm3 in volume. Canopy height cumulative distributions for each site illustrated ecologically-significant differences in ecosystem structure. Strong coefficients of determination (r2 from 0.64 to 0.95) supported prediction of above-ground biomass from canopy volume. Canopy volumes, above-ground biomass and carbon stocks were shown to be sensitive to spatial changes in the structure of vegetation communities. The grain of data produced and sensitivity of this approach is invaluable to capture even subtle differences in the structure (and therefore function) of these heterogeneous ecosystems subject to rapid environmental change. The results demonstrate how products from inexpensive UAS coupled with SfM photogrammetry can produce ultra-fine grain biophysical data products, which have the potential to revolutionise scientific understanding of ecology in ecosystems with either spatially or temporally discontinuous canopy cover.
Article
Full-text available
The effect of climatically-driven plant phenology on mammalian reproduction is one key to predicting species-specific demographic responses to climate change. Large ungulates face their greatest energetic demands from the later stages of pregnancy through weaning, and so in seasonal environments parturition dates should match periods of high primary productivity. Interannual variation in weather influences the quality and timing of forage availability, which can influence neonatal survival. Here, we evaluated macro-scale patterns in reproductive performance of a widely distributed ungulate (mule deer, Odocoileus hemionus) across contrasting climatological regimes using satellite-derived indices of primary productivity and plant phenology over eight degrees of latitude (890 km) in the American Southwest. The dataset comprised > 180,000 animal observations taken from 54 populations over eight years (2004-2011). Regionally, both the start and peak of growing season ("Start" and "Peak", respectively) are negatively and significantly correlated with latitude, an unusual pattern stemming from a change in the dominance of spring snowmelt in the north to the influence of the North American Monsoon in the south. Corresponding to the timing and variation in both the Start and Peak, mule deer reproduction was latest, lowest, and most variable at lower latitudes where plant phenology is timed to the onset of monsoonal moisture. Parturition dates closely tracked the growing season across space, lagging behind the Start and preceding the Peak by 27 and 23 days, respectively. Mean juvenile production increased, and variation decreased, with increasing latitude. Temporally, juvenile production was best predicted by primary productivity during summer, which encompassed late pregnancy, parturition, and early lactation. Our findings offer a parsimonious explanation of two key reproductive parameters in ungulate demography, timing of parturition and mean annual production, across latitude and changing climatological regimes. Practically, this demonstrates the potential for broad-scale modeling of couplings between climate, plant phenology, and animal populations using space-borne observations.
Article
Full-text available
The McMurdo Dry Valleys of Antarctica are a unique yet threatened polar biome. Cyanobacterial mats form a large part of the standing biomass in the McMurdo Dry Valleys and are therefore an indicator of ecosystem productivity and health. They are, however, patchily distributed, and this has hampered spatial ecology studies due to the logistical challenges of ground-based field sampling. Here, we report the application of remote sensing using a fixed-wing unmanned aerial vehicle (UAV) and GIS spatial mapping to identify cyanobacterial mats, estimate their extent and discriminate between different mat types. Using the Spalding Pond area of Taylor Valley as a test site, we were able to identify mats on soil surfaces within the hyporheic zone, as well as benthic mats below the water surface. The mapping also clearly identified the footprint of campsites and walking trails on soils, and we highlight the potential of this technique in monitoring human impact in this fragile ecosystem.
Article
Part I. Introduction: 1. Global land-use and land-cover change: an overview Part II. Working Group Reports: 2. A wiring diagram for the study of land use/cover change: Report of Working Group A 3. Towards a typology and regionalization of land-cover and land-use change: Report of Working Group B 4. Land-use and land-cover projections: Report of Working Group C Part III. Changes in Land Use and Land Cover: 5. Forests and tree cover 6. Grasslands 7. Human settlements Part IV. Environmental Consequences: 8. Atmospheric chemistry and air quality 9. Soils 10. Hydrology and water quality Part V. Human Driving Forces: 11. Population and income 12. Technology 13. Political-economic institutions 14. Culture and cultural change Part VI. Issues In Data and Modeling: 15. Modeling land-atmosphere interactions: a short review 16. Modeling global change in an integrated framework: a view from the social sciences 17. Data on global land-cover change: acquisition, assessment, and analysis Appendices Index.