◇ 第28回 地理情報システム学会 学術研究発表大会 プログラム ◇
28th GISA Timetable
IAG'i セッション:参加費無料 IAG'i: Free entry 무료 免費
GISA 企画・ハンズオンセッション:参加費無料 GISA Special, Hands-on Sessions: Free entry 무료 免費
GISA 口頭発表セッション:参加費必要 GISA Oral Presentations: Fee requested 유료 収費ポスター Poster 포스터 海報
10月19日 (土/토/六) Oct 19 Sat
時間 Time
Room A @ 201
Room B @ 203
Room C @ 204
Room D @ 301
Room E @ 302
Room F @ 303
Room H @ 205
ポスター会場 Poster @ 403-404
16:10-16:30
閉会式 Closing (優秀発表賞の表彰を含む) 폐막식 閉幕式HTML program, English, Chinese, Korean translations by AI, Hisatoshi
発表題目一覧 (題目をクリックすると概要を表示します。セッション単位での表示/非表示の切り替えも可能です)
Title and abstract of presentations (To show the abstract, click on the title)
발표 제목 일람 (제목을 클릭하면, 개요가 표시됩니다)
標題的一覧 (点撃標題, 可以表示要旨)
10月19日 (土) 9:00〜11:00
C-1: IAG'i (1) Chair: TSUTSUMIDA, Narumasa
C-1-1 Mapping Urban Areas Using the Landsat-8 Image Acquired in Daegu, South Korea by Using Machine Learning and Urban Index
PDF Yun-Jae Choung, Eung-Joon Lee, Hyung-Wook Choi
Mapping urban areas is important for monitoring urban development and preventing urban disasters. In this research, the two algorithms such as machine learning and urban index were employed for mapping the urban areas using the Landsat-8 image acquired in Daegu, South Korea through the following steps. First, the machine learning technique was employed for detecting the urban areas from the given Landsat-8 image. Next, the urban areas were also detected from the urban index image generated using the multispectral bands of the Landsat-8 satellite image. Finally, the accuracy of these detected urban areas was separately assessed.
C-1-2 Aerial Imagery-based Estimation of Marine Farming Facilities Capacity in Wando-gun, South Jeolla Province
The marine farming business is conducted in seas, making it difficult to identify the state of marine farms, and the illegal marine farming facilities worsen the fishing environment, and lead to the production and distribution of low-quality farmed marine products thus harming the appropriate supply of marine products and stabilization of prices. Therefore, this study used high-resolution aerial images of Wando-gun, South Jeolla Province, Korea in estimating the facility capacity for laver, sea mustard, abalone, fish and oysters so the marine farms state was identified along with the identification of illegal marine farming. To that end, the marine farming approval book was arranged according to the standardized table so as to build property DB. The location and area map of the water surface was digitized based on the GPS coordinates, and thus a spatial DB was constructed. The DB then was integrated into a single GIS database and superimposed on the aerial imagery. Further, to verify the accuracy of this study, the administrative data of the Ministry of Maritime Affairs and Fisheries and Wando-gun was compared. This study enables the estimation of the itemized facility capacity, making it possible to ensure a systematic management of marine farms, and to serve as the basic research data and estimate the output for production statistics agencies.
C-1-3 Extracting Urban Mobility QoL Indicators and Individual Activity Pattern from Mobile Phone-based Human Mobility Trajectories
For the human-centric society and improving individual Quality of Life (QOL), extracting and analyzing key QOL indicators are crucial. In this paper, we aim to develop a model to extract key QOL mobility indicators from a city-scale low-sampling rate mobile phone GPS data. The model estimates stochastic individual activity from mobility pattern and we evaluate the result based on census and other related panel survey data. The individual key indicators can be used for spatiotemporal analysis and also useful for aggregated analysis in the different spatiotemporal units. We showcase the advantage of our model using a web-based analytic dashboard.
北米のUCGIS(University Consortium for Geographic Information Science)は、2016年にGIS&T BoK(Geographic Information Science & Technologies Body of Knowledge 2006年初版)の改定版を発行した。改定版は、WEBで提供されており、現在も更新が重ねられている。演者らは、2015年から日本の大学のGIS実習で利用可能な教材を開発してきた。本教材は、既存研究の成果に加え、近年の地理情報技術の利用事例の多様化も踏まえて構成した。本研究では、本教材の充実度を評価することを目的に、GIS&T BoKにおいて追加、修正された項目のうち、実習と関連する項目を抽出し、本教材の項目と比較した。
C-2-1 An Estimation of the Usage of Public Facilities Using Mobile Phone Data
PDF Honglei Wang, Kashiyama Takehiro, Omata Hiroshi, Yoshihide Sekimoto
The population of Nanto city is declining year by year due to the sub-replacement fertility and the aging of the population and the migration from small and medium-sized cities to metropolitan areas. Population decline has had a major impact on the social structure and economic development of the city of Nanto. The reduction of residents in Nanto leads to the idling of many communal facilities, as well as the reduction of local tax revenue and the increase of the operation and maintenance cost of public facilities. The decrease in the population has imposed a serious financial burden on the local government. To address the problem, the government of Nanto has proposed a plan to restructure public facilities in which officials prepared to sell underutilized facilities to the private sector for reducing public facilities. In order to effectively distinguish public facilities with different utilization rates and predict the number of visitors. This paper will use mobile phone data and relevant models to statistically analyze and predict the frequency and number of visitors to public facilities by local residents, so as to help the government formulate a public facility restructuring plan.
C-2-2 Multi-indicator Eco-environmental Vulnerability Assessment in Africa
Eco-environmental vulnerability evaluation plays a key role in providing useful information about ecological and environmental background for designing suitable policy measures to improve and restore the environment. Africa is the continent facing many challenges and suffers from various drastic environmental problems, including drought, deforestation, water pollution, Argo-industrial agriculture, overfishing and overgrazing, etc. Here we quantify the status of eco-environmental vulnerability in African Countries (57 countries), employing 16 indicators across six domains, including hydrometeorology, natural hazards, socioeconomics, land resources, topography, and climate change. The eco-environmental vulnerability is scaled into six levels consisting of potential, light, slight, medium, heavy, and very heavy. Results show that different African Countries have widely varying patterns of eco-environment vulnerability requiring different unique performances for environmental improvement. Among African Countries, a largest fraction of very heavy vulnerability level is attributed to Ethiopia (27.96%), followed by Madagascar, Kenya, Nigeria, Mozambique, Tanzania, and Congo (16.61%, 7.09%, 5.67%, 5.30%, 5.04%, 2.48% respectively). In general, high vulnerability levels are concentrated in Southeast Africa where are of low-income nations with high population and growing rate, and suffer from high frequencies of natural hazards. These findings help us understand the key ecological and environmental characteristics of African Continent to assist policy-makers to set improvement targets on specific areas and adopt effective practices, while keeping track of other aspects of eco-environment.
C-2-3 Drought monitoring on the Awash river basin of Ethiopia
Droughts have frequently occurred throughout Awash River Basin in recent years due to land degradation, high population density, and wetland degradation. The Awash River Basin is the most important river basin in Ethiopia covering a total area of 110,000 km2 and home of around 10.5 million inhabitants. Under the influence of climate change, the air temperature is rising and consequently affects the spatiotemporal pattern of precipitation and land use. Two widely used drought indices namely the Standardized Precipitation Index (SPI) and Normalized Difference Vegetation Index (NDVI) are chosen to address these issues. The SPI computed at different time scales indicate the different forms of drought. The monthly gridded data of Climate Hazards Group Infrared Precipitation with Stations precipitation, and Moderate Resolution Imaging Spectroscopy Terra NDVI products were used to assess the spatial and temporal meteorological drought. Furthermore, a non-parametric Mann-Kendall method was used to test for a monotonic trend at each grid level, and magnitude was estimated by the Sen’s method. Results show that the most frequent droughts occurred in the middle and lower Awash Basin during the period of analysis (2001-2018). The SPI 12 results indicate that severe hydrological drought events occurred in the middle Awash Basin during the years 2009, 2011, and 2015. Similarly, a large area of significant NDVI declining was found in those years, which coincide with the crop growing seasons. Pixel-based trend analysis revealed significant decreasing trends ( ) of rainfall and NDVI occurred in the middle and lower parts of the basin.
C-2-4 Exploring Spatial Scale by Interactive Map for Geographically Weighted Correlation
PDF Joseph Percival, Narumasa Tsutsumida, Daisuke Murakami, Takahiro Yoshida, Tomoki Nakaya
Geographically weighted correlation and partial correlation analyses are used to map spatial variations of correlations in a spatial data set. The geographically weighted approach uses a moving-window kernel running across geographical space and calculates a statistical model and summary statistics with distance-decayed weighted data. Although some bandwidth optimization approaches, such as cross-validation, have been proposed, it is important to allow geospatial statistic practitioners to easily explore the influence of spatial scale (localness) on analyses. We demonstrate our developed interactive map as a user-friendly visualization tool to analyze geographically weighted correlation and partial correlation analyses in a web browser. Census data with over 200 variables in Tokyo in 2005 were applied for the demonstration.
This paper introduces a generalized conceptual framework for evaluating eco-environmental vulnerability due to any influential variables of interest, including nature and manmade disturbances. Freely accessible remote sensing and census dataset are used to derive the driving variables with aid of Geographical Information System (GIS). Analytic Hierarchy Process (AHP) or Machine Learning (ML) can be applied to judge the importance and assign the weights of the influential variables. The designed framework is well suitable for eco-environmental assessment at a various spatial scale and a useful tool for assessing and monitoring long-term eco-environmental status due to any impacting factors. Finally, we present results of some case studies that utilize the generalized framework for eco-environmental vulnerability assessment in Vietnam, Indonesia and global scale due to nature and human disturbances.
Large-scale comparison of landforms in Japan, Korea and Taiwan using DEMs
Recently, green infrastructure (GI) is attracted as a new concept about disaster mitigation. In the study, we show the future land use policy about flood disaster mitigation, using GI, by the results of the living environment evaluation with flood risk. Target area is the Osato district in Kaiyo town, Tokushima prefecture. The district is located at the mouth of Kaifu river. We evaluated the living environment using the proximity of living facilities and the results of flood depth estimation. From the results of the evaluation, at the central area of the town, the living environment was high, but at the same area, there are many vary of the flood depth. Inland flood depth on the farmland where is the west side of the central area was high. In the future, high living environment area with high flood depth needs to be control new development. And farmland near the central area has important role to mitigate flood risk as a GI, so new development in the area is to be prohibited.
C-3-2 Development of 3-Dimensional Visualization Function of Rainfall Radar Data Using Unity 3D Terrain
This study presents the development of the function designed for 3D visualization of rainfall radar data using the terrain of the unity engine. The provided rainfall radar data is about the rainfall data provided by radars operated in Seoul. The data has an extension of tif, and a resolution of 1000*1000. In order to create a terrain, first, tif files were converted into png and RAW files. When converting png files, tif values were converted into RGB color values and stored according to the arbitrary legendary reference table. The RAW files were converted in the same way, but they are converted not into RGB but into white and black values and used. Based on the legend, the higher value was converted into whiter, and the lower value was converted into blacker. The terrain model was created in a RAW file, and using the png data converted in the Material’s Transparent Shader, the 3-dimensional rainfall radar was implemented. To ensure the utilization of the relevant data, the base map used the Google map, and the Alpha value was designated in the Terrain Material of the created radar to achieve a semi-transparent visualization of the radar. The 3-dimensional rainfall radar, created by this function, offers the strengths of enabling the user to view the data in various rotating directions and to ensure a multi-dimensional analysis. Further, since the conversion is based on the initial tif file values, it has nearly no error rate of visualized data. The function, if further improved through a survey of user requirements, will facilitate the analysis of rainfall data.
C-3-3 A Study on the Development of Visualization System for the Urban Internal Waters Flooding Using Unreal Engine
Natural disasters caused by global warming are on the rise, and water disaster damage is the biggest in Korea. To minimize these damages, accurate information delivery and prompt response are required. The purpose of this study was to develop a virtual monitoring system based on an Unreal engine that can intuitively identify the disaster situations that occur during water ingress and enable rapid decision making. The research area was targeted at Gwanghwamun area of Seoul City, which is expected to be damaged in case of flooding due to its dense facilities, and other free 3D object model files, including Unreal Marketplace, were used to establish a base map. The Leveling function was developed to intuitively check the level of the water in the built 3D base map, and Shader was developed and applied to express the water level in a walkdown. In addition, buoyancy was applied to 3D objects to achieve floating and flooding effects. The development of such a three-dimensional urban water immersion visualization system will help prevent damage efficiently by predicting the situation in the affected areas that cannot be identified due to power loss and paralysis of communication facilities in the event of domestic demand flooding and using the system as a reference for decision.
C-3-4 Mining Heatstroke Hotspot for Inbound Tourists in Tokyo
PDF Tianqi Xia, Satoshi Miyazawa, Xuan Song, Haoran Zhang, Kyoung-Sook Kim, Ryosuke Shibasaki
Summer is the popular season for inbound tourism in Tokyo during which a lot of events with be held. Nevertheless, as the city with four distinct seasons, the tourists who visited Tokyo during summer vocation are likely to suffer from heatstroke when enjoying outdoor traveling. In order to help local government improve travel quality and better prepare for the upcoming 2020 Tokyo Olympic Games, this research aims to extract tourism regions that have higher risk of heatstroke via heterogeneous data. First, the candidate POI are extracted based on the popularity of POI, then the popular routes are extracted by combining one or several neighboring POIs. With the regions generated by routes and POIs, the risk of heatstroke are evaluated by several indicators.
C-4-1 A Study on the Development of Satellite Image Color Correction Algorithm Using Open Sources
PDF Seo-I Cheon, Seung-Wook Choi, Kyeong-Hwan Choi
Image mosaicking can be handled by ENVI, Erdas Imagine, ArcGIS and other commercial software, but they are designed centering on the implementation of functions, resulting in producing very different outputs depending on the user’s software handling skills. Thus, there is a need to develop automatic image mosaicking software based on the map sheet. This study presents the development of the automatic mosaicking and color correction software for the national land satellite information under the project of establishment and operation of the satellite information utilization center, and also conducts the development of color correction algorithm. The color correction used the histogram matching method. In order to check the image stretching state of the NIR bands of N number of images in a different color distribution, the standard deviation of each NIR band was calculated, and the NIR band with the biggest standard deviation was made to become the reference for selection. The reference NIR was used to correct the color between the other NIR bands, and a merged NIR band was created through image mosaicking. The same method was applied to the color correction of the remaining R, G, and B bands. The color correction algorithm was developed based on the open source GDAL library using Python. This study targeted the low-capacity images, and the function is implemented manually. In order to boost the utilization of satellite images in the development of a color correction algorithm, the accuracy should be verified and the speed improvement and other relevant algorithms should be improved according to standard deviations, targeting the N number of large-capacity images so as to automate the function.
C-4-2 Study on the Shape of Tracks of Tropical Storms in the North West Pacific Basin
Quantitative analysis on the sinuosity of 959 tropical storm (TS) tracks is carried out utilizing four decades (1977-2016) of storm data and an established track sinuosity metric in the North West Pacific (NWP) ocean basin. Track sinuosity is the measure of deviation of a TS from its straight-line geodesic path. More sinuous track following TSs are considered to be more damaging due to less precise prediction of their movements by models. Sinuosity measurements are further categorized into four quartiles and mapped within a GIS environment. Statistical analysis is done to detect spatial and temporal trends (monthly to decadal) of track sinuosity and its relationship with fundamental parameters of TSs. Strong enhancement of storm track sinuosity is observed as we move early to late typhoon season months i.e., from the month of July to October. A short term (one- to three-year) cyclic pattern of varying track sinuosity of TSs is detected in the NWP basin. Significant longitudinal shift (from 1100–1400 E to 1300– 1600 E) in the positions of genesis points of majority of TSs is observed as the track sinuosity of TSs rises. Investigation of track sinuosity based on the warm/cold phase of the El Niño/Southern Oscillation (ENSO) reveals crucial information. The warm phase of ENSO is found to be associated with a greater number of TSs with higher sinuous tracks in the NWP basin, which are mostly accumulated in the Eastern part of the NWP basin. The study provides crucial information for disaster risk assessment, mitigation and preparedness.
C-4-3 Finding Optimal Restoration Process of Road Transportation System with Deep Q-network – in Case of Western Japan Flooding –
PDF Soohyun Joo, Yoshiki Ogawa, Yoshihide Sekimoto
Western Japan was hit by heavy rain from 28th July to 8th of June, 2018. It had caused tremendous damage to the living and industrial infrastructure. To make citizens return to their original condition as soon as possible, government has been doing recovery activity pursuant to the recovery plan. Although there are various criteria for determining recovery priority, these priority is not always a guarantee of achieving their object. They have no choice but to select one standard because it is impossible for person to find out optimal solution of complex environment using limited information. Using Deep Q-network, we could know the reward that we can get right now, as well as the sum of future rewards that expect us to get from that state we face. That is, we would be able to choose the action that maximizes that function each time, and use it to obtain the optimal policy function. In this research, our objective is to make travel time of all ODs restore to its original condition as soon as possible. Therefore, we would try to find out optimal policy of road-network’s restoration using multi-agent model with DQN for our goal.
C-4-4 Landscape and Sociodemographic Factors Affecting Urban Floods
Floods are major social and environmental concerns in many urban areas. We investigated how changes in land cover, sociodemographic conditions, and landscape factors affect flood damage in major urban areas of South Korea. Using historical maps and GIS analysis, we showed flood damages increased in those areas where rapid urbanization happened without coordinated urban planning. High flood damage areas are not spatially randomly distributed, and the hotspots of high damage areas changed over time. Additionally, human modifications of natural channels further exacerbated flood risks during the development stage and subsequent periods. This study underscores the importance of understanding the historical geographical conditions, and how humans either increased or reduced flood damage through social and technical interventions. Findings of the study have implications for resilient flood management in urban areas that are currently facing the dual challenges of land densification and climate change-induced heavy precipitation.
C-4-5 A Study on the Development of the Multi-Dimensional River Information System Designed for Providing Tailored Information for Each User
PDF Kyoung-Seop Kim, Gyeong-Min Kang, Jang-Bae Lee
Korea is currently operating diverse river space information-providing systems but they provide only 1 or 2-dimensional fragmentary information. They are also operated individually, making it difficult to acquire the information so an integrated system for the management of river information is required. This study developed the STREAM (SuiTable 3D for River Establishing Application Management) system designed to provide the 3-dimensional information tailored to users in order to develop a system aimed at the integrated management of river facilities and spatial information. To construct 3-dimensional topographic information, a DEM (digital elevation model) was created by extracting contours and elevation points from digital topographic maps and applying the TIN interpolation method. Meshes were created based on the created DEM, so a unity-based 3-dimensional topography was constructed, and high-resolution orthographic images were mapped to construct 3-dimensionmal topographic information. Extracted is the text of the existing river facilities information, river resources information, hydrophilic information, and road traffic information, and it was extracted as images using the unity 3D. The extracted information was parsed so it was developed to acquire the information wanted by the user. It will later be linked to the river facility life cycle management technology, high-precision river facility information, and standardized 3D BIM in order to provide the information on the river facility management. The developed system is expected to improve the easiness of providing the river information required by the user, and to display 3-dimensional river information so as to provide intuitive information on rivers, thus increasing the utilization of river and hydrophilic information by non-specialists.
We constructed a three-dimensional CG system to guess drawing points of paintings in real time with linking view and Digital Elevation Model, satellite images, aerial photographs, modern and old maps. Using this system, it is possible to estimate the past painters's landscape drawing points. This system configuration is as follows. First of all, the role of display is played by blender with GIS plugin which is free 3DCG software, we adjusted the position of camera, angle of view, and focal length, with watching the change of map and landscape. Although the drawing in late Edo period was expressed with exaggeration in elevation, using our system, it was confirmed that the position and shape of the mountain were accurately expressed.
<PCおよびネット環境はESRIジャパンにて用意します> データ分析が身近なものになってきているが、分析ツールに手こずったり、分析レポートが目的になっていませんか?簡単な操作でExcelより楽にグラフ表示やデータ分析が行えます。また、空間・非空間データの可視化・空間分析し、その試行プロセスを繰り返してデータが持つ特徴を最大限に導き出せます。Insights for ArcGISを使ってデータ分析してみませんか?
近年、歩道の状況や街路樹の有無など、マイクロスケールでのウォーカビリティ評価にGoogle Street Viewが有効であることが示されてきた。ただし、多くの研究は監査員による画像の評価であり、広い範囲を分析するには依然として作業コストがかかる。本研究は、ディープラーニング手法を用いてGoogle Street Viewからウォーカビリティ評価に関連するオブジェクトを自動的に抽出し、オブジェクト抽出結果と監査員によるウォーカビリティ評価結果の関連性を分析した。
近年MaaS(Mobility as a Service)の概念が進み、特に短距離交通において既存の公共交通手段に囚われない相乗り交通やシェアサイクル、キックスケーターといった多様化とそれを支えるITサービスによってシームレス化が進んでいる。こうした交通手段にはそれぞれ交通需要の密度・分布といった特性と、道路・地形などの地理的特性によって相対的な優位性が異なり、地域によって最適な交通手段の構成が存在する。特に、短距離交通手段の需要密度が高い都市部において、その類型を把握することは交通サービス体系の適正化が可能になると考えられる。 本研究では東京・神奈川の都市部を対象に、建物の密度や建物情報に属するテナント名称を用いてビルをカテゴライズする建物指標、地域の交通利便性を評価するアクセシビリティ指標、人口・世帯構成といった居住者指標、道路・地形情報を用いた類型化を行い、グループごとの交通需要パターンと比較・分析することで、交通サービス整備方針や市街地整備の評価において有用な分類指標・手法を検討することを目的とする。
SfM(Structure from Motion)写真測量とは複数の写真から被写体の形状を復元する技術です。特に、無人航空機(ドローン)による撮影と調和的です。SfM写真測量により、オルソモザイク画像(写真地図)やDSM(数値表面モデル)を作成することができます。ハンズオンでは、技術の概要と使いこなすコツを解説し、写真の撮影からオルソモザイク画像の出力までの最も基礎的な一連の作業を行います。 ノートPC(Win/mac)と電源アダプタが必要です。
ポスターセッション Poster Session Click the button to show/hide abstracts
For the human-centric society and improving individual Quality of Life (QOL), extracting and analyzing key QOL indicators are crucial. In this paper, we aim to develop a model to extract key QOL mobility indicators from a city-scale low-sampling rate mobile phone GPS data. The model estimates stochastic individual activity from mobility pattern and we evaluate the result based on census and other related panel survey data. The individual key indicators can be used for spatiotemporal analysis and also useful for aggregated analysis in the different spatiotemporal units. We showcase the advantage of our model using a web-based analytic dashboard.
P-12 二次医療圏から考えるエイズ診療拠点病院の配置
PDF 今橋 真弓・横幕 能行
【背景と目的】HIV/エイズ診療において全国どの地域でも高度な診療が受けられるように383の拠点病院が制定されている。近年、HIVと共に生きる人々(people living with HIV以下PWH)の予後改善に伴い、抗HIV療法の提供に加え、PWHの居住地域における長期療養環境整備が課題となっている。そこで、本研究では、一般の入院に係る医療を提供する基準で定められた二次医療圏別にエイズ診療拠点病院の配置について解析を行い、エイズ診療に関して拠点病院の地域における機能を検証し、今後の果たすべき役割について考察する。 【方法】定期通院者数の回答があった373施設を対象にして、拠点病院住所、提供できる医療サービス内容(できることスコア)、定期通院者数について、二次医療圏ごとに収集し、地図上に表した。二次医療圏ごとの拠点病院数、できることスコアの平均値、定期通院者数について多変量解析を行い、定期通院者数が拠点病院数、できることスコアに関連があるかどうか解析を行った。 【結果】定期通院者数は東京都区西部・大阪市・東京都区中央部・名古屋・福岡糸島・横浜西部医療圏で500人以上を超えていた。拠点病院を有する二次医療圏は192医療圏でそのうち2つ以上の拠点病院を有する医療圏は90医療圏だった。80%以上の拠点病院が提供できると回答した医療サービスには救急医療・抗HIV療法の維持・外来一般診療・入院加療があり、二次医療圏ごとの定期通院者数は二次医療圏ごとのできることスコア平均値および病院数と関連があることが示された(それぞれp=0.049、p<0.001)。 【考察】現在、二次医療圏におけるPWHの診療提供は特定施設に集中・集約しているが、今後は地域医療機関との機能分担による医療提供により、よりよいPWHのセーフティネット構築を目指す必要がある。
P-13 The Fuzzy Method for Vessel Speed Limit and Pollution Reduction on Summer in Keelung Port
When "Sustainable Development" has become the global trend as the most important topic. In the meantime, "Green Port" also has been emphasized the attention of the internal harbors. In the maritime management, government officers have gradually paid attention to the air pollution from ship and port activities. Green port program will be developed to achieve the continuous environmental protection, social and economic benefits through different management measure. This study integrates the AIS data, Marine Geography Information System (MGIS) and Fuzzy distinguish theory to propose the Optimum Speed for inbound vessels, not only to reduce the harbor air pollution and also ensure the safety of entry and exit of the port for all vessels. These seven algorithm blocks are used for the Fuzzy Logic Control (FLC) input. Including Turning Capacity (TC), Crush Stop Capacity (CSC), Wind and Current Effect (WACE), Ship Operating Capacity Index (SOCI), Environment Influence Index (EII), Ship Maneuvering Capacity Index (SMCI), Green Port Index (GPI); and one Optimum Speed (OS) output. This method provides the suggesting Optimum Speed for the captain and Vessel Traffic Service (VTS) as the reference navigation speed. Hence, while speed down for reducing the air pollution, also keeping the navigation safety and improving the port efficiency in simultaneously.
P-14 Habitat suitability empirical model of albacore tuna in the South Atlantic Ocean
This study developed an empirical habitat suitability model for identifying the optimal habitat of immature albacore tuna in the South Atlantic Ocean (SAO) by using multisatellite remote sensing data and Taiwanese longline fisheries data between 2009 and 2015. A geometric mean model with three environmental variables is suggested to be appropriate for explaining the habitat variance of albacore in the SAO. A monthly albacore mean catch per unit effort (CPUE) of larger than 50 individuals/1,000 hooks occurred from April to August and was distributed over 25–40°S. The optimal range of environmental variables in the sea surface temperature, sea surface salinity, and sea surface chlorophyll-a concentration for the habitat of albacore are suggested to be approximately 17–19°C, 34.6-35 psu, and 0.33–0.41 mg/m3, respectively. We discovered a significant positive relationship between the suitable habitat area and nominal CPUE of immature albacore in the SAO (r = 0.912). Overall, frontal structures play a major role in the stimulation of potential albacore habitats.
Developments of city give a disproportionate impact on the decline of vegetation cover, causing surface temperature to rise. Prior to explore the importance of urban green cover for UHI mitigation, this study aims to analyze the relationship between land cover fraction and land surface temperature. We estimate Jakarta comprises of 69% of impervious, 23% of vegetation, and 6% of soil. The mean LST was 33.12oC with spatial variations associated with vegetation cover (r2=0.75).
P-25 Habitat suitability model of Uroteuthis edulis using remote sensing data and GIS in the South East China Sea
Squid (Uroteuthis edulis) was one of the important commercial species of the Taiwanese Torch-lighted fishery in the northern waters, Taiwan. The modeled oceanographic data of sea surface temperature (SST), sea surface height (SSH), sea surface salinity (SSS), mix layer depth (MLD), current intensity (current), and satellite-based oceanographic data sea surface chlorophyll-a (SSC) with catch data collecting from Taiwanese Torch-lighted fisheries were simultaneously collected during 2015-2017. The Generalized Additive Models (GAMs) and maximum entropy models (MaxEnt) was used to evaluate the effects of oceanographic conditions on the formation of potential fishing zones for Squid and then to explore the spatial variability of these features in the northern waters, Taiwan. The results from the habitat suitability index Model revealed its potential for predicting the spatial distribution of squid and highlighted the use of multispectral satellite images for describing potential fishing zones. The optimal range of hydrological variables in SST, SSC, SSS, SSH, MLD, and CUR, respectively, for the habitat of Squid were 26-28°C, 0.09-1 mg m-3, 34.4-35.8 psu, 0.5-0.7 m, 10-20 m, and 0.2-0.4 m/s. The spatial potential fishing zone patterns were explained predominantly by SSC is the most influential factor of geographic distribution in boreal dominant season (May to October). The geographic information system maps of fishing period of the predicted HSI values were overlapped by the observed CPUE in 2018, suggesting that the model can be used as a tool for reliable prediction of potential fishing grounds with the development of management regulations.
近年MaaS(Mobility as a Service)の概念が進み、特に短距離交通において既存の公共交通手段に囚われない相乗り交通やシェアサイクル、キックスケーターといった多様化とそれを支えるITサービスによってシームレス化が進んでいる。こうした交通手段にはそれぞれ交通需要の密度・分布といった特性と、道路・地形などの地理的特性によって相対的な優位性が異なり、地域によって最適な交通手段の構成が存在する。特に、短距離交通手段の需要密度が高い都市部において、その類型を把握することは交通サービス体系の適正化が可能になると考えられる。本研究では東京・神奈川の都市部を対象に、建物の密度や建物情報に属するテナント名称を用いてビルをカテゴライズする建物指標、地域の交通利便性を評価するアクセシビリティ指標、人口・世帯構成といった居住者指標、道路・地形情報を用いた類型化を行い、グループごとの交通需要パターンと比較・分析することで、交通サービス整備方針や市街地整備の評価において有用な分類指標・手法を検討することを目的とする。
Mountainous regions, surrounded by various natural environments, are sensitive to anthropogenic disturbances. This study aims to investigate recent changes in built-up areas and population in Liping county, a typical mountainous region in China since 2000 by Landsat time series and township-level census data. Overall, we found that built-up areas have continuously expanded whereas population has not increased, which may imply excessive human impacts on this region.
Recently, unmanned aerial vehicles (UAVs) are wildly used in the field of aerial photogrammetry. For a photogrammatic application, it is important to select a suitable software in order to build a reliability of the digital 3D models that are created from images from UAV. In this research we focused on comparison of three software solutions for processing UAV images and analysis of their influence on the quality of the 3D models. Two test sits, river side and landslide areas are chosen, and three software packages, Pix4D mapper, Agisoft PhotoScan and Context Capture are used in this comparison. The accuracy of each of the software packages was compared by quantifying the error in the control points and check points between the model and the field survey. The result shows that different software has its advantages and disadvantages depend on applications.
P-50 Characteristic Analysis of GNSS Real-Time Kinematic for Mobile Platforms
The Real-time Kinematic (RTK) GNSS positioning has predominantly been applied to short baseline applications due to the spatial decorrelation of the measurements between the reference station and the rover. The performance of the solutions was investigated for two types of RTKs; a single-baseline RTK (SRTK) and the network-based RTK (NRTK). The characteristics of RTK were analyzed in this study, especially with the baseline lengths between two receivers. The accuracy and stability of the RTK are highly dependent on the success rates of the ambiguity resolution, which is considerably stable up to 50km of baselines. The network-based RTK was conducted for three currently operating correction information, that is, Virtual Reference Station (VRS), Flächen-Korrektur-Parameter (FKP), and the State Space Representation (SSR). The initialization time and the stability were inspected for both static and kinematic environments, and the accuracy was analyzed by the reference coordinates obtained from the Mobile Mapping System (MMS). The NRTK shows a better performance in terms of success rate of ambiguity resolution. The actual VRS is generated within about 2 m from each CORS, and the amount of horizontal error corresponds to the bias of VRS locations. The SSR correction generally shows a similar performance as VRS/FKP, but the NRTK positioning was unstable during the period with rising satellites.
P-51 Influence of uncertain address information on spatial statistical analysis
PDF Ikuho Yamada
Due to privacy and confidentiality concerns, it is not unusual that address information of individuals that are available to researchers contains certain ambiguity and/or incorrectness. Data providers may aggregate address information based on administrative boundaries, and survey respondents, for example, may hesitate to fill in their precise address in a questionnaire survey. This study examines how such uncertainty in address information may influence results of spatial statistical analysis through simulation-based experiments using Moran’s I statistic as an example. Both synthetic data and actual survey results are utilized in the experiments.
P-56 Realtime Monitoring of Dangerous Slopes based on 3D Maps
PDF Jin Duk LEE, Ki Tae CHANG, Kon-Joon BHANG, Tae Kyung BAEK
For predicting the risk of collapse and inducing the evacuating inhabitants around dangerous slopes, the realtime slope monitoring & management system is needed to be constructed. The functions which are needed for meeting the requirements of the system, that is basic function, monitoring function, search function, spatial analysis function and site management function and so forth, were analyzed and reflected in screen design of the Web-EOC(emergency operating center). To grasping intuitionally current status of slope areas, the Web-EOC slope monitoring & management system was constructed based on 3D GIS and 3D maps called Vworld, which has been operated by the Ministry of Land, Infrastructure and Transport, Korea. And then application experiments was carried out through a testbed installation.
Recently, green infrastructure (GI) is attracted as a new concept about disaster mitigation. In the study, we show the future land use policy about flood disaster mitigation, using GI, by the results of the living environment evaluation with flood risk. Target area is the Osato district in Kaiyo town, Tokushima prefecture. The district is located at the mouth of Kaifu river. We evaluated the living environment using the proximity of living facilities and the results of flood depth estimation. From the results of the evaluation, at the central area of the town, the living environment was high, but at the same area, there are many vary of the flood depth. Inland flood depth on the farmland where is the west side of the central area was high. In the future, high living environment area with high flood depth needs to be control new development. And farmland near the central area has important role to mitigate flood risk as a GI, so new development in the area is to be prohibited.