Data Descriptor | Open

RiceAtlas, a spatial database of global rice calendars and production

  • Scientific Data 4, Article number: 170074 (2017)
  • doi:10.1038/sdata.2017.74
  • Download Citation
Published online:


Knowing where, when, and how much rice is planted and harvested is crucial information for understanding the effects of policy, trade, and global and technological change on food security. We developed RiceAtlas, a spatial database on the seasonal distribution of the world’s rice production. It consists of data on rice planting and harvesting dates by growing season and estimates of monthly production for all rice-producing countries. Sources used for planting and harvesting dates include global and regional databases, national publications, online reports, and expert knowledge. Monthly production data were estimated based on annual or seasonal production statistics, and planting and harvesting dates. RiceAtlas has 2,725 spatial units. Compared with available global crop calendars, RiceAtlas is nearly ten times more spatially detailed and has nearly seven times more spatial units, with at least two seasons of calendar data, making RiceAtlas the most comprehensive and detailed spatial database on rice calendar and production.

Design Type(s)
  • data integration objective
  • database creation objective
  • observation design
Measurement Type(s)
  • agricultural calendar
  • agricultural production
Technology Type(s)
  • digital curation
Factor Type(s)
  • geographic location
Sample Characteristic(s)
  • Afghanistan
  • Algeria
  • Angola
  • Australia
  • Azerbaijan
  • Bangladesh
  • Belize
  • Benin
  • Bhutan
  • Bolivia
  • Brazil
  • Brunei Darussalam
  • Bulgaria
  • Burkina Faso
  • Burundi
  • Cambodia
  • Cameroon
  • Central African Republic
  • Chad
  • Chile
  • China
  • Colombia
  • Comoros
  • Costa Rica
  • Cote d'Ivoire
  • Cuba
  • Democratic Republic of the Congo
  • Dominican Republic
  • Ecuador
  • Egypt
  • El Salvador
  • Ethiopia
  • Fiji
  • France
  • Gabon
  • Gambia
  • Ghana
  • Greece
  • Guatemala
  • Guinea
  • Guinea-Bissau
  • Guyana
  • Guyane
  • Haiti
  • Honduras
  • Hungary
  • India
  • Indonesia
  • Iran
  • Iraq
  • Italy
  • Jamaica
  • Japan
  • Kazakhstan
  • Kenya
  • Kyrgyzstan
  • Laos
  • Lebanon
  • Liberia
  • Madagascar
  • Malawi
  • Malaysia
  • Mali
  • Mauritania
  • Mexico
  • Morocco
  • Mozambique
  • Myanmar
  • Nepal
  • Nicaragua
  • Niger
  • Nigeria
  • North Korea
  • Pakistan
  • Panama
  • Papua New Guinea
  • Paraguay
  • Peru
  • Philippines
  • Portugal
  • Republic of Congo
  • Republic of South Africa
  • Reunion Island
  • Romania
  • Russia
  • Rwanda
  • Saint Lucia
  • Saudi Arabia
  • Senegal
  • Sierra Leone
  • Solomon Islands
  • Somalia
  • South Korea
  • Spain
  • Sri Lanka
  • Sudan
  • Suriname
  • Swaziland
  • Taiwan Province
  • Tajikistan
  • Tanzania
  • Thailand
  • The Philippines
  • Timor-Leste
  • Togo
  • Trinidad and Tobago
  • Turkey
  • Turkmenistan
  • Uganda
  • Ukraine
  • United States of America
  • Uruguay
  • Uzbekistan
  • Venezuela
  • Viet Nam
  • Zambia
  • Zimbabwe
  • rice field

Background & Summary

Rice is the world’s most important food crop. It is harvested from over 163 million ha in more than 100 countries ( It is grown in diverse cropping systems and environments—from single crop systems in temperate and tropical regions in both rainfed and irrigated conditions, to intensive monoculture in irrigated areas in the tropics where rice is grown two or three times per year.

Although information on the distribution of global rice production by region and country can be derived from readily available statistics (e.g.,,, information on its distribution within a year is often lacking. Rice area and production statistics are available only at the national level for some countries; if these are available at the subnational level, the statistics are often by year and not by season. Linking information on production and area to the crop calendar can help analyze spatio-temporal variation in rice production. This can contribute to an improved ability to answer questions about food security. For example, this information, together with data on climate shocks and rice stocks, can be used to better assess seasonal and geographic variation in rice supply to mitigate shortfalls in rice availability at certain times of the year. Furthermore, information on where and when rice is planted is needed to quantify the potential risk of abiotic and biotic stresses during the rice-growing seasons, and to model the effects of global climate change and technological change on rice yield and production. In summary, a globally complete and spatially explicit rice crop calendar linked to area and production estimates is a valuable global public good.

Several rice crop calendars exist (, 1,2,3). Some are limited to a few countries or have regional coverage, whereas others are global. Regional resources include crop calendars for Latin America and the Caribbean1 and for Africa (, which have information on the planting and harvesting periods of rice and other major crops by agro-ecological zone. The database on rice for Africa includes 26 countries, and that for Latin America includes 24. The calendar for Latin America is outdated, whereas that for Africa does not include many rice-producing countries. Global calendars have been developed recently2,3. The calendars by Sacks et al.2 lack detail at the subnational level, especially in developing countries. The MIRCA2000 (ref. 3) calendar is monthly, gridded, and available for irrigated and rainfed rice, but does not adequately cover rice areas that are cultivated more than once a year. In both global calendars, some areas with rice grown in two seasons have data for only one season. Also, they include only a maximum of two seasons—the main and second season—inadequately covering some of the world’s most important rice areas with three distinct cropping seasons, such as in Bangladesh, where aman rice (main rainy season) is harvested in November-December; boro (dry season) in April-May; and aus in July-August. Parts of Vietnam also have three cropping seasons—winter-spring, spring-summer, and summer-autumn—and this is also the case in parts of China (early and late seasons for double-cropped rice areas; and middle for single-cropped rice areas) and India.

Because of the need to develop a spatially explicit global database of rice calendars that includes detailed information on rice areas with more than one rice crop in a year, we compiled the most detailed available datasets of rice planting and harvesting dates by growing season in all rice-producing countries, and linked the database to subnational production data. ‘RiceAtlas’ provides a spatial and seasonal distribution of the world’s rice production. RiceAtlas contributes to the GEOGLAM (Group on Earth Observations Global Agricultural Monitoring)4 initiative and regional partnerships, such as the Asian Rice Crop Estimation and Monitoring initiative (Asia-RiCE), by providing information for agricultural monitoring requirements, satellite data acquisition plans, and global crop outlook.


Rice calendar

The rice calendar in RiceAtlas is based on various published sources such as global and regional datasets, international and national publications, online sources, and unpublished data sources such as expert knowledge. Collaborators from various countries contributed new datasets, which were used to revise or validate the initial database that was compiled from existing sources (Table 1, Supplementary Table 1).

Table 1: Main data sources for rice-growing regions*

We collected data on the start, peak, and end dates of sowing or transplanting, and the start, peak, and end dates of harvesting of rice for all seasons in all rice-growing countries. In cases where peak planting and harvesting dates were not available, we estimated those to be at the midpoint between the start and end dates. If only the peak planting dates were available, we assumed the start and end dates of planting to be 15 days before and 15 days after the peak date, respectively. The same procedure was used to estimate the start and end dates of harvesting if only peak harvesting dates were available. Planting in a region is not done on a single date but the length of the planting window varies between regions. In the absence of information, we set the planting window to 30 days. This can be revised when better information becomes available. Where available, additional data such as crop establishment method and seedling age for transplanted rice were recorded.

To describe RiceAtlas and compare its calendar with existing regional (; ref 1) and global (Sacks et al.2 and MIRCA20003) datasets, we used the following metrics:

  1. Coverage. This refers to the number of rice growing countries with data.

  2. Spatial detail. This refers to the number of spatial units for which data is available. To avoid double counting, we merged adjacent spatial units that have the same calendar. In the case of MIRCA2000, rice calendars are available for irrigated and rainfed rice. If both were available for one spatial unit, only the irrigated calendar was considered in the count of the spatial units.

  3. Seasonal detail. This is the number of spatial units with calendars for two or more seasons. This is a metric for the temporal completeness of the crop calendar.

  4. Resolution. We used measures similar to those used by Deichmann5:

  5. Overall spatial resolution =Landarea(1000km2)Numberofspatialunits

  6. Rice area resolution =Ricearea(1000ha)Numberofspatialunits

Rice production

We compiled rice production and area (henceforth referred to as production data or production statistics) from various sources such as national statistics agencies, agriculture departments, and the FAO (,; Table 1, Supplementary Table 1). Because available rice production data from different countries do not refer to the same years, we used the average of the last three years of available data, and adjusted these, such that the national production totals match the 2010–2012 average production from FAO ( A three-year average retains the recent production level and accounts for interannual fluctuations in production.

Spatial and seasonal analysis of rice production

Rice production data were linked with the crop calendar data through their locations using administrative boundaries in the GADM database of the Global Administrative Areas (version 2.8; There were two cases where the compiled data needed adjustment:

  1. Different levels of spatial resolution. For example, production statistics were available at one level (e.g., first level subdivisions) and the crop calendar was available at the a more detailed level (e.g., second level subdivisions) (35 countries);

  2. Mismatch in seasonal information, for example, the crop calendar reported double cropping, but the production data referred to only a single cropping season (33 countries).

In cases where the calendar data were spatially more detailed than the production data, we disaggregated the latter using expert knowledge where available, or by assuming equal production over the entire area or season. Conversely, if production data were more detailed than the calendar data, we disaggregated the latter and assumed the same calendar for all disaggregated spatial units. If the production statistics explicitly referred to only a single cropping season but the rice calendar had more than one season, rice production was attributed to the main season only. Because of the differences in the level of detail of available data for both rice calendar and production, the spatial detail of RiceAtlas varies across continents and countries (Figures 1 and 2, Table 2).

Figure 1: Spatial units covered in RiceAtlas and the number of rice-growing seasons.
Figure 1
Figure 2: Peak rice planting months by season.
Figure 2

Season 1 refers to the main rice-growing season, that is, the season with the highest rice production. Season 2 has the second highest rice production and Season 3 the least.

Table 2: Level of detail of RiceAtlas by geographic region.

The production data were distributed proportionally to the corresponding months based on sowing or transplanting, and harvesting dates (start, peak, and end) per season, with the peak planting or harvesting months having greater weight and progressively lesser weights for months away from the peak. For example, if the planting window was three months and the peak planting was the middle month, the first and last months were given equal weights of 0.25 each whereas the peak planting month was given 0.5. If production data for a region was available annually but there were two or more known crop seasons in a year in that area, production data were first disaggregated by season (equally, if seasonal data were not available) then distributed to months based on months of harvest. Area data, on the other hand, were distributed based on the months of both planting and harvesting to estimate the monthly area with a standing rice crop. Global monthly rice-growing areas, as well as regional and within-year distribution of rice production at harvest time, show the production peak and lean periods by region (Figures 3 and 4).

Figure 3: Monthly rice crop area in different geographic regions, 2010-2012.
Figure 3
Figure 4: Rice production by geographic region and month of harvest, 2010-2012.
Figure 4

Code availability

R code used to compute for the total planted area or production by month for each country is available from Data Citation 1: Harvard Dataverse

Data Records

RiceAtlas is a spatial database of rice calendars and production in 115 countries, with its attributes given in Table 3. RiceAtlas version 1.0 can be downloaded from the IRRI Dataverse Repository (Data Citation 1: Harvard Dataverse

Table 3: Attributes of RiceAtlas.

RiceAtlas will be improved by including more detailed data for selected countries when these become available and accessible. New versions will be periodically uploaded to the Repository.

Technical Validation

Expert knowledge was used to validate and correct the crop calendars. More than 50 persons contributed data and/or validated the crop calendars. Digital and analogue country-level maps and tabulations were provided to rice experts to verify the data for countries they were familiar with. Maps and tables were provided to members of the Temperate Rice Research Consortium during their review and planning meeting at IRRI headquarters on 8−9 November 2013. The data from Africa were reviewed by 27 rice scientists during the AfricaRice Science Week on 9−13 February 2015 in Cotonou, Benin. Their comments were addressed and their revisions were included in the current version. All authors validated the crop calendars for their respective regions of expertise.

RiceAtlas v1.0 has 2,725 spatial units in total. It has 2,209 unique contiguous spatial units for the rice calendar and is nearly 10 times more spatially detailed than previously published global rice calendars (Table 4). It has 904 spatial units with data for at least two rice growing seasons. This is almost seven times more than those of published global rice calendars and nearly 10 times more for Asia. RiceAtlas, therefore, greatly improves coverage of intensively cultivated rice areas in the region. In addition, based on comparison among rice calendars in terms of overall spatial resolution and rice area resolution, RiceAtlas is the most comprehensive and detailed global rice calendar currently available.

Table 4: Level of detail of RiceAtlas and selected global or regional rice calendars.

RiceAtlas will be updated as more detailed data become available and accessible. To date, the effort has concentrated on collecting and validating data for Asia and Africa. Arrangements have been made to update data for Latin America through the International Center for Tropical Agriculture (CIAT) and its network of partners in the region.

The spatial detail of RiceAtlas varies across continents and countries, and for some applications a higher spatial resolution could be desirable. To create a more homogenous and higher spatial resolution database, the available data could be used to build predictive models to downscale the rice calendar data to a higher spatial resolution (e.g., 1 km2 grid cells). Such models could use climate data and satellite images. For example, Moderate Resolution Imaging Spectroradiometer (MODIS) images have been used to detect key phenological stages of the rice crop including start of season6. The use of time series of satellite images can also allow for the detection of annual variation, changes in cropping intensity and shifts in planting dates.

Additional Information

How to cite this article: Laborte, A. G. et al. RiceAtlas, a spatial database of global rice calendars and production. Sci. Data 4:170074 doi: 10.1038/sdata.2017.74 (2017).

Publishers note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    FAO. Calendario de cultivos: America Latina y el Caribe (FAO, 2006).

  2. 2.

    , , & Crop planting dates: an analysis of global patterns. Global Ecol. Biogeogr. 19, 607–620 (2010).

  3. 3.

    , & MIRCA2000—global monthly irrigated and rainfed crop areas around the year 2000: a new high-resolution data set for agricultural and hydrological modelling. Global Biogeochem. Cy. 24, GB 1011 (2010).

  4. 4.

    GEO. Progress on GEOGLAM Implementation: First steps towards implementation 2013-2014 Phase I and II (2013).

  5. 5.

    . A Medium Resolution Population Database for Africa (1994).

  6. 6.

    , , & Multi-year monitoring of rice crop phenology through time series analysis of MODIS images. Int. J. Remote Sens. 30, 4643–4662 (2009).

  7. 7.

    , , & Mapping rice areas of South Asia using MODIS multitemporal data. J. Appl. Remote Sens 5, 053547 (2011).

  8. 8.

    , , , & Estimating crop yield potential at regional to national scales. Field Crop. Res. 143, 34–43 (2013).

  9. 9.

    , , & . Rice Almanac 3rd edn (International Rice Research Institute, 2002).

Download references

Data Citations

  1. 1.

    Laborte, A. G. Harvard Dataverse (2017)


This work was supported by the CGIAR Global Rice Science Partnership (GRiSP) and Policies, Institutions, and Markets (PIM), and the Global Futures Project.

We thank the following people who contributed datasets and/or checked and validated data:


Neemi Beser, Parvesh Kumar Chandna, Muralli Krishna Gumma, Nyo Me Htwe, David Johnson, Inez Slamet-Loedin, David Raitzer, Ben Samson, and Dule Zhao


Komlan Adigninou Ablede, Cyriaque Akakpo, Moundibaye Dastre Allarangaye, Essowedeou Sekou Ani, Idriss Baggie, Oladele Samuel Bakare, Raphael Kwame Bam, Ibrahim Bassoro, Adamou Bassou, Belay Abera Bayuh, Joseph Bigirimana, Madiama Cisse, Wilson Dogbe, Henri Gbakatchetche, Habibou Gueye, Famara Jaiteh, Geophrey Kajiru, Alain Kalisa, Nianankoro Kamissoko, Fanny Lunze, Illiassou Mossi Maïga, Buri Mohammed Moro, Yonnelle Moukoumbi, Rosemary Murori-Mutegi, David Nanfumba, Alexis Ndayiragije, Raymond Rabeson, Kéita Sékou, and Louis Yameogo

Latin America and the Caribbean:

Gonzalo Carracelas, Daniel Jimenez, Viviana Lorena Becerra Velasquez, and the Mexican Rice Council

Rest of the world:

Massimo Biloni, Russell Ford, and Kent MacKenzie

Author information


  1. Social Sciences Division, International Rice Research Institute (IRRI), Los Baños 4031, Laguna, Philippines

    • Alice G. Laborte
    • , Mary Anne Gutierrez
    • , Jane Girly Balanza
    • , M.V.R. Murty
    • , Lorena Villano
    • , Jorrel Khalil Aunario
    •  & Andrew Nelson
  2. Sustainable Productivity Enhancement Program, Africa Rice Center (AfricaRice), 01 BP 2031, Cotonou, Benin

    • Kazuki Saito
    •  & Sander J. Zwart
  3. Institute for Electromagnetic Sensing of the Environment, Italian National Research Council, Via Bassini 15, Milan 20133, Italy

    • Mirco Boschetti
  4. Plant Breeding Division, International Rice Research Institute (IRRI), Los Baños 4031, Laguna, Philippines

    • Russell Reinke
  5. Environment and Production Technology, International Food Policy Research Institute (IFPRI), Washington, District Of Columbia 20005, USA

    • Jawoo Koo
  6. Environmental Science and Policy, University of California, Davis, California 95616, USA

    • Robert J. Hijmans
  7. Department of Natural Resources, ITC - Faculty of Geo-Information Science and Earth Observation of the University of Twente, PO Box 217, 7500 AE Enschede, The Netherlands

    • Andrew Nelson


  1. Search for Alice G. Laborte in:

  2. Search for Mary Anne Gutierrez in:

  3. Search for Jane Girly Balanza in:

  4. Search for Kazuki Saito in:

  5. Search for Sander J. Zwart in:

  6. Search for Mirco Boschetti in:

  7. Search for M.V.R. Murty in:

  8. Search for Lorena Villano in:

  9. Search for Jorrel Khalil Aunario in:

  10. Search for Russell Reinke in:

  11. Search for Jawoo Koo in:

  12. Search for Robert J. Hijmans in:

  13. Search for Andrew Nelson in:


A.G.L. coordinated the data collection and verification, performed analyses, and drafted the Data Descriptor. M.A.G. compiled data from various sources, estimated monthly rice production, performed analyses, prepared the figures, coordinated the data collection and verification, and drafted a section of the Data Descriptor. J.G.B. compiled data from various sources, estimated monthly rice production, performed analyses, and drafted a section of the Data Descriptor. K.S., S.J.Z. provided and validated data for Africa, coordinated the data validation for Africa, and edited the Data Descriptor. M.B. provided and validated data for Europe and edited the Data Descriptor. M.V.R.M. validated the data for South and Southeast Asia. L.V. estimated monthly rice production and drafted a section of the Data Descriptor. J.K.A. developed routines to calculate monthly distribution of rice area and production. R.R. validated data and coordinated data validation for temperate rice-growing countries. J.K. provided data for Latin America and the Caribbean. R.J.H. provided data for various countries and edited the Data Descriptor. A.N. conceived the project and drafted sections in and edited the Data Descriptor. All authors have read and approved the final version of the Data Descriptor.

Competing interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to Alice G. Laborte.

Supplementary information

Word documents

  1. 1.

    Supplementary Table 1

Creative Commons BYThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit The Creative Commons Public Domain Dedication waiver applies to the metadata files made available in this article.