Refine your search: Clear all
Featured:
Locations:
More
Formats:
More
Organisations:
More
Tags:
More
Licenses:
More
  • 600+ Downloads
    Updated 10 March 2017 | Dataset date: March 08, 2017-March 08, 2017
    This dataset updates: Never
    Please note that the windspeed and track dataset only covers the part where the this was still a tropical cyclone. For explanation see below caveats. Due to this we will not release a priority index, since windspeed data is missing for most of the country. Dataset of windspeed and track was kindly provided by University College London. The rainfall data is calculated based on GPM. It is the accumulated rainfall from March 6th midnight to March 10th 10:00am Madagascar time.
  • 300+ Downloads
    Updated 1 March 2017 | Dataset date: October 07, 2016-October 07, 2016
    This dataset updates: Never
    Partial Assessment and 3W matrix ,Hurricane Matthew. as of 07.10.2016
  • 800+ Downloads
    Updated 18 February 2017 | Dataset date: February 17, 2017-February 17, 2017
    This dataset updates: Never
    Dataset contains windspeeds in miles per hour.
  • 30+ Downloads
    Updated 12 December 2016 | Dataset date: November 24, 2016-November 24, 2016
    This dataset updates: Never
    This report describes preliminary building damage analysis carried out by UNITAR-UNOSAT covering Area 1 (Jeremie and Roseaux Commune), Area 2 (Abricot, Dame-Marie, Anse d'Hainaults and Les Irois Communes), Area 3 (Corail, Pestel, Beaumont and Roseaux Communes), and Area 4 (Tiburon Commune) for a total area of approximately 1,200 Km2. Building damage analysis, including a rapid assessment of transportation network conditions and locations of spontaneous people gathering sites, was conducted by comparing the post-disaster satellite images (Pleiades acquired on 7/10/2016 for AOI1, Pleiades acquired on 12/10/2016 for AOI2, Pleiades acquired on 09/11/2016 for AOI3 and Worldview-2 acquired on 9/10/2016 & 17/10/2016 for AOI4) with available pre-disaster images (WorldView-1 on 08/12/2014, 01/05/2015, 09/05/2015 and 16/06/2015; Worldview-2 on 17/07/2016, 28/11/2014 and 14/06/2015 and Worldview-3 on 17/10/2015). UNOSAT's preliminary analysis shows a total of 40,696 buildings/structures with visible damages and approximately 508 locations with visible road obstacles and/or access constraints. In addition, 1,497 temporary people gathering sites have been identified within the analysed areas (Area 1, Area 2, Area 3 and Area 4).
  • 2200+ Downloads
    Updated 7 December 2016 | Dataset date: November 22, 2016-November 22, 2016
    This dataset updates: Never
    Blog post about this prediction can be found here: http://bit.ly/2fWF2jq The predicted priority index of Typhoon Haima is produced by a machine learning algorithm that was trained on four past typhoons: Haiyan, Melor, Hagupit and Rammasun. It uses base line data for the whole country, combined with impact data of windspeeds and rains, and trained on counts by the Philippine government on people affected and houses damaged. First run The Priority Index is a 1-5 classification that can be used to identify the worst hit areas: those that need to be visited for further assessments or support first. Second run The model now predicts two things: a weighted index between partially damaged and completely damaged, where partially damaged is counted as 25% of the completely damaged. This has proven to give he highest accuracy. the precentage of total damage (damaged houses versus all houses) The absolute number of houses damaged / people affected is insufficiently validated at the moment, and should just be used for further trainng and ground-truthing. Data sources: Administrative boundaries (P_Codes) - Philippines Government; Published by GADM and UN OCHA (HDX) Census 2015 (population) - Philippine Statistics Authority; received from UN OCHA (HDX) Avg. wind speed (km/h) - University College London Typhoon path - University College London Houses damaged - NDRRMC Rainfall - GPM Poverty - Pantawid pamilyang pilipino program (aggregated) For the second run of the algorithm we also included: Roof and wall materials New geographical features The result of different models can be found in the file 'Typhoon Haima - performance of different models - second run.csv' A note on how to interpret this. date running date alg_date same alg_model name of the algorithm used alg_predict_on name of the learning variable alg_use_log i s the learning variable transformed in log code_version version of the learn.py code All the columns with feat_ indicates the importance of that feature, if not present that feature was not used. learn_matrix name of the learning matrix with the 5 typhoons run_name unique run name (pickle files and csv files have this name for this model) typhoon_to_predict name of a new typhoon to predict val_accuracy accuracy based on 10 categories of damage 0% 10% 20% … val_perc_down perc of underpredicted categories val_perc_up perc of overpredicted categories Val_best_score best r2 score Val_stdev_best_score error on best score based on the CV Val_score_test r2 score on the test set (this should be around +- 5% of the previus number to not overfit Val_mean_error_num_houses average error on the number of houses val_median_error_num_houses median val_std_error_num_houses std deviation of the errors (lower is better) Algorithm developed by 510.global the data innovation initiative of the Netherlands Red Cross.
  • 500+ Downloads
    Updated 25 November 2016 | Dataset date: November 22, 2016-November 22, 2016
    This dataset updates: Never
    Estimated population movement in Haiti as of 22 November 2016. The estimated distribution of people for whom their home Commune/Section Communale in the pre-hurricane period was in either Grande Anse, Sud or Nippes départment, and as of 22 November had moved to another Commune/Section Communale. Estimates are based on movements of de-identified Digicel SIM cards which made or received at least one call pre-hurricane and in the week up to 22 November 2016. The SIM card movements are combined with available population data derived from estimates for the year 2015.
  • 400+ Downloads
    Updated 21 November 2016 | Dataset date: November 08, 2016-November 08, 2016
    This dataset updates: Never
    Estimated population movement in Haiti as of 8 November 2016. The estimated distribution of people for whom their home Commune/Section Communale in the pre-hurricane period was in either Grande Anse, Sud or Nippes départment, and as of 8 November had moved to another Commune/Section Communale. Estimates are based on movements of de-identified Digicel SIM cards which made or received at least one call pre-hurricane and in the week up to 8 November 2016. The SIM card movements are combined with available population data derived from estimates for the year 2015.
  • 1400+ Downloads
    Updated 15 November 2016 | Dataset date: October 24, 2016-October 24, 2016
    This dataset updates: Never
    NDRRMC report of October 28th 2016
  • 900+ Downloads
    Updated 4 November 2016 | Dataset date: October 24, 2016-October 24, 2016
    This dataset updates: Never
    Summary of estimated population movement in Haiti as of 24 October 2016. The estimated distribution of people for whom their home Section Communale in the pre-hurricane period was in either Grande Anse, Sud or Nippes départment, and as of 24 October had moved to another Section Communale. Estimates are based on movements of de-identified Digicel SIM cards which made or received at least one call pre-hurricane and on 24 October 2016. The SIM card movements are combined with available population data derived from estimates for the year 2015.
  • 400+ Downloads
    Updated 3 November 2016 | Dataset date: October 20, 2016-October 20, 2016
    This dataset updates: Never
    This data is created using GPM data. It contains the accumulated precipitation for the periods October 11-18th, October 18-20th and the total over both periods.
  • 2000+ Downloads
    Updated 2 November 2016 | Dataset date: October 25, 2016-October 25, 2016
    This dataset updates: Every week
    This data is about damaged houses in the Philippines after Typhoon Haima (Lawin)
  • Updated 31 October 2016 | Dataset date: October 27, 2016-October 27, 2016
    This dataset updates: Never
    This map illustrates potentially damaged satellite-detected areas and related damage density in the southwestern part of Haiti in Sud Department, Haiti. The UNITAR-UNOSAT analysis used WorldView-2 satellite imagery acquired on the 9 and 17 October 2016. The UNITAR-UNOSAT analysis, identified 3,678 potentially damaged structures within this map extent. Additionally, 87 road obstacles and 131 population gathering sites within this area were identified in the analysis. Please note that some areas were cloud covered and could not be analysed. This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to UNITAR-UNOSAT.
  • 10+ Downloads
    Updated 31 October 2016 | Dataset date: October 27, 2016-October 27, 2016
    This dataset updates: Never
    This map illustrates satellite-detected potentially damaged buildings in Tiburon town, Sud department, Haiti. The UNITAR-UNOSAT analysis used a WorldView-2 satellite image acquired on the 17 July 2016 as a pre-image and a Worldview- 2 satellite image acquired on the 17 October 2016 as a post-image. The UNITAR-UNOSAT analysis identified 648 potentially damaged structures within the map extent of which 611 were identified inside the town of Tiburon. The depiction and use of town boundaries acquired from Wikimapia, are not warranted to be error- free nor do they imply official endorsement or acceptance by the United Nations. This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to UNITAR - UNOSAT.
  • 10+ Downloads
    Updated 31 October 2016 | Dataset date: October 20, 2016-October 20, 2016
    This dataset updates: Never
    This report describes preliminary building damage analysis carried out by UNITAR-UNOSAT covering Area 1 (Jeremie and Roseaux Commune) and Area 2 (Abricot, Dame-Marie, Anse d’Hainaults and Les Irois Communes) for a total area of approximately 650 Km2. Building damage analysis, including a rapid assessment of transportation network conditions and locations of spontaneous people gathering sites, was conducted by comparing the post-disaster satellite images (Pleiades acquired on 7/10/2016 for AOI1 and 12/10/2016 for AOI2) with available pre-disaster images (WorldView-1 on 01/05/2015, 09/05/2015 and 08/12/2014; Worldview-2 on 17 July 2016).
  • 10+ Downloads
    Updated 31 October 2016 | Dataset date: October 21, 2016-October 21, 2016
    This dataset updates: Never
    This report describes preliminary building damage analysis carried out by UNITAR-UNOSAT covering the towns and its surrounding areas over Baracoa, Maisi, Imias, and Cajobabo. Building damage analysis was conducted with the post-disaster satellite images (Pleiades acquired on 7/10/2016, 10/10/2016, and 11/10/2016).
  • Updated 31 October 2016 | Dataset date: October 20, 2016-October 20, 2016
    This dataset updates: Never
    This map illustrates potential satellite-detected damaged structures in Baracoa town, in Guantanamo Province, Cuba. The UNITAR-UNOSAT analysis used a Pleiades satellite image, acquired on the 11 October 2016, and identified 3,147 damaged structures inside the town of Baracoa.This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to UNITAR-UNOSAT.
  • 20+ Downloads
    Updated 31 October 2016 | Dataset date: October 18, 2016-October 18, 2016
    This dataset updates: Never
    This map illustrates potentially damaged satellite-detected areas, related damage density and the identified road obstacles in the southwestern part of Haiti in Grande Anse and Sud departments, Haiti. The UNITAR-UNOSAT analysis used a Pleiades satellite image acquired on the 12 October 2016. The UNITAR-UNOSAT analysis combined with Copernicus analysis, identified 9,173 potentially damaged structures within this map extent. Additionally, 123 road obstacles and 255 population gathering sites within this area were identified in the analysis. Please note that some areas were cloud covered and could not be analysed. This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to UNITAR-UNOSAT.
  • Updated 31 October 2016 | Dataset date: October 18, 2016-October 18, 2016
    This dataset updates: Never
    This map illustrates potential satellite-detected structures in Cajobabo town and the surrounding area in Guantanamo Province, Cuba. The UNITAR-UNOSAT analysis used a Pleiades satellite image, acquired on the 7 of October, 2016. The analysis identified 450 damaged structures within the map extent of which 211 were identified inside the town of Cajobabo. This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to UNITAR - UNOSAT.
  • 20+ Downloads
    Updated 31 October 2016 | Dataset date: October 15, 2016-October 15, 2016
    This dataset updates: Never
    This report highlights the preliminary building damage analysis including a rapid assessment of transportation network conditions and locations of spontaneous people gathering sites covering Jérémie Commune and surrounding areas.The analysis was conducted by comparing the post-disaster satellite images (Pleiades acquired on 7/10/2016) with available pre-disaster images (WorldView-1 on 01/05/2015 and 08/12/2014).
  • Updated 31 October 2016 | Dataset date: October 14, 2016-October 14, 2016
    This dataset updates: Never
    This map illustrates satellite-detected areas potentially damaged and related density and the identified road obstacles in Jérémie city and surrounding areas in Grande Anse department, Haiti. The UNITAR-UNOSAT analysis used a Pleiades satellite image acquired on the 07 October 2016 and Worldview-1 image acquired on 01 May 2015 and 08 December 2014. The UNITAR-UNOSAT analysis in the outskirts of Jeremie combined with Copernicus analysis in Jeremie city, identified 13,013 potentially damaged structures within this map extent. The UNITAR-UNOSAT analysis identified 134 road obstacles and 800 population gathering sites within the map extent.This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to UNITAR-UNOSAT.
  • Updated 31 October 2016 | Dataset date: October 13, 2016-October 13, 2016
    This dataset updates: Never
    This map illustrates potential satellite-detected damaged structures in Maisi town and the surrounding area in Guantanamo Province, Cuba. The UNITAR-UNOSAT analysis used a Pleiades satellite image, acquired on the 7 October 2016, and identified 417 damaged structures within the map extent; 230 were identified inside the town of Maisi .This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to UNITAR-UNOSAT.
  • Updated 31 October 2016 | Dataset date: October 12, 2016-October 12, 2016
    This dataset updates: Never
    This map illustrates potential satellite-detected damaged structures in Imias and the surrounding area in Guantanamo Province, Cuba. The UNITAR-UNOSAT analysis used a Pleiades satellite image, acquired on the 10 October 2016, and identified 552 damaged structures within the map extent and 603 damaged structures in the covered area. This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to UNITAR-UNOSAT.
  • 500+ Downloads
    Updated 31 October 2016 | Dataset date: October 25, 2016-October 25, 2016
    This dataset updates: Never
    Summary of Digicel Haiti mobile network functionality as of 25th October 2016. Proportion of Digicel Haiti network (radio cells) with normal function per commune (admin level 2) as of 25th October 2016. French version provided.
  • 1000+ Downloads
    Updated 20 October 2016 | Dataset date: October 20, 2016-October 20, 2016
    This dataset updates: Never
    Data provided by Mark Saunders at University College London
  • 300+ Downloads
    Updated 18 October 2016 | Dataset date: October 18, 2016-October 18, 2016
    This dataset updates: Never
    Partial Assessment and 4W matrix ,Hurricane Matthew. as of 18.10.2016