METEOR Explorer - Building Exposure Map of Nepal

Publication date

Nov 02, 2020

Version

2020-02-15

License

Creative Commons BY-NC-SA 4.0 (CC BY-NC-SA 4.0)

Preliminary Exposure Map for Nepal, by ImageCat.

Click on the map to view a distribution of costs by construction type.

This data was collected for Nepal as part of the Modelling Exposure Through Earth Observation Routines (METEOR) project and is a Level 3, regionally enhanced data set. The purpose of collecting the data is CAT modeling and loss estimation. Modeling potential exposure helps mitigation and response planning by identifying regions of high risk or greater exposure before a natural disaster occurs. Repurposing for any reason other than assessing risk is not recommended. The data presents the estimated number of buildings, building area, and rebuilding value at a 3-arcsecond grid resolution for Nepal (approximately 90 meters, depending on where on earth). The results were created through a process of spreading the number of buildings from census data, at the VDC census level, to the 3 arcsecond level by a statistical assessment of moderate resolution EO data; which is described in more detail below. The estimated building count at any given area is a result of statistical processes and should not be mistaken as a building count. The structural classes of buildings used for risk assessment are estimated given the building wall and roof material classes surveyed through the census. Structural classes for urbanized areas such as Kathmandu were developed through field observations using stratified sampling and engineer and local expert verification.

Data Quality - Lineage

Dasymetric Mapping Evaluation:

Visual review and adjustment by 3 different analysts not involved in the process- junior level, mid-level, and senior level. Comparison with several dasymetrically mapped products, including Landscan and WorldPop, review in Google Earth with particular attention to mountainous areas, review against SAR, GUF, and other products used for dasymetric mapping.

Structural Distribution Evaluation:

Structural Distribution was developed through the implantation of the stratified sampling method by Kathmandu Living Labs (KLL), and reviewed by GEM, ImageCat, and NSET. Several changes to structural classes were made. In addition, the structural mapping from the VDC census data or IPUMS data was validated through mapping the structural classes and review of the results with NSET and ImageCat. The distribution of structural classes closely reflected known patterns of construction practices with respect to elevation, proximity to India, and urbanity.

Persons per household Evaluation:

Given detailed census level data available, no additional validation was performed.

Development Pattern Evaluation:

The development patterns were reviewed by 5 different parties. In general, development patterns were not as distinct in Nepal as in many countries, and the structural mapping did not vary as significantly by the development patterns as with region, which was captured in detail by the census and validated by NSET.

Building Height and Area Evaluation:

Building height and size were generated using highly detailed micro census data from IPUMS, HOTOSM building height samples, and OSM building footprints from selected regions. The resulting data was reviewed spatially on maps by NSET and ImageCat. At a grid level, the estimates were further compared with the number of structures digitized in the OSM database as a “floor.” In outer areas, particularly in the Western districts, there were more structures than could be accounted for in the census data. It is unknown whether this is due to recent development or bias in the census data. No corrective measures were taken.

Replacement costs Evaluation:

Replacement costs were provided by NSET. Given the local source of data and the recent rebuilding costs after the 2015 Gorkha earthquake, no additional validation was performed.

Processing Step Descriptions:

Dasymetric Mapping (figure 3.2)

Dasymetric mapping is the process of spreading the number of buildings from census data, at the village development committee (VDC) census unit level to the 3-arcsecond level, by a statistical assessment of moderate resolution Earth Observation (EO) data. To collect EO indicators of settlements and the density of buildings, various remote sensing data sets were used. These included:

NOAA night-time light annual composite (VIIRS) resampled from the 15-arcsecond grid cell to 3-arcsecond
Oak Ridge National Laboratory Landscan ambient population (LSCAN) resampled from the 15-arcsecond grid cell to 3-arcsecond
JRC Global Human Settlement Layer (GHSL-Landsat) derived from Landsat imagery resampled from the 15-arcsecond grid cell to 3-arcsecond
DLR Global Urban Footprint (GUF) resampled from 12-meter grid cells to a 3-arcsecond percentage of human presence raster grid
JRC GHSL derived from Sentinel-1 SAR (GHSL-SAR) resampled from 20-meter grid cells to a 3-arcsecond percentage of human presence raster grid
CIESIN-Facebook High Resolution Settlement Layer (HRSL) resampled from 1-arcsecond grid cells to a 3-arcsecond percentage of human presence raster grid
JRC mosaiced Sentinel-1 dual polarization bands (SAR B1, B2, B3)resampled from 20-meter resolution to a 3-arcsecond mean raster grid per band and a maximum mean value of the 3 bands (SAR-MaxMean).
An indicator of OSM data throughout the country- building count and maximum building height were calculated from building footprint polygons and aggregated up to create 3-arcsecond grids.

These remotely sensed earth observation products and building footprint aggregates establish the distribution statistics for dispersing buildings by urban density and determining the development pattern type throughout the country. Each of the EO products individually, and in a combination, act as weights to disperse the known number of households per VDC by structural type and development pattern to 3-arcsecond grid cells. For example, with development patterns 1 or 2 (resembling rural or single-family residential communities) the even building distribution of the VDC is reallocated only to grid cells within the VDC associated to human settlement. For determining the weights for distribution within a given VDC tract, several machine learning algorithms were run using the EO to develop a prediction model. In the case of Nepal with complex terrain, GUF was highly weighted, and support vector was determined to be the most effective AI tool, as determined by visual inspection. For developments with higher populations or building density, the reallocation of buildings becomes more complex and requires a more detailed examination of the structural types and mapping schemes.

Masking process: To prevent unpopulated areas from being considered as settlements, especially in the highly mountainous terrain, a mask was created by combining the extents of night-time light (VIIRS), ambient population (LSCAN), and GHSL-Landsat that have been reclassified to inhabited vs. uninhabited using a minimum threshold value determined by visual inspection. These minimum values correspond to even the sparsest human presence. This mask was used to subset the high-resolution Global Urban Footprint (GUF), Sentinel-1 SAR based GHSL product, Sentinel-1 mosaic with dual polarization (Band1 = VH, Band2= linear comparison of VH/VV, Band3 = VV), and CIESIN-Facebook High Resolution Settlement Layer (HRSL) that go into the machine learning process to come up with the development patterns; subsetting the data sets decreases the processing time.

Development Pattern Creation: Development patterns are patterns of construction in a given country that typify the building structure development and density as much as possible. They sometimes correspond with land use, but not always. The development patterns are determined by a structural engineer working with GIS analysts to conduct a web reconnaissance exercise using Google Earth, and structural distribution web searches to characterize the urbanity density and development patterns for each country. For Nepal, the ImageCat engineer characterized 8 development pattern types (see below Development Pattern Descriptions).

To characterize building density in more populated areas, analysts digitized sample training development patterns polygons in the top 10 most populated cities in the country. The basemap vintage and source used during digitization vary by region and zoom level. However, the most current high-resolution satellite images are used. The training polygons and the moderate resolution EO products described above are used in a machine learning process (CART algorithm, Random forest, and Support Vector) for assigning the development patterns throughout the country, which informs the estimated building density. The intensity of urbanity correlates to both the building density and the structural distribution (see Structural Distribution Processing Step for more details).

Development Pattern Descriptions:

Development Pattern 1: Rural development found outside of city boundaries and is typically associated with agricultural development. The regions typically consist of small, remote villages with single roads in and out. Buildings are typically spaced far apart and are almost exclusively 1 to 2 stories. Local materials and construction practices are generally used and performed in these areas.

Development Pattern 2: This development pattern reflects areas typically dominated by single family residential structures. Commercial properties, such as local markets, are present, however residential structures are the primary occupancy. The built-up area is denser than rural class 1, however open land (yards, vacant lots, etc.) are present and can be observed via satellite imagery. All structures are low-rise, with most in the 1 to 2 story range.

Development Pattern 3: This development pattern is representative of regions with dense residential and commercial development. Apartments are typically located above first floor commercial properties. Structures are predominantly low to mid-rise, with an occasional high-rise structure located within the development pattern. Buildings are tightly spaced.

Development Pattern 4: This development pattern is typically associated with extremely dense, informal settlements. They are usually found within boundaries of large cities, and are typically comprised of very small (<100 m^2) standalone structures with little to no space between adjacent buildings. The settlement is unplanned, therefore there is no organization to the configuration of building layouts. Almost all structures are 1-story, and are typically erected using cheap and accessible local materials.

Development Pattern 5: Development pattern 5 is characterized by urban areas predominantly occupied by low to mid-rise residential and commercial structures. An occasional high-rise apartment or office building may be present. These developments are typically found near or around major city centers. Buildings are tightly spaced and are fairly regular in shape.

Development Pattern 6: This development pattern is the central business district of urban areas within the major cities. The region is occupied by low to high-rise apartments and commercial offices. Most structures are under 7-stories, however high-rise (8+ stories) can be found within the region. Building footprints are larger than most non-industrial development patterns. This development pattern will be found only in major cities and along the major, paved roads.

Development Pattern 7: This development pattern is characterized by areas dominated by ports, mining or industrial activities. Structures are typically closely spaced and regular in shape. A majority of buildings within these regions are warehouses, rectangular shape and single story. Smaller low-rise, office and commercial structures can also be found on site.

Development Pattern 8: This development pattern is typically located within the urban region and is comprised of large developments, such as universities. The built-up environment is typically comprised of low to mid-rise structures with large building footprints.

Number of Buildings (figure 3.3):

The total number of buildings per structural class, per 3 arcsecond grid was inferred using a combination of IPUMS, Nepal census data, OSM survey data, and aggregated OSM building raster data sets (as mentioned in dasymetric mapping). For the rural development pattern type the height distribution from IPUMS, specifically heights for unreinforced brick masonry UFB-1 and UFB-5, was used with the average building footprint area from 3-arcsecond aggregated OSM raster to determine the average total building area. Using the Nepal census count of household per village development committee (VDC) by wall material type a relationship between the average total building area using structural type was created to calculate total number of buildings for the rural zones. For all other non-rural development pattern designated area the Humanitarian OpenStreetMap Team (HOTOSM) in-situ building survey (sampling strategy discussed in Structural Distribution section) was used to establish the relationship between the structural type and building story height by development pattern to determine total number of buildings.

The VDC-level census spreadsheets and GIS data were provided by Sharad Wagle from the National Society for Earthquake Technology (NSET) in Nepal.

Structural Distribution (figure 3.4):

A structural engineer conducts a web reconnaissance survey of available data regarding the structural type and distribution of the country. Various engineering websites and standards (World Housing Encyclopedia [WHE], Prompt Assessment of Global Earthquakes for Response [PAGER], Global Earthquake Model [GEM]) were used to establish a preliminary structure type distribution. These preliminary distributions are validated through in-country building code sources (Nepal National Building Code) and Google StreetView survey.

After the web reconnaissance the structural engineer begins to formulate the mapping scheme by development pattern. Using the 2011 Nepal VDC-level census data the rural development pattern was assigned a mapping scheme using the building wall material type and number of household value. The building height values for UFB-1 and UFB-5 within the rural development pattern zones were obtained from IPUMS data set. For the remaining non-rural development patterns, the field survey data collected by HOTOSM was used to establish the mapping scheme for each of those development pattern types.

The data below provides a description of the method of inferring structural class from the census data, used in non-urban areas, for each GEM taxonomy classification.

C99/LFINF+DNO/HBET:1,3 - Mapped to structures with walls identified as "Cement bonded bricks/stone". The amount apportioned to reinforced concrete buildings is calculated as the percent of structures with “RCC with pillar” foundations from the “Cement bonded bricks/stone” + “RCC with pillar” foundations count.

C99/LFINF+DNO/HBET:4,7 - Mapped to structures with walls identified as "Cement bonded bricks/stone". The amount apportioned to reinforced concrete buildings is calculated as the percent of structures with “RCC with pillar” foundations from the “Cement bonded bricks/stone” + “RCC with pillar” foundations count.

C99/LFINF+DNO/HBET:8,20 - Mapped to structures with walls identified as "Cement bonded bricks/stone". The amount apportioned to reinforced concrete buildings is calculated as the percent of structures with “RCC with pillar” foundations from the “Cement bonded bricks/stone” + “RCC with pillar” foundations count.

MUR+ADO/HBET:1,3 - Directly mapped to structures with walls identified as "unbaked brick".

MUR+CL99+MOC - Mapped to structures with walls identified as "Cement bonded bricks/stone". The amount apportioned to “cement bonded bricks” is calculated as the ratio of structures with foundations of “Cement bonded bricks/stone” of the total number of structures with foundations of “Cement bonded bricks/stone” + “RCC with pillar." To distinguish between “Cement bonded bricks” and “Cement bonded stone”, roof data was incorporated. Buildings with “Thatch/straw” and “galvanized iron” roofs and "Cement bonded bricks/stone" walls are classified as stone masonry. Buildings with “Tile/slate”, “RCC”, “Wood/planks” and “Mud” roofs and "Cement bonded bricks/stone" walls are classified as fired brick masonry.

MUR+CL99+MOM - Mapped to structures with walls identified as "Mud bonded bricks/stone”. To distinguish between “Mud bonded bricks” and “Mud bonded stone”, roof data was incorporated. Buildings with “Thatch/straw” and “galvanized iron” roofs and "Cement bonded bricks/stone" walls are classified as stone masonry. Buildings with “Tile/slate”, “RCC”, “Wood/planks” and “Mud” roofs and "Cement bonded bricks/stone" walls are classified as fired brick masonry.

MUR+STRUB+MOL - Mapped to structures with walls identified as "Cement bonded bricks/stone". The amount apportioned to “cement bonded bricks” is calculated as the percent of structures with “Cement bonded bricks/stone” foundations from the “Cement bonded bricks/stone” + “RCC with pillar” foundations count. To distinguish between “Cement bonded bricks” and “Cement bonded stone”, roof data was incorporated. Buildings with “Thatch/straw” and “galvanized iron” roofs and "Cement bonded bricks/stone" walls are classified as stone masonry. Buildings with “Tile/slate”, “RCC”, “Wood/planks” and “Mud” roofs and "Cement bonded bricks/stone" walls are classified as fired brick masonry.

MUR+STRUB+MOM - Mapped to structures with walls identified as "Mud bonded bricks/stone”. To distinguish between “Mud bonded bricks” and “Mud bonded stone”, roof data was incorporated. Buildings with “Thatch/straw” and “galvanized iron” roofs and "Cement bonded bricks/stone" walls are classified as stone masonry. Buildings with “Tile/slate”, “RCC”, “Wood/planks” and “Mud” roofs and "Cement bonded bricks/stone" walls are classified as fired brick masonry.

W+WWD - Directly mapped to structures with walls identified as "Wood/planks" and “Bamboo”.

The HOTOSM sampling strategy followed the stratified sampling and Bayesian updating approach suggested in: Porter K., Z. Hu, C. Huyck and J. Bevington (2014), User guide: Field sampling strategies for estimating building inventories, GEM Technical Report 2014-02 V1.0.0, 42 pp., GEM Foundation, Pavia, Italy, doi: 10.13117/GEM.DATA-CAPTURE.TR2014.02.

The HOTOSM building survey data was verified by a local in-country engineer (Sharad Wagle) and in-house engineer (Michael Eguchi). All of the mapping schemes are then mapped to the PAGER standard structural types. These structure types are overlaid with the manually delineated development pattern sample polygons to create a refined mapping scheme. A final round of sanity checking is conducted by ImageCat engineers.

The mapping schemes below provide the structural mapping scheme for each development pattern. The structure type key can be found below.

Development Pattern 4: C99/LFINF+DNO/HBET:1,3 - 0.0769|C99/LFINF+DNO/HBET:4,7 - 0.044|MATO/LN - 0.2308 -MUR+CL99/HBET:1,3 - 0.5385|W - 0.0549|W+WWD - 0.0549

Development Pattern 6: C99/LFINF+DNO/HBET:8,20 - 0.0773|C99/LFINF+DNO/HBET:1,3 - 0.1701|C99/LFINF+DNO/HBET:4,7 - 0.4485 - S/LFINF - 0.0258|MUR+CL99/HBET:1,3 - 0.1443|MUR+CL99/HBET:4,7 - 0.134

Building Height (figure 3.5):

The building height values for rural development pattern zones, specifically to UFB-1 and UFB-5, were obtained from IPUMS data set. For the remaining non-rural development patterns, the field survey data collected by HOTOSM was used to establish the building height by structure type per development pattern type.

Total Building Area (figure 3.6):

The total building was calculated using a combination of IPUMS, Nepal census data, OSM survey data, and 3-arcsecond aggregated OSM building raster data sets. For the rural development pattern type the height distribution from IPUMS, specifically heights for unreinforced brick masonry UFB-1 and UFB-5, was used with the average building footprint area from 3-arcsecond aggregated OSM raster to determine the average total building area. For all other non-rural development pattern designated zones the Humanitarian OpenStreetMap Team (HOTOSM) in-situ building survey was used to establish the total building area by structure type per development pattern type using the surveyed building footprint and height values.

Replacement Cost (figure 3.7):

The replacement cost values were provided by partner NSET which are derived by an in-country survey in the surrounding urban and rural regions of Kathmandu. Below are the average replacement cost values in USD:

Adobe - $9.00

Reinforced Concrete Framed Building (RCC) with infill - $27.00

Engineered Reinforced Concrete Framed Building (RCC) - $32.00

Informal - $5.00

Light Steel Building - $16.00

Steel Frame Building w/ URM Infill - $24.00

Load Bearing Brick and Mud Mortar Masonry - $12.22

Load Bearing Brick and Cement Mortar Masonry - $19.74

Load Bearing Stone and Mud Mortar Masonry - $9.40

Load Bearing Stone and Cement Mortar Masonry - $10.34

Wood and Timber - $8.00

General information

Publication date

Version

License

Abstract