1
1. Abstract
Vegetation maps play an important role in understanding vegetation dynamics at
Mono Lake, California. These maps act as a basis for studying succession, helping
establish correlation between vegetation and other environmental influences, and
providing important indications of how plants respond to disturbances. Current
vegetation maps for Mono Lake do not measure up to the standards of vegetation maps
that are accurate, easily understood and of high quality.
Employing
Rabbitbrush
(Chrysothamnus nauseous) and Big Sagebrush
(Atremisia tridentata) as study species, ground-truthing methods were utilized to take
flora samples at Mono Lake, California. Results showed that along a 500 meter transect,
moving away from the lakeshore, both the percent cover and number of individuals of
Rabbitbrush decreased where as Big Sagebrush increased. Topographic data was also
gathered along the transect, and found to be increasing, with the exception of a few level
data points that represented old beach ridges. Finally, the study examined how current
Geographic Information Systems (GIS) techniques aid in the making of effective
vegetation maps and how they could be applied at Mono Lake.
2
2. Introduction
Mono Lake is an interesting area to do research, due to its unique geology,
wildlife, and political history. Mono Lake is a terminal saline lake, located in the Great
Basin and Range Province of the Western United States (Figure 1).
Figure 1. The boxed in area represents Mono Lake’s location in the Great
Basin (Hart, 1996)
The area’s geologic past includes volcanism and fault activity, which is still present
today, and glaciation. Mono’s wildlife is very distinctive; the lake is home to the alkali
fly and brine shrimp, which are found only a few places in the world. Politically, its past
is full of controversy, which revolves around the taking of Mono Lake’s freshwater
tributaries by the city of Los Angeles.
Mono Lake is an appropriate place to conduct a study concerning GIS and
vegetation mapping because since the early 1900’s, the state of California has been taking
water from the Mono Basin. The city of Los Angeles Department of Water and Power
began diverting Mono Lake’s tributaries in 1941. As a result, the volume of the lake was
halved and the salinity doubled. In 1962, the lake had dropped 25 vertical feet and in
3
1995 had dropped 40. However, the perseverance of dedicated people and the help of
the courts, enabled Mono Lake to be regain many of its tributaries. In 1993, it was
ordered that the lake level eventually be restored to 6, 390 feet or higher (Hart, 1996,
www.monolake.org.).
Developing a vegetation database and using GIS would be beneficial as well as time
saving to the mapping of Mono Lake plant communities. With the lake levels increasing
on a yearly basis, it would easy to update maps and make changes that correspond to
those taking place in the environment. Using GIS techniques for mapping would also be
an efficient way to keep track of the fluctuating lake levels. The objectives of the study
conducted at Mono Lake relating to vegetation mapping and GIS are:
1. Conduct vegetation samples in order for the researchers to learn the types of groundtruthing that has to accompany map making, and get a feel for the amount of time it
would require to do a vegetation survey.
2. Provide background information on current vegetation present and Mono Lake, GIS,
and how vegetation mapping and GIS techniques have been integrated.
3. Discuss the results and draw conclusions.
3.0 Background Information
3.1. The Vegetation of Mono Lake
There are at least seven types of vegetation previously described in studies done
at Mono Lake, which current maps reflect (Figure2). These vegetation types have been
described from personal observations made by Burch et al (1977) and by Vorster (1985).
Figure 2(next page) Current vegetation map of Mono Lake(Mono Lake Committee,
1999)
4
However, Constantine (1993) also categorized the vegetation into six categories (Figure
3). The major difference in the categorization of these studies is that Burch (1977) and
Vorster (1985) considered transition zones, areas where major vegetative communities
overlap and Constantine (1993) did not.
Figure 3. A Current Plant Distribution of Mono Lake (Constantine, 1993)
The major types of plant communities present at Mono Lake are as follows as described
by Constantine (1993) are as follows:
(See also Figure 2 attached as the previous page)
3.1.1) Alkaline Sink Scrub/ Alkaline Herb - This community is characterized by
alkaline crusted soils and low moisture availability.
The vegetation is primarily
herbaceous and halophytic (salt-loving) in nature. The dominant species are Rabbitbrush
(Chrysothamnus nauseosus), Greasewood (Sarcobatus vermiculatu), and Saltgrass
(Distichlis spicata).
5
3.1.2) Freshwater Marsh/ Wet Marsh – This type of community is characterized by
the water table at the surface originating from non lake sources, a presence of standing
water, and nutrient rich soil that is oxygen deprived. The dominant species are Monkey
Flower (Mimulus guttatu), Common Cattail (Typha latifolia), Foxtail Barley (Hordeum
jubatum) and other grasses and sedges.
3.1.3) Sagebrush Scrub/Dry Scrub - This community is the most common in the Mono
Basin. It is characterized by a sparse herbaceous layer and a shrub cover of > 5%. Due
to the volcanic activity in the area, the soils are usually deep, well drained, and nonalkaline.
Plants in this community must be able to tolerate a variety of extreme
conditions and most are equipped with large deep taproots. The dominant species are Big
Sagebrush (Artemisia tridentata), Bitterbrush (Purshia tridentata), Giant Blazing Star
(Mentzelia laevicaulis), and a few Desert Peach (Prunus andersonii).
3.1.4) Pinyon-Juniper Woodland - Juniper woodland typically occurs at higher
altitudes, where the climate is colder and wetter. Woodlands are classified by being open
forests where trees and intermixed with shrubs and grasses. The presence of Pinyon and
Juniper is dependent on the moisture availability, altitude, and soil depth. The dominant
species in this community are Pinyon Pine (Pinus monophylla), Utah Juniper (Juniperus
osteosperma), Mormon Tea (Ephedra virdis), and Cut-leaf Mahogany (Cercocarpus
ledifolius).
3.1.5) Jeffrey Pine Forest - This plant community does not occur directly near the lake,
but in the surrounding mountains. The largest stand of Jeffrey pine in the world occurs
near the southeast end of the lake. The dominant species are Jeffrey Pine (Pinus jeffreyi),
Lupin (Lupinus spp), and Prickly Phlox (Leptodactylon pungens).
3.1.6) Riparian Forest - These forests occur along the streams that run down the
mountains and into Mono Lake. The soils are layers of fine and coarse material that has
been deposited by varying stream levels. This community is characterized mainly by
6
deciduous species.
Plants in this community must be able to withstand periodic
flooding. These communities also provide great habitat for wildlife. The dominant
species are Quaking Aspen (Populus tremuloides), Black Cottonwood (Populus
trichocarpa), and Coyote Willow (Salix exigua).
3.2 The Importance and Use of Vegetation Maps
Vegetation is a fundamental aspect of the environment, animals use it as habitat, it is
a renewable resource that provides food and fuel for humans, and it is very influential in
affecting other earth processes such as the carbon and nitrogen cycles. Vegetation maps
are important because they can model the current vegetation of an area and can be used to
make predictions about future vegetation changes. Vegetation maps provide in part for
the basis of understanding the world around us and for solving complex problems, such
as management. Therefore, it is important to be able to map vegetation thoroughly.
However, it is important to remember that any vegetation classification whether it be in
map form or not, is a simplification of a complex reality of gradients and mosaics in
vegetation (Whittaker, 1970).
Many vegetation maps of the past and present have traditionally focused on what the
map communicates rather than analysis. Demers (1991) explains that the purpose of
most maps is to produce a visual thematic pattern that corresponds to the vegetation
classification employed.
There is not much quantitative map information available
except for measurements taken manually because the classification process often
combines numerous vegetative attributes in a given area (Demers, 1991).
In the past, vegetation maps were constructed based on observations of mappers and
scientists who surveyed the vegetation using a variety of field methods. Consequently,
the maps were not as accurate, were fairly subjective, and it took a long time to make
them.
For example, in Mammoth Cave National Park, scientists spent three years
7
surveying all the vegetation (University of Idaho, 1999). However it is important to
note that no single method of mapping vegetation is best because the methods used are
largely determined by the purposes for which the map will be used (Kiichler, 1988).
Many maps have been made using time consuming visual interpretation of aerial
photographs, planimeters, and field-collected data. As time has progressed, vegetative
cover has been determined by spectral data from satellites, which can produce maps
much faster but far less detailed. Most manual mapping with satellite data is done with
film products that are not optimally suited for mapping heterogeneous natural vegetation
mainly because the film has already been radiometerically and geometrically rectified
(University of Idaho, 1999). Problems that occur can be terrain shadowing and poor
spectral separation of cover types, which requires additional data and high quality
products. In the long, this run means higher costs (University of Idaho, 1999). There are
problems with correcting spectral separation because many types of vegetation have the
same spectral signature and therefore usually receive the same class label even though
these communities should be classed as separate entities. At this point, Geographic
Information Systems (GIS) can be very useful.
3.3 A Brief Overview of Geographic Information Systems (GIS)
Geographic Information Systems are computer based systems for storage, display,
and manipulation of spatial data (Clarke, 1997). GIS is useful for vegetation mapping
because it can perform complicated overlays and spatial analysis that historically have
been time consuming and very difficult. GIS databases make it possible to model,
combine, and analyze spatial relationships between data sets for the characterization and
mapping of plant communities
8
The common use of GIS is the storage of thematic data layers, such as vegetation
that can be superimposed onto other data layers. GIS data is commonly stored in a
computer in two forms: raster or vector. Vector data represents spatial information with
lines in coordinate space and stored as points, line, and polygons (Figure 4) and can have
textual information stored with them in accompanying databases. Lines are stored as
strings of coordinates and spatial relationships can be computed if needed. The vector
model permits the closest digital approximation to the original map and also retains
spatial relationships such as network linkages, object areas, perimeters, and shared
boundaries (Clarke, 1997, University of Idaho, 1999). Raster data divides space into
fields and assigns each field a unique value, usually displayed in a square lattice or grid
(Figure 4). Raster structure is convenient for imaging systems such as digital satellite
imagery. (Figure 5) (Clarke, 1997 and University of Idaho).
Figure 4. A represents a heterogeneous area modeled in a vector type and B
represents the same area modeled in a raster type
(University of Idaho, 1999).
9
Figure 5. The boxed in area represents Mono Lake in the satellite
image (www.monolake.org).
Current vegetation mapping is utilizing integration of GIS techniques, satellite data,
software programs, and ground-truthing. Recently, scientific studies being conducted
have illustrated the usefulness of GIS techniques in vegetation mapping. A study in the
foothills of San Diego used a combination of ARC/INFO (a well-known GIS package)
and a statistical analysis program called S-Plus to map vegetation (S-Plus, 1999). The
team of professors and students analyzed Landsat Thematic Mapping satellite data to
map general vegetation categories, and to obtain more detailed vegetation types, they
collected field data for each location. ARC/INFO was then used to bring all the data
together and to determine the most probable vegetation growing at each site on the map
(S-Plus, 1999).
In Norway, mapping plant communities is accomplished using infrared aerial
photographs, GIS, and digital elevation models (DEM). The study utilized groundtruthing in which vegetation was sampled systematically along transects that covered all
the major plant communities. A Global Positioning System (GPS) unit was also used to
determine the positions of the samples, which ensured that the sample area could be
10
precisely matched to the corresponding area on the infrared aerial photographs. The
infrared photographs were then transformed into a raster format with a scanning device.
This information was then inserted into a GIS.
Nilsen, et al, (1999) explained that the
modeling method operates by crosslinking field data (i.e. the defined plant communities)
and the raster layers of the GIS. Probability models were built to establish links between
GIS data layers, some of which were derived from the DEM, and plant communities
resulting from classification of field data.
4 Methods
4.1 Study Area
The study area was located on the south/southeast side of Mono Lake, just west of
Navy Beach, southeast of Lee Vining, California. The plant communities sampled are
classified as Alkaline Sink Scrub and Big Sagebrush Scrub (Figure 6).
Figure 5. The boxed in area represents the location where vegetation samples
were taken. (Hart, 1996)
4.2 Field Methods
One continuous transect 500 meters in length was run south from a point on the
shore of Mono Lake, west of Navy Beach. Circular quadrats with a radius of 5.6 meters
11
were used to sample shrub vegetation (Kent and Coker, 1994). One circular quadrat
was set down at random intervals based on a random number table (Zar, 1974). In each
quadrat the number of both Rabbitbrush (Chrysothamnus nasueous) and Big Sagebrush
(Artemisia tridentata) were counted and the percent cover of each species was estimated.
Topographic data was also gathered along the transect using a laser level (See Appendix
A). Global Positioning Systems data was gathered using a GPS unit and an exact position
for each quadrat was established.
4.2.1 Vegetation Description
As a result of the limited time for this study, only one transect was sampled. Two
plant species, Rabbitbrush (Chrysothamnus nauseosus) (Figure 7) and Big Sagebrush
(Atremisia tridentata) (Figure 8) were considered in the study. These two species were
chosen because of their abundance and because they are easily recognized.
Figure 7. Rabbitbrush (Chrysothamnus nauseosus) ( Bowers, 1993)
12
Figure 8. Big Sagebrush (Artemisia tridentata) (Bowers, 1993)
5. Results
5.1 Vegetation Data
Results from the vegetation samples are displayed in Table 1. The data reveals
that the percent cover and number of individual Rabbitbrush (Chyrsothamnus nauesous)
was highest closest to the shoreline and decreased with increasing distance from the
shore.
The percent cover and number of individuals of Big Sagebrush (Artemisia
tridentata) was lowest closest to the shoreline and increased with increasing distance
from the shore. However, quadrats 2,7, and 10 did not follow the general trend.
Table 1. Represents Vegetation Data Collected in the Field
Quadrat
Number
Point of
Quadrat on
Transect in
Meters
% cover of
Rabbitbrush
1
2
3
4
5
6
7
8
9
10
11
0
13
43
81
101
157
195
218
223
244
273
0
70
35
85
72
45
20
15
10
10
<1
% Cover of
# of
# of
Big
individuals of individuals of
Sagebrush Rabbit brush
Big
Sagebrush
0
0
<1
0
0
5
0
0
0
11
30
0
21
19
32
36
25
7
16
11
15
2
0
0
2
0
0
14
0
1
1
4
12
13
12
13
14
15
16
17
18
296
328
364
397
456
485
496
0
<1
0
0
0
0
0
35
55
70
50
75
81
79
0
1
0
0
0
0
0
17
10
23
27
30
33
29
5.2 Topographic data
A topographic profile is presented in Figure 9. Results showed that topography
increased with increasing distance form the shoreline. However, there were areas where
the change was very small. These locations represent old beach ridges and manifested
themselves in the form of terraces.
Location along transect
(m)
Topographic Profile
500
400
300
200
100
0
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57
Height (m)
Figure 9. Graph represents topographic profile
5.3 Global Positioning Systems (GPS ) Data
Raw GPS data was also gathered. The exact position of every qaudrat sampled was
taken using a GPS unit. Figure 10 shows an image of the transect with each circle
representing a quadrat.
14
Figure 10. Represents the raw GPS data gathered along the transect.
6. Discussion
The results from the vegetation samples taken at Mono Lake showed a density
gradient of vegetation moving away from the lake shore. The occurrence and percent
cover of Big Sagebrush (Atremisia tridentata) increased from the shore of the lake and
Rabbitbrush (Chrysothamnus nauseous) decreased. This could be due to Rabbitbrush
being more salt tolerant and having higher water demands than Big Sagebrush.
It is
important to note that the time that elapsed while gathering this information was over two
hours and covered only a small portion of the vegetation present. Time is a factor for
people who make maps to consider when taking on the daunting task of ground-truthing a
large area of vegetation.
Another factor that could have influenced the results is topography. Results
showed that topography also formed an increasing slope along the transect. An increased
slope may have increased the vegetation’s exposure to higher amounts of wind and
weather elements. Big Sagebrush is a hearty plant and may be more tolerant to harsh
15
weather conditions, thus being able to thrive in elevated areas. Along the topography
gradient there were three spots where the ground leveled out and vegetative cover
decreased. These areas represent remnants of a previous lake shore or beach ridge. The
reduced amount of vegetation may possibly be due to an increased salt and mineral
deposit left behind by evaporated lake water.
Raw GPS data was gathered so that if needed, locations of field sites could be
directly located on a map or satellite imagery.
The GPS positions allow direct
comparisons of existing map information to actual field locations and thus map
discrepancies can be easily resolved.
7. Conclusion
The field methods conducted in the study gave the researchers a clear idea of how
ground-truthing methods for vegetation mapping are conducted, the methodology, and
the amount of time involved. The study showed that Big Sagebrush(Artemisia tridentata)
and Rabbitbrush (Chrysothmanus nauseous) form a distribution gradient moving away
from the shore of Mono Lake. Topography also showed a gradient, increasing south of
the shoreline. The was study conducted to see how GIS technology aids advancing the
process of vegetation mapping. After examining Mono Lake, it is evident that it would
be a good place to utilize GIS techniques to make high quality, accurate, and easy to
update vegetation maps
16
Appendix A
Location along transect(m) Actual Height (m)
0
3
5
10
15
18
21
25
30
32
34
38
41
44
45.8
46
50
60
70
80
120
130
140
150
160
170
180
190
200
210
220
230
237
240
244
250
260
270
280
290
300
310
320
330
340
350
360
370
0
0.11
0.01
0.18
0.44
0.49
0.54
0.71
1.04
1.44
1.57
1.8
2.17
2.45
2.93
2.69
3.19
3.55
3.93
4.3
4.54
5.24
5.94
6.22
6.81
7.75
8.43
8.73
8.73
8.94
9.32
9.66
9.99
10.38
10.99
10.97
11.56
12.49
12.82
13.38
13.95
14.34
14.84
15.01
15.33
15.82
16.07
16.5
17
380
390
400
410
420
430
440
450
460
500
16.7
16.95
17.29
17.36
17.44
17.59
18.13
20.18
20.78
18.63