================ CEMADEN datasets ================ Updated: 2025-02-15 Introduction: What is CEMADEN? ------------------------------ .. _cemaden-observational-data: http://www2.cemaden.gov.br/ .. _cemaden-ped-platform: https://ped.cemaden.gov.br/ CEMADEN (National Center for Monitoring and Alerts of Natural Disasters) is an institution linked to the Ministry of Science, Technology and Innovation (MCTI) that works in the monitoring of risks and the issuance of alerts of natural disasters in Brazil. Its main mission is to preserve lives and reduce socioeconomic damage by anticipating extreme events, such as landslides, floods, flash floods, and droughts. CEMADEN collects environmental data from various sources and sensors spread throughout the national territory, through the Observational Network, which comprises meteorological, hydrological, rainfall, geotechnical stations, among others. How to access the data: PED Platform ------------------------------------ The observational data collected by CEMADEN are available to the public through the `Data Delivery Platform (PED) `_ [2]_. In it, anyone can make data requests, respecting some rules: * Limit of requests: * External users: up to 12 requests per minute. * Institutional partners: up to 180 requests per minute. * Scheduled requests (histories) are available for download for 30 days . * Accounts that have been inactive for more than one year are automatically removed. .. note:: It is mandatory to cite the source of the data when using them: DATA FROM THE CEMADEN/MCTIC OBSERVATIONAL NETWORK. Data structure -------------- The data made available in the `PED `_ [2]_ follow a clear organization, both in format and classification: * Each station file can contain data in two categories: * **Daily** (e.g., accumulated rainfall per day); * **Non-daily** (e.g. hourly, real-time data). * The stations also have different **types**: * "C" → Acqua * "U" → Agrometeorological * "G" → Geotechnical * "H" → Hydrological * "A" → Rainfall * "B" → Rainfall B * "T" → Test Types of sensors ^^^^^^^^^^^^^^^^ Each type of station monitored by CEMADEN presents data according to some sensors. These sensors monitor different environmental variables. See the :ref:`Sensor types table ` for the codes associated with the environmental variables monitored by each station type. .. _tabela-sensores: .. table:: Sensor types for CEMADEN stations :widths: auto :align: center ======== =============================================== Code Description ======== =============================================== 10 Rain 20 Level 60 Air temperature 90 Relative humidity 180 Wind speed 190 Wind direction 210 Solar radiation 240 Precipitation Intensity 260 Minimum Level 270 Maximum Level 280 Balance Radiation 290 Soil Temperature Level 1 300 Soil Temperature Level 2 310 Soil Temperature Level 3 320 Soil Temperature Level 4 330 Soil Moisture Level 1 340 Soil Moisture Level 2 350 Soil Moisture Level 3 360 Soil Moisture Level 4 370 Maximum daily air temperature 390 Maximum daily relative humidity 400 Minimum daily relative humidity 410 Maximum daily wind speed 420 Wind direction at maximum daily speed 430 Daily prevailing wind direction 440 Soil Temperature Level 1 Maximum Daily 450 Soil Temperature Level 1 Daily Minimum 460 Soil Temperature Level 2 Maximum Daily 470 Soil Temperature Level 2 Daily Minimum 480 Soil Temperature Level 3 Maximum Daily 490 Soil Temperature Level 3 Daily Minimum 500 Soil Temperature Level 4 Maximum Daily 510 Soil Temperature Level 4 Daily Minimum 520 Soil Moisture Level 1 Maximum Daily 530 Soil Moisture Level 1 Daily Minimum 540 Soil Moisture Level 2 Maximum Daily 550 Soil Moisture Level 2 Minimum Daily 560 Soil Moisture Level 3 Maximum Daily 570 Soil Moisture Level 3 Minimum Daily 580 Soil Moisture Level 4 Maximum Daily 590 Soil Moisture Level 4 Minimum Daily 600 Daily Accumulated Precipitation 610 Soil Moisture Level 5 620 Soil Moisture Level 6 630 Soil Moisture Level 5 Maximum Daily 640 Soil Moisture Level 5 Minimum Daily 650 Soil Moisture Level 6 Maximum Daily 660 Soil Moisture Level 6 Minimum Daily ======== =============================================== .. note:: Important: Not all station types have data for the Daily or Non-Daily categories and neither do the same sensors. Each type can provide different data, depending on its specific sensors. Processing Script ----------------- The data made available via PED are later processed into a structure that combines information from several stations, however, maintaining the separation by Type of station, Category (Daily / Non-daily), Type of sensor and Year, covering the period from 2013 to 2024. The period covered includes information from 2013 onwards, as this is the year of registration of the first station registered in the CEMADEN system. After processing, the data files are segmented according to the structure defined above, being organized into directories according to the subdivisions by type, category, sensor and year. .. image:: _static/images/cemaden_folders_1.png :width: 700 :alt: Structure of CEMADEN data processed For rainfall stations, hourly data are accumulated to form daily series, which allows the analysis of the volume of accumulated daily rainfall, calculation of indicators such as SPI and SPEI and comparison with other precipitation data, such as those made available by the National Water Agency (ANA). List of stations ---------------- CEMADEN provides the list of stations registered via PED through a Web request service, returning data that can be structured in table format. These data correspond to the records of stations registered in the CEMADEN system, and which have descriptive and geolocation information. When carrying out the analysis, all stations registered until 12/31/2023 were considered, so that, in 2024, there may be stations with at least 1 year of registered data. Among the 3873 stations registered by the deadline. The list of stations made available by CEMADEN has a structure similar to the one shown below: +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | | cidade | codestacao | codibge | data_instalacao | id_estacao | latitude | longitude | nome | rede_sigla | tipoestacao_descricao | uf | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 0 | BRASÍLIA | 530010805A | 5300108 | 2015-02-09 19:09:43.717 | 7878 | -15.876800 | -47.962700 | Nucleo Bandeirante | CEMADEN | Pluviométrica | DF | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 1 | BRASÍLIA | 530010804A | 5300108 | 2015-02-09 17:23:13.357 | 7877 | -15.623700 | -47.846900 | Sobradinho | CEMADEN | Pluviométrica | DF | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 2 | BRASÍLIA | 530010803A | 5300108 | 2015-02-09 17:15:08.558 | 7876 | -15.781966 | -47.998073 | Cidade Estrutural | CEMADEN | Pluviométrica | DF | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 3 | BRASÍLIA | 530010802A | 5300108 | 2015-02-09 17:05:58.246 | 7875 | -15.818400 | -48.153300 | Ceilandia | CEMADEN | Pluviométrica | DF | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 4 | BRASÍLIA | 530010801A | 5300108 | 2014-12-29 17:39:06.077 | 7635 | -15.827000 | -48.023000 | Águas Claras | CEMADEN | Pluviométrica | DF | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 3868 | JORDÃO | 120032801A | 1200328 | 2018-08-14 21:36:09.819 | 9674 | -9.190133 | -71.950808 | Centro | CEMADEN | Pluviométrica | AC | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 3869 | BRASILÉIA | 120010401A | 1200104 | 2018-08-14 20:39:49.214 | 9673 | -11.012469 | -68.740939 | Centro | CEMADEN | Pluviométrica | AC | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 3870 | PORTO VELHO | 110020502A | 1100205 | 2013-12-13 14:23:10.344 | 3014 | -8.762160 | -63.907421 | Porto Velho | CEMADEN | Pluviométrica | RO | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 3871 | PORTO VELHO | 110020501A | 1100205 | 2013-12-12 18:04:09.767 | 3151 | -8.742165 | -63.904242 | AC São Sebastião | CEMADEN | Pluviométrica | RO | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ | 3872 | PIMENTA BUENO | 110018901A | 1100189 | 2018-08-18 15:58:30.208 | 9740 | -11.683234 | -61.182871 | Apidia | CEMADEN | Pluviométrica | RO | +------+---------------+------------+---------+-------------------------+------------+------------+------------+--------------------+------------+-----------------------+-----+ The dictionary with all the variables available on CEMADEN's station registration form is available in :ref:`Station registration variables ` below: .. _station-variable-table: .. table:: Variables available in CEMADEN station records :widths: auto :align: center ======================== =============================================================== ========== Variable Description Type ======================== =============================================================== ========== altitude Station altitude in meters float64 cidade Name of the city where the station is located object codestacao Station code according to CEMADEN object codibge IBGE code of the municipality (7 digits) int64 cota_alerta River level alert quota (confirm) float64 cota_atencao River level attention quota (confirm) float64 cota_transbordamento River overflow quota (confirm) float64 data_instalacao Station installation date and time object dh_cadastro Date and time of registration of the station object dh_inicio_inativo Date and time of the start of the inactive status object dh_ultima_remessa Date and time of last data shipment object id_estacao Unique station identifier int64 id_rede Observational network identifier int64 id_tipoestacao Station type identifier int64 latitude Latitude of the station in decimal degrees float64 longitude Longitude of the station in decimal degrees float64 nome Station name object offset Distance (in meters) between the river bottom and the sensor float64 rede_sigla Station network acronym (e.g. CEMADEN) object tipoestacao_descricao Description of the type of station (e.g. Rainfall) object uf Federative Unit (UF) of the station object ======================== =============================================================== ========== The data regarding the list of stations were obtained through the python code developed to make requests to the CEMADEN system via PED. However, to make these requests, it is necessary to register with the PED, with login and password, so that a temporary access token is generated, necessary to make data requests. .. note:: You can download the complete list of registered stations here: `CEMADEN registered stations <_static/lista_estacoes_cemaden.csv>`_ Geographical coverage by type of stations ----------------------------------------- The stations of the CEMADEN Observational Network are distributed in the Brazilian national territory, with their highest concentration near the coast (east). Each station is categorized according to different types. The :ref:`Table of station frequencies by type ` presents the frequencies and percentages of CEMADEN stations registered as of 12/31/2025, according to their respective types. .. _station-type-frequency: .. table:: Number and percentage of CEMADEN stations by type :widths: auto :align: center ======================= ========== ============ Station Type Frequency Percentage (%) ======================= ========== ============ Rainfall 3100 80.04 Acqua 447 11.54 Hydrologic 148 3.82 Geotechnical 96 2.48 Agrometeorological 82 2.12 ======================= ========== ============ The table with :ref:`station frequencies by region ` presents the frequencies and percentages of CEMADEN stations registered as of 12/31/2025, according to the Brazilian regions. .. _station-by-region: .. table:: Number and percentage of CEMADEN stations by Brazilian region :widths: auto :align: center ========== ========== ============= Region Frequency Percentage (%) ========== ========== ============= Southeast 1694 43.7 Northeast 1252 32.3 South 647 16.7 North 175 4.5 Midwest 105 2.7 ========== ========== ============= Among the Brazilian regions, the distribution of CEMADEN stations is as follows: Southeast Region with 1,694 stations (43.7%), Northeast Region with 1,252 (32.3%), South Region with 647 (16.7%), North Region with 175 (4.5%), and Midwest Region with 105 (2.7%). The table with :ref:`station frequencies by state ` presents the frequencies and percentages of CEMADEN stations registered as of 12/31/2025, according to the Brazilian states. .. _station-by-state: .. table:: Number and percentage of CEMADEN stations by Brazilian state :widths: auto :align: center ======================== ====================== =============== Federative Unit Number of Stations Percentage (%) ======================== ====================== =============== São Paulo 707 18.25 Minas Gerais 436 11.26 Rio de Janeiro 371 9.58 Pernambuco 371 9.58 Santa Catarina 364 9.40 Bahia 303 7.82 Ceará 188 4.85 Espírito Santo 180 4.65 Rio Grande do Sul 170 4.39 Paraná 113 2.92 Maranhão 99 2.56 Piauí 76 1.96 Amazon 72 1.86 Paraíba 72 1.86 Rio Grande do Norte 67 1.73 Pará 65 1.68 Alagoas 59 1.52 Goiás 52 1.34 Mato Grosso 25 0.65 Mato Grosso do Sul 23 0.59 Tocantins 17 0.44 Sergipe 17 0.44 Acre 6 0.15 Amapá 6 0.15 Roraima 6 0.15 Federal District 5 0.13 Rondônia 3 0.08 ======================== ====================== =============== The ten states with the highest number of registered stations are: São Paulo with 707 stations (18.25%), Minas Gerais with 436 (11.26%), Rio de Janeiro and Pernambuco both with 371 (9.58%), Santa Catarina with 364 (9.40%), Bahia with 303 (7.82%), Ceará with 188 (4.85%), Espírito Santo with 180 (4.65%), Rio Grande do Sul with 170 (4.39%), and Paraná with 113 (2.92%). The table with :ref:`station-type-by-region` presents the frequencies and percentages (per row) of CEMADEN station types distributed across the Brazilian regions. .. _station-type-by-region: .. list-table:: Frequencies and percentages of CEMADEN station types by Brazilian region :header-rows: 1 :widths: 22 12 20 14 14 18 :align: center * - Region - Acqua - Agrometeorological - Geotechnical - Hydrologic - Rainfall * - Southeast Region - 31 (1.83%) - 4 (0.24%) - 57 (3.36%) - 80 (4.72%) - 1522 (89.85%) * - Northeast Region - 416 (33.23%) - 78 (6.23%) - 28 (2.24%) - 29 (2.32%) - 701 (55.99%) * - South Region - 0 (0.0%) - 0 (0.0%) - 11 (1.7%) - 30 (4.64%) - 606 (93.66%) * - North Region - 0 (0.0%) - 0 (0.0%) - 0 (0.0%) - 5 (2.86%) - 170 (97.14%) * - Midwest Region - 0 (0.0%) - 0 (0.0%) - 0 (0.0%) - 4 (3.81%) - 101 (96.19%) The table shows the frequency and percentage (by row) of each CEMADEN station type across the five Brazilian regions. Figure :ref:`station-distribution-map` shows the spatial distribution of the 3.873 stations registered in CEMADEN as of 12/31/2023. .. _station-distribution-map: .. figure:: _static/images/station_distribution_cemaden.png :alt: Spatial distribution of CEMADEN stations :align: center :width: 700 Spatial distribution of the 3.873 CEMADEN stations as of 12/31/2023. The stations of the type 'Pluviometric B' are categorized as pluviometric in the cadastral list, while the 'Test' stations do not have a record in the registration form although they have data made available via PED. Data availability ----------------- The data were downloaded through a request registered with the PED. After evaluation and approval steps, a link is generated containing files in csv format, one per station. However, the number of files made available differs from the number of stations registered in the CEADEN system and therefore an evaluation was carried out seeking to understand how many registered stations have data made available via PED and how many registered stations have data available for download, considering the date of 03/25/2025, when the station data were downloaded after approval of the request. In all, of the 4027 stations with available data, 3867 have registration in the list of stations, while 160 stations have data, but are not included in the registration list. Conclusion ---------- CEMADEN is an essential pillar in the policy of prevention of natural disasters in Brazil. Through its robust data collection infrastructure and the PED platform, researchers, public managers, and society have access to valuable information to mitigate risks and plan response actions. .. rubric:: References .. [1] Centro Nacional de Monitoramento e Alertas de Desastres Naturais – CEMADEN. Observational Network Data of CEMADEN/MCTIC [Internet]. Brazil: CEMADEN; [cited 2025 May 19]. Available from: http://www2.cemaden.gov.br/ .. [2] Centro Nacional de Monitoramento e Alertas de Desastres Naturais – CEMADEN. Data Delivery Platform - PED [Internet]. Brazil: CEMADEN; [cited 2025 May 19]. Available from: https://ped.cemaden.gov.br/ **Contributors** +-------------------+----------------------------------------------------------------------+ | Marcos Eustorgio Filho | Center for Data and Knowledge Integration for Health (CIDACS), | | | Instituto Gonçalo Moniz, Fundação Oswaldo Cruz, Salvador, Brazil| +-------------------+----------------------------------------------------------------------+ | Danielson Neves | Center for Data and Knowledge Integration for Health (CIDACS), | | | Instituto Gonçalo Moniz, Fundação Oswaldo Cruz, Salvador, Brazil| +-------------------+----------------------------------------------------------------------+