OR/14/064 EO information services to be delivered
| Kessler, H, and Dearden, R. 2014. Scoping study for a Pan-European geological data infrastructure: D3.4: technical requirements for serving 3D geological models. British Geological Survey, OR/14/072. |
 |
Service specifications
This section addresses the EO service specifications in so far as they can be identified prior to direct user access, and before all of the datasets have been acquired and subsequently assessed for quality and accuracy. Where possible, the thematic content, resolution, coverage, timeliness, accuracy and format of the products/services are outlined in Table 5.
| Service | Service coverage | Thematic content | Resolution | Timeliness | Format | Comment |
| Land use/land cover mapping | St. Vincent and the Grenadines, St. Lucia and Grenada | i) Landcover — we shall seek to capture vegetation, sealed surfaces, basic road network & water bodies.
ii) Water features — rivers, streams, lakes, water bodies, watersheds. |
Spatial resolution will be typically 1 pixel. This cannot be ascertained until the imagery is delivered and its quality is assessed | The EO data shall not be older than 3 years, although older imagery will be used for context | Raster — land use/land cover. Vector — water features |
Limited ground checking will be undertaken whilst validating the landslide inventory. 80% of the area of interest will be covered. |
| Hazard mapping to support landslide risk assessment | Grenada St. Lucia | i) Landslide inventory map of Grenada at 1:20 000, and St. Lucia at 1:20 000 with key areas (no more than 50%) at 1:10 000.
ii) DEM generated using ASTER |
i) Landslide inventory at 1:10 000 to 1:20 000 scale.
ii) DEM at 30 m horizontal resolution & vert. accuracy 5–10 m. |
Inventory EO data will not be older than 2010. | Inventory — vector DEM — raster grid |
Multiple years of EO will be used for the landslide inventory to provide context of active and non-active features. Ground truthing will be on a sample basis. |
| Digital Elevation Model (DEM) | Belize (or parts thereof) | i) High resolution DEM over all of Belize ii) Precise DEM over a limited area |
i) High resolution DEM will be 30 m for at least 80% of Belize, and 20 m for 8952 km2. ii) Precise DEM will be in the order of ~1 m. We cannot be certain until the data are acquired and assessed. |
i) 30 m DEM derived from 2010 imagery. ii) 20 m DEM from SPOT data from within 5 years of present. |
Raster grids | Nationwide DEM will cover at least 80% of the territory. The EO data for the precise DEM have been tasked for acquisition by Pleiades in triplet mode — this acquisition is dependent on cloud-free conditions. |
The format of the products will be delivered in the most appropriate mode for that product, which is sometimes defined in the SoW e.g. it states that the water features of Service 1 will be in vector format. In general, all products will be delivered in digital format, although analogues copies will be delivered at least to comply with ESA specifications. The data will be readily ingestible into standard GIS. Additional formats, such as 3D PDFs could be produced if requested by the users as they have been found in other projects to increase understanding and usage of 3D EO data.
Regarding the thematic content of the landslide inventory, BGS has defined the following attributes (Table 6) and iterated them with ITC to ensure that both organisations will use compatible attributes when producing their inventory. Not all of the attributes will be completed, depending on the quality of the imagery and the terrain conditions.
| # | Description | Identifier | Type | Dimension | Comments |
| 1 | FID | ||||
| 2 | Shape | text | 10 | polygon | |
| 3 | Id | ||||
| 4 | Landslide ID | LID | number | 7 | polygon identifier that can be related to landslide database entry point |
| 5 | Location District | DISTR | text | 30 | district name |
| 6 | Location Locale | LOCAL | text | 30 | locality name |
| 7 | Movement type | TYPE | code | 2 | … (not entered), FL (flow), SR (rotational slide), SP (planar slide), SU (undifferentiated slide), FA (fall), TO (topple), SP (spread), UN (undefined) |
| 8 | Morphology | MORPH | code | 1 | L, S, T, A (Landslide undifferentiated, Scarp, Transport zone, Accumulation zone) |
| 9 | confidence | CONF | code | 1 | H, M, L (High, Medium, Low) |
| 10 | 2010 | 2010 | code | 1 | N, I, A — (Not present — No slide visible), (Inactive — The slide can be recognized but no activity visible, in the form of disrupted vegetation or bare surface), A (Active — Slide shows clear signs of recent activity in the form of bare surfaces, disrupted vegetation etc) |
| 11 | 2011 | 2011 | code | 1 | N, I, A — (Not present — No slide visible), (Inactive — The slide can be recognized but no activity visible, in the form of disrupted vegetation or bare surface), A (Active — Slide shows clear signs of recent activity in the form of bare surfaces, disrupted vegetation etc) |
| 12 | 2012 | 2012 | code | 1 | N, I, A — (Not present — No slide visible), (Inactive — The slide can be recognized but no activity visible, in the form of disrupted vegetation or bare surface), A (Active — Slide shows clear signs of recent activity in the form of bare surfaces, disrupted vegetation etc) |
| 13 | 2013 | 2013 | code | 1 | N, I, A — (Not present — No slide visible), (Inactive — The slide can be recognized but no activity visible, in the form of disrupted vegetation or bare surface), A (Active — Slide shows clear signs of recent activity in the form of bare surfaces, disrupted vegetation etc) |
| 14 | 2014 | 2014 | code | 1 | N, I, A — (Not present — No slide visible), (Inactive — The slide can be recognized but no activity visible, in the form of disrupted vegetation or bare surface), A (Active — Slide shows clear signs of recent activity in the form of bare surfaces, disrupted vegetation etc) |
| 15 | FIELD CHECK | FIELD | TEXT | 50 | free text |
The coverage of the precise DEM i.e. the area that has been tasked for Pleiades triplet acquisition is illustrated in Figure 2. This area was agreed in discussion with the WB. Acquisition of imagery is highly dependent on cloud-free conditions.
Relevance to user requirements
The hazard characteristics row of Table 1 summarises the hazards that are of importance to the users of the services in each territory, while Table 3 lists the current geospatial information sources that are available to address those characteristics. This is the situation that EO-RISC endeavours to address. Additional information regarding user requirements and the existing geospatial information sources may be known by the CHARIM project, and BGS will have the opportunity to liaise with the user community at the CHARIM meeting in late September 2014.
The products/services delivered by EO-RISC will certainly be of benefit in their own rights, Cecy Castillo (Science Dept. Chair at the University of Belize) commented that they ‘will definitely be able to make use of these datasets’ and that the project is an ‘exciting development’ while Carren Williams (Principal Land Information Officer, Department of Lands and Surveys, Belize) stated that ‘the information will be very beneficial for the country’. However, the EO-RISC project will dovetail with CHARIM to produce the Caribbean Handbook for Risk Information Management with BGS staff already identified to contribute to several sections. This handbook will help to build capacity of governments in the region to generate landslide hazard and risk information, and to apply this in disaster risk reduction use cases. The target audience of the handbook is technical staff from government organisations and private consultants, representatives from government sectors such as Ministries of Planning and Public Works, representatives of National Disaster Management Organisations, representatives from NGO’s dealing with risk reduction, representatives from communities and World Bank consultants.
Advantages/benefits
It is too early in the project to provide detailed comments on the advantages/benefits of the EO Information services to be delivered — these will be finalised in D3 Operational Documentation. Initial comments can confirm that the general advantages of EO data hold true in this geographic region:
- EO data are remotely derived i.e. non-intrusive;
- The imagery and the processing/interpretative methodology is generally objective and consistent for other areas of similar terrain and data availability;
- Archive data can be accessed to provide historic data (i.e. a baseline). This is particularly important for dynamic features such as landslides where they may appear intermittently in the record, but the combination of their appearances in history (and currently) enables models to be constructed regarding possible future occurrences;
- Programmes such as GEO and Copernicus support long term image acquisition, therefore the methodology used in EO-RISC can be implemented for monitoring programmes within the eoworld initiative and beyond.
- Projects elsewhere (e.g. EO-MINERS; www.eo-miners.eu) have shown that products developed primarily from EO data sources can provide authoritative and objective bases for discussion between stakeholders interested in environmental conditions.
Limitations/constraints
It is too early in the project to provide detailed comments on the limitations/constraints of the EO Information services to be delivered — these will also be finalised in D3 Operational Documentation. Nevertheless, as with every EO-based project, we are constrained by data availability and cost. In this situation there is not a suitable archive of radar data for InSAR analysis to be undertaken, so we are limited to optical data and the associated techniques.
Pleiades very high resolution data (including triplets for Belize) have been tasked, however the time constraints of this project mean that it is possible that these data will not be acquired as the climate at this time of year generally results in significant cloud cover.
The EO services lack some input from the user community due to the postponement of the meeting at Washington DC. This has meant that EO-RISC is dependent upon feedback from other projects until BGS can liaise directly with the users in late September 2014.
Guidelines for use
It is too early to provide guidelines on the utilisation of the EO Services. These will be finalised in D3 Operational Documentation. However, it is recommended that we take advantage of the WB CHARIM project which has a larger remit in the area. The input of EO data will be advertised and promoted via the Caribbean Handbook for Risk Information Management. BGS will also working directly with local users (where resources permit) to embed and stimulate the use of the EO-RISC services as appropriate.
 |