Ph.D cand., Biological Environmental Engineering,
Graduate School of Agricultural and Life Sciences, University of Tokyo
1-1-1 Yayoi, Bunkyo-Ku, Tokyo, 113-8657
Tel: +81-3-5841-5345 Fax: +81-3-5841-8169
E-mail: [email protected] [email protected]
Associate Prof., Institute of Policy and Planning Science,
University of Tsukuba 1-1-1 Tennou-dai, Tsukuba, Ibaraki, 305-8573
Tel: +81-298-53-5005 Fax: +81-298-55-3849
Research Associate., Institute of Low Temp.
Sci. Hokkaido University Sapporo, 060-0819
Tel: +81-011-706-5490 Fax: +81-011-706-7142
Prof., Biological Environmental Engineering,
Graduate School of Agricultural and Life Sciences,
University of Tokyo 1-1-2 Yayoi, Bunkyo-Ku, Tokyo, 113-8657
Tel: +81-3-5841-5370 Fax: +81-3-5841-8169
Kushiro wetland located in the northeast Japan is designated as Ramsar Convention, wetlands of international importance. Plant ecologists have urged the importance of preservation of the vegetation characterized by high biodiversity and high spatial heterogeneity, as well as the necessity of intensive monitoring of distribution according to the environmental changes and the disturbance dynamics. Recent advances of remote sensing technologies have provided various information of community types in wide range, in spite of the inaccessibility in the field. The satellite and airborne are available platforms for acquiring the various information of wetland such as land cover and vegetation types . Aerial photographs have been widely used as the remotely sensed data source for wetland information acquisition historically . The textural features and superior spatial resolution of aerial photography make a very useful data source for detailed wetland mapping . However the lack of spatial resolution, derived from the complicated structures in wetland, resulted in extracting relatively coarse vegetation unit. It was not feasible for identifying similar or specific vegetation types. Hence, without close observation it is difficult to identify analogous vegetation types dominated typically in this wetland. Since wetlands are often spatially characterized by steep ecological gradients with vegetation units narrower than the pixel size of current sensors . Thus, wetland delineation up to species level usually needs more intensive tasks in the field trips  and the platforms restricts the utility of remotely sensed imagery with different spatial resolution for mapping wetlands. Therefore, more proximate platforms to the earth such as a balloon is necessary for detail monitoring to obtain finer resolution data, since the units of wetland vegetation are often smaller than resolutions of satellite sensors. The advantages of utilization of balloons cost low and ease of use. Nevertheless, to our surprise, very few studies of wetland observation have been conducted by using balloon photographs as a tool to extract the training data of vegetation, especially in wetland. The paucity of such studies probably causes the uncertainty for extracting unreliable training data of vegetation. In this study, we tested two tethered helium filled balloons as platforms for monitoring of wetland vegetation in Kushiro wetland. The aim of our study was to evaluate the potential of the utility for aerial balloon photographs to classify wetland vegetation.
ur study site is located in Akanuma marsh in Kushiro wetland (N 43°6′ 19″, E 144°21′ 23″). It is legally appointed as the specially preserved area and is characterized by high biodiversity and high spatial heterogeneity of vegetation. Kushiro wetland is the largest in Japan, spreading over 18,290 ha.( Figure 1). Low moor is spread around lake Akanuma, and transitional and high moors are spread away from the lake. Alunus japonica and Phragmites communis dominated mainly in low-moor, Myrica gale var. tomentosa and Carex spp. dominated in transitional moor, Sphagnum bog, alpine plants and dwarf shrubs dominated in high moor as broad categories, respectively.
Figure 1. Study site. Kushiro Wetlnd, northeast Hokkaido in Japan
The method comprises five main stages, 1) ground truth and taking balloon photos 2) image calibration and mosaicking 3) photo interpretation and classification for base map. 4) application of balloon mosaic photos as training data for classification of aerial NIR color video images. 5) application of SPOT (HRV) images for the classification of whole area of wetland. Processing of satellite imagery and aerial balloon photography were undertaken in the ERDAS Imagine version 8.4 software (REDAS Inc.) digital image processing system and Arc View version 3.2 software (ESRI Inc.).
Figure 2. Ground truth points, vegetation research and
GPS survey, July 26-27 in 2001 and Jun 30 – July 2 in 1998
In the first step, The vegetation research (July 26 – 27 in 2001 and Jun 30 – July 2 in 1998) was carried along the wooden path way in the direction of North-South, 59 points of 1m×1m size quadrates and 52 points of 10m×10m size quadrates were set. While in the direction of East-West, 52 points of 20m×20m size quadrates were set every 20m. The ground truth points were located in the ” Figure 2 “. The position coordinates were measured using two sets of ground positioning system (GPS, Trimble Corp., 4600LS) by carrier phase DGPS. We tried balloons as platforms for monitoring of wetland vegetation to make vegetation map in detail. Aerial color photographs were taken by two tethered helium filled balloons from nadir view at 100-200 m height on the 2nd of July in 1998 ( Table 1 ). The observation area covers 230m×460m. In the second step, the photos were scanned (600dpi) and corrected for geometric distortions and combined into georeferenced image mosaics (15cm / pixel). The mosaicking of aerial balloon photographs was undertaken in the MOSAIC module ERDAS IMAGIN. These images were projected onto the UTM coordinates system by resampling and digital ortho-photo maps. In the third step, the digital processed vegetation map was constructed by mosaicked photos with manual photo interpretation and field validation was conducted by Arc View GIS version 3.2 software. In the fourth step, NIR color video (SILVACAM, VTT Automation Co., Finland) sequence images (30cm / pixel) were taken by an airplane (Dornier., Do228-200, National Aerospace Laboratory of Japan) from nadir view on 30 June, 1998 at 900 m height( Table 2 ). One 750×480 pixel size image was taken in every 1/30 second and the 500 video sequence images were overlaid using an algorithm  of tie points extraction based on image matching and registered to UTM WGS 84 coordination system using ground control points (GCPs). In the fifth step, Two remotely sensed satellite data sets were obtained. SPOT2 (HRV) images were acquired on 10 June 1998 and 17 June 2000. Images were ortho-rectified. Resampling was performed using the nearest neighbor algorithm. The whole of the Kushiro wetland vegetation was classified using the base map., June, July and August. The slight seasonal change was detected to cover the whole of the vegetation community types into the genus and species level. Lastly, the accuracy assessment was done about both of the base map and the satellite images.
Table 1. Performance of tethered helium filled ballons.
|altitude:||100 – 200 m|
|coverage size / shot :||15 m × 20 m / scene and 50 m × 70 m / scene|
|mosaicked coverage area :||230m × 460m|
|number of available georeferenced scenes :||23 / 66 shots|
|focal length of still camera :||28mm|
|resolution :||15 cm / pixel|
|permanent flight duration / balloon :||48hours|
|actual flight duration / day :||4-5 hours|
|permanent weight :||10kg|
Table 2. Performance of the airborne color near infrared (CNIR) video image
|coverage size:||250 m × 1025 m|
|bands:||Red (760-900nm), Green(580-680nm), Blue(490-580nm)|
|survey airplane:||Dornier., Do228-200|
|CNIR video:||SILVACAM, VTT Automation Co., Finland|
|observation angle:||nadir and off-nadir, 45 degree zenith|
|resolution:||30 cm / pixel|
|film size:||36 mm,|
|number of overlaid images:||500 scenes|
|image size:||750 × 480 pixel|
Results and conclusions
Aerial balloon photo interpretation and vegetation map
The interpretation of aerial balloon mosaicked photographs with high resolution enabled us to identify the vegetation up to specific ten types and twenty-seven categories of individual vegetation in each wetland type. ” Figure 3 ” and ” Figure 4 ” shows the all vegetation map and preservative vegetation map constructed by photo interpretation, respectively. We could identify 58species in all. 10types and 27categories of individual vegetation communities and 5types and 11 categories for conservation species( Figure 5 ). The individual categories and vegetation types were as follows: 3 types and 5 individual categories in the low moor (1. auiherbosa type; Potamogeton octandrus. poir, 2. shrub and herbaceous type; Alnus japonica, Alnus japonica – Phragmites communis, Alnus japonica – Carex spp.-Phragmites communis, Alunus japonica – Carex spp.), one type and one category in transitional moor, (3. herbaceous type; Carex spp.), 3 types and 6 categories in the low-transitional moor, (4. shrub, small shrub and herbaceous type; Myrica gale var. tomentosa, Myrica gale var. tomentosa – Carex.spp – Chamaedaphne calyculata Moench – Vaccinium uliginosum, 5. small pool, small shrub and herbaceous type; Carex stipata (Muhlenberg) – Phragmites communis, Carex spp. – Vaccinium uliginosum, Carex middendoffii, 6. small shrub and herbaceous type; Carex middendoffii – Chamaedaphne calyculata Moench – Vaccinium uliginosum ), one type and one category in transitional moor, (7. herbaceous type; Carex spp.), 3 types and 13 categories in the high moor, ( 8. pool, moss, herbaceous and alpine plant type; pool – Sphagnum fuscum, pool – Eriophorum gracile – Eriophorum vaginatum, pool – Sphagnum fuscum – Eriophorum vaginatum, pool – Equiserum fluviatile(L.) – Menyanthes trifoliate, pool – Hosta rectifolia – Loiseleuria procumbens – Empetrum nigrum, 9. moss and dwarf shrub type; Sphagnum fuscum – Sphagnum subfuvum, Sphagnum magellanicum, Sphagnum fuscum- Sphagnum papillocum, Sphagnum – Carex spp.- Loiseleuria procumbens – Empetrum uliginosum, Sphagnum fimbriatum – Sphagnum subsecundum-Carex spp, 10. small pool and herbaceous type; Equiserum fluviatile(L.) – Eriophrum gracile – Menyanthes trifoliate, Equiserum fluviatile(L.) – Eriophorum gracile, Equiserum fluviatile(L.) – Phragmites communis – Chamaedaphne calyculata Moench). As for Carex spp., it included Carex lasiocarpa, Carex pseudo curaica, Carex middendoffi, Carex myngbyei. We could discriminate the difference of the distribution of Equiserum fluviatile(L.), Carex spp. and auiherbosa. We could detect the difference of the phenology of Carex spp. This difference represented the darker brown patches (non-heads or forming of the heads) and light brown-pink patches (heads with flowers) in the transitional moor. And moreover, we could discriminate the vitality, the status of the condition of vegetation in Phragmites communis. This represented the brown-darker green patches (weak or deteriorated leaves, dominated in the dry inland away from the Akanuma pond) and the blue-green patches (vigorous leaves, dominated within 100 m from Akanuma pond). And further more, we clearly identified the four species of Sphagnum mosses, Sphagnum fuscum, Sphagnum subfuvum, Sphagnum magellanicum, and Sphagnum papillocum. It was effective to classify especially among the vegetation such as small shrub mixed with herbaceous plants, moss bog with pools and dwarf shrubs with sedge, moss and alpine plants by comparing to the intensive ground truth data.
Figure 3. Vegetation map in Akanuma marsh in
Kushiro Wetland constructed by balloon mosaicked photo interpretation
Figure 4. Conservation vegetation map in Akanuma marsh in Kushiro Wetland constructed by balloon mosaicked photo interpretation.
Airborne CNIR video images clustered
The CNIR video images were clustered (ISODATA) by using aerial balloon mosaicked photo as the training data. Figure 6 shows the clustered image taken from Nadir angle. On the supervised classification, the training area in the video image could be chosen as pure pixels according to the photo manual interpretation. Akanuma marsh vegetation was classified into the following 13 community types; 3 types in the low-moor ; (1). Aquihebosa plants, (2). Alunus japonica, Alunus japonica – Phragmites communis, Alunus japonica – Carex.spp, Alunus japonica – Carex.spp – Phragmites communis, (3). Carex.spp – Phragmites, 4 types in the transitional-moor ; (1). Myrica gale var. tomentosa, (2). Myrica gale var. tomentosa-perennial plant, (3). Carex.spp-prennial plant (4). Carex, 6 types in the high-moor; (1). Sphagnum bog, (2). Sphagnum bog – small shrub, (3) Sphagnum-perennial herb, (4). Sphagnum bog – alpine plant – monocots grass, (5). Sphagnum bog – alpine plant (6). dwarf shrub – alpine plant. Although the coverage size of the aerial mosaicked photo is smaller than the airborne CNIR video image, we could identify the typical communities. Compared to the results of vegetation base map, we could not directly identify the component of sub species in the video image taken from Nadir angle. (such as Sphagnum dominated species in high moor and Myrica gale var. tomentosa in the transitional moor.) While monocots (hard) grass were easily identified into the species level as well as the base map.
Figure 5. Legend of final vegetation map, showing 10 types of specific vegetation
and 27 categories of individual categories
in each wetland type. (*)This mark show the category includes conservation species.
Figure 6. Airborne CNIR video image taken from nadir angle and clustered (ISODATA) by using
base map constructed by aerial balloon
As for the supervised classification of SPOT images ( Figure 7), the phenology of Carex spp mixed shrub was clearly detected than Chamedaphne calyculata Moench or Alunus japonica. These results are important because this level of detailed classification up to the species level and categories could not be retrieved without utilizations of balloons. Since this classification up to species level could not be identified from any of the satellite image data sets directly for small-scale studies. By using balloon as the platform, it was possible for us to take aerial photographs whenever we want to conduct the experiment. We could monitor the change of phenology and the status of the condition among the vegetation of green grasses and sedge. It was certain that balloon technique helped us provide the additional information of analyzing the SPOT images.
The utilization of balloons as the platform of monitoring the terrestrial material is one of the basic techniques in the remote sensing methods. However, it is one of a most effective technique to get the reliable information of vegetation for detailed classification. The availability of these kinds of mapping, as the status of vegetation and the detailed community types will be indispensable and feasible to the conservation for such as biodiversity and spatial heterogeneity. Since wetlands come under the pressure of human activity, it will help us apply to the observation of disturbance dynamics, restoration and invasive species.
Figure 7. Supervised classification of Spot 2 (HRV) images. The training map
composed by the mosaicked photo was used for the classification.
Images were aquired on 10 June 1998 and 17 June 2000.
- K. R. Harvery and G. J. E. Hill, “Vegetation mapping of a tropical freshwater swamp in the Northern Territory, Australia: a comparison of aerial photographs, Landsat TM and SPOT satellite imagery,” International Journal of Remote Sensing, vol.22, pp.2911-2925, 2001.
- J. R. Jensen, E. J. Christensen, and R. Sharits, “National wetland mapping in South Carolina using airborne multispectral scanner data,” Remote Sensing of Environment,vol.16, pp.1-12, 1984.
- R. Kadmon and R. Harari-Kremer, “Studying long-term vegetation dynamics using Digital Processing of Historical Aerial Photographs,” Remote Sensing of Environment, vol.68, pp.164-176, 1999.
- C. A. Jennings, P. A. Vohs, and Dewey M. R, “Classification of wetland area along the upper Mississippi River with aerial videography,” WETLANDS vol.12, pp.163-170, 1992
- R. M. Johnston and M. M. Barson, “Remote sensing of Australian wetlands: an evaluation of Landsat TM data for inventory and classification,” Australian Journal of Marine and Freshwater Research, vol.44, pp.223-232, 1993.
- C. Juan, J. D. Jordan, and C. Tan, “Application of airborne hyperspectral imaging in wetland delineation,” Proceedings of Asian Conference of Remote Sensing, vol.2, pp.834-839, 2000. 2000.
- Kushida, K., K. Yazawa, T. Tamaru, M. Fukuda, K. Yoshino, G. Takao, S. Kuniyoshi, Y. Okada, and A. Tokairin: “Multiangular measurements with aerial video sequence imagery in Kushiro-shitsugen”, International Archives of Photogrammetry and Remote Sensing, Vol. 32, pp. 55-59, 1999