Bioecology of fall armyworm Spodoptera frugiperda (J. E. Smith), its management and potential patterns of seasonal spread in Africa

GENERAL INFORMATION 1. Title of Dataset: Bioecology of fall armyworm Spodoptera frugiperda (J. E. Smith), its management and potential patterns of seasonal spread in Africa CBFAMFEW I 2. Author Information A. Principal Investigator Contact Information Name: Saliou Niassy Institution:ICIPE Address: DUDUVILLE KASARANI Email: [e-mail] B. Associate or Co-investigator Contact Information Name: Sevgan Subramanian Institution: ICIPE Address: Duduville, icipe Email: [e-mail] C. Alternate Contact Information Name: Institution: Address: Email: 3. Date of data collection (single date, range, approximate date): 4. Geographic location of data collection: 5. Information about funding sources that supported the collection of the data: We thank FAO, which made the data used in this paper freely available. The authors express their gratitude to farmers, extension agents and all technical staff members involved in the data collection in the different countries. USAID/OFDA contributed to the national capacity that enabled the data to be generated. SHARING/ACCESS INFORMATION 1. Licenses/restrictions placed on the data: No, data is open access 2. Links to publications that cite or use the data: None 3. Links to other publicly accessible locations of the data: None 4. Links/relationships to ancillary data sets: Partnership with FAO 5. Was data derived from another source? Yes. A. If yes, list source(s): FAO, FAMEWS 6. Recommended citation for this dataset: Community based fall armyworm monitoring, forecasting and early warning (CBFAMFEW) DATA & FILE OVERVIEW 1. File List: Niassy_et_al_FAMEWS_Plos.xlsx 2. Relationship between files, if important: All collected using FAMEWS App 3. Additional related data collected that was not included in the current data package: Rainfall data 4. Are there multiple versions of the dataset? no A. If yes, name of file(s) that was updated: i. Why was the file updated? ii. When was the file updated? METHODOLOGICAL INFORMATION 1. Description of methods used for collection/generation of data: 2.1. Survey sites The CBFAMFEW project was implemented in six eastern African countries (Burundi, Ethiopia, Kenya, Rwanda, Tanzania and Uganda) between August 2017 and August 2019. In each country, five maize-growing districts were selected, and in each district, 10 villages were selected for data collection, with two Community Focal Persons (CFPs) per village. The selected districts in each country for the study are presented in Figure 1, and the locations for larval scouting and trapping of FAW males in the six countries are presented in Figure 2. 2.2. Community-Based Fall Armyworm Monitoring, Forecasting, Early Warning and Management System (CBFAMFEW) The CBFAMFEW project established a network that allowed each selected village to have a trained CFP with a smartphone and a pheromone trap to collect FAW data. The CFPs were trained in the use of the FAO-developed FAMEWS mobile application (app) for data input on FAW incidence. CFPs were also trained on how to interpret data and provide timely advice to villagers and early warning of FAW attacks. Further training on regular field scouting and detecting the pest at various stages was also provided to the CFPs. To coordinate data collection and transmission, mobile phones were loaded with the FAW monitoring application app, FAMEWS. The FAMEWS app is available in Google Play in 13 languages, and was downloaded by farmers and extension workers. In total, 56 mobile phones were provided per country (1 mobile phone per village x 50 villages + 1 mobile per district officer x 5 officers + 1 mobile for the National Coordinator). Hence a total of 336 mobile phones were deployed in the six countries. Parallel to FAW population monitoring activities in East Africa, similar initiatives were conducted in Ghana, Liberia, Zambia, Mozambique and South Sudan using similar approach. Data collected from these countries using the FAMEWS app were also included with data obtained from the six eastern African countries to map the FAW density across a wider area of SSA. 2.3. Pheromone traps and field scouting For monitoring adults, two universal bucket traps (Unitrap) were installed in each village. Traps baited with FAW pheromone were deployed just after planting and monitoring started after the emergence of plant seedlings to detect the first arrival of moth pests. Traps and lure (lure blend: Z9-14Ac (81.7%); z11-16Ac (17.54%); Z7-12Ac (0.5%) and z9-12Ac (0.25%)) were supplied by Russell IPM Ltd, Unit 45, First Avenue, Deeside Industrial Park Deeside, Flintshire, CH5 2NU United Kingdom. The FAW monitoring kit also contained a toxicant which immobilized the moths once attracted to the device. One trap was placed inside the maize field and a second one outside the maize fields. The trap was hung from a suspended pole about 1.5 m above the ground, and one trap was used for every 0.5–2 ha. The pheromone lure was replaced every 3–6 weeks. The traps were checked and emptied weekly, and trapped moths were then sorted to identify FAW. Field scouting was conducted at least twice a week, from the seedling and early whorl stages of the maize crop. This was also the time that farmers and extension workers sampled for egg masses, larvae, damage symptoms and the presence of other pests such as the African Armyworm (AAW) Spodoptera exempta (Walker) and stemborers. The maize fields were scouted using a “W” pattern approach, which involved sampling 10 consecutive plants at five different spots along the “W” transect (20). FAW and non-target moth counts from pheromone traps, and field scouting data were recorded and entered into the FAMEWS app. The mean percentage of plants infested with FAW, AAW, and stemborers, was automatically tabulated. Data were also collected on the condition of traps and whether the lures (pheromones) had been replaced. Additional data collected with the FAMEWS app included dates, country, geolocation, crop information (variety, planting date, irrigated or rain-fed, fertilizers used, and crop growth stage), general health of the crop, the cropping system (e.g. mono/intercropping, seasonal, rotation and push-pull), pest management practice adopted (chemical pesticides or biopesticides), and rainfall. Push-pull encompasses intercropping maize with the legume Desmodium spp. (Desmodium intortum for climate-smart push-pull or Desmodium uncinatum for conventional push-pull) and a border row of Napier grass Pennisetum purpureum or Brachiaria brizantha cv Mulato II (for climate-smart push-pull) around the plot; both Desmodium sp. and Napier grass are perennial fodder plants [18]. Seasonal cropping is a farming practice in which the same crop is grown in the same area for a number of growing seasons. In the case of maize, it is mainly rain fed, and the land remain fallow between seasons. While crop rotation is the practice of growing a series of different types of crops in the same area across seasons. 2.4. FAW monitoring data validation and processing A national coordinator in each country was appointed to validate the data collected by the CFPs through FAMEWS, and who uploaded to the global platform maintained by FAO (http://www.fao.org/fall-armyworm/monitoring-tools/famews-global-platform/en/) through the global coordinator. The data presented in this study were downloaded from the FAO global platform, which has the entire database collected for the FAMEWS app between January 2018 to June 2019. The various data entries were officially requested from the FAO, cleaned and analyzed in Microsoft Excel and R software 3.6.1 [27]. 2.5. Weather data and mapping methodology Monthly rainfall data for 2018 and 2019 were sourced from Climate Hazards Group Infrared Precipitation with Station data (CHIRPS). CHIRPS incorporates satellite imagery with in-situ station data to create 1 km resolution gridded rainfall time series data (www.chc.ucsb.edu/data/chirps/). The geo-referenced points were used to extract monthly rainfall records (proxy data) in millimeters of the FAW infested areas in the six countries using the point sampling tool in QGIS software (http://qgis.osgeo.org). The FAW density was plotted against monthly rainfall data to establish whether there was a relationship between rainfall patterns and the abundance of FAW in the six countries. We interpreted effects of rainfall from a pest management perspective, whereby downpours, for instance could influence the pest flight and neonate movement and feeding on maize plants [28, 29]. Rainfall can both negatively and positively affect the general plant health, particularly in rain-fed cropping systems. Spatial distribution of FAW density (number of insects per 1 km2) in target countries, based on both trap and scouting data, were represented using heat maps. The FAW traps and scouting georeferenced data dating from January 2018 to June 2019 were organized on a quarterly basis based on the collection date. The points were clustered quarterly for the year 2018 to the 2nd quarter of 2019. The heat maps were developed using the kernel density tool in QGIS software (http://qgis.osgeo.org). Kernel density estimation is a powerful non-parametric technique for estimating probability density function of variables. The tool calculates the density of point features around each output raster cell. The density was calculated based on an accumulation of the number of FAW geo-referenced records in a sampled location, with a higher number of FAW records resulting in a higher value in the heat map. The heat maps were developed with the assumption that the sampling protocol for scouting and installation of traps was spatially unbiased. This analysis helps in identifying the hot-spot areas with high infestation rates. On the other hand, we utilized a maize cropping calendar which was freely available from FAO Global Information and Early Warning System (GIEWS: http://www.fao.org/giews/countrybrief/index.jsp) to explore the influence of maize sowing, growing, harvesting and fallow periods on FAW adults density during the main and second seasons. For a regional overview of FAW population dynamics, we also utilized FAW occurrence and density data from 11 African countries (six in East Africa and five in other regions) that are available in the FAO global platform. Further, data collected from pheromone traps (universal bucket traps) in the eastern and other African countries were quarterly mapped at the regional level. Specifically, we overlaid the quarter FAW density heat maps in 2018 and 2019 to show the change in FAW density over time in each location, and differences in FAW density between locations at a point in time. Further, we mapped the all sites in the region where FAW was consistently observed across all the quarters in the year (i.e. year-round distribution). 2. Methods for processing the data: 2.6 Statistical analysis Trap counts were modelled using the Negative Binomial Model (NBM) which accommodates overdispersion of integer counts data. The NBM was used to evaluate the effect of main crop types, rotation crop and cropping systems on FAW trap counts. Incident rate ratios (IRR) were estimated for the different levels of each factor relative to a chosen reference level of a factor in question. IRR is a relative measure of incidence rate, such that IRR = 1 means no effect of the exposure, IRR > 1 means positive effect of exposure and IRR < 1 means negative effect of exposure. Similary, the NBM was also used to study the effect of crop phenology on trap counts. For main crop factor, maize was used as a reference category while for cropping systems, we used seasonal cropping as a reference level. “Beans” was used a reference group for the crop rotation factor. The effect of maize crop stage on FAW trap counts and larval counts (as proportion of infested plants) was analysed for each country separately in view of high variability between countries. While the FAW trap counts were analysed using the NBM, proportion of larval infested plants were modeled using quasi-binomial model. All tests were performed at the 5% significance level. 3. Instrument- or software-specific information needed to interpret the data: Analyses were performed using R version 3.6.1 [27]. 4. Standards and calibration information, if appropriate: 5. Environmental/experimental conditions: N/A 6. Describe any quality-assurance procedures performed on the data: FAMEWS APP 7. People involved with sample collection, processing, analysis and/or submission: The CBFAMFEW project established a network that allowed each selected village to have a trained CFP with a smartphone and a pheromone trap to collect FAW data. The CFPs were trained in the use of the FAO-developed FAMEWS mobile application (app) for data input on FAW incidence. CFPs were also trained on how to interpret data and provide timely advice to villagers and early warning of FAW attacks. Further training on regular field scouting and detecting the pest at various stages was also provided to the CFPs. To coordinate data collection and transmission, mobile phones were loaded with the FAW monitoring application app, FAMEWS. The FAMEWS app is available in Google Play in 13 languages, and was downloaded by farmers and extension workers. DATA-SPECIFIC INFORMATION: 1. Number of variables: 48 2. Number of cases/rows: 109712 3. Variable List: ID date date by month cropFieldSizeUnit scouting traps locationName cropVariety cropPlantingDate rainLastDate country userCountry region latitude longitude cropMain cropFertilizer cropIrrigation cropStage cropHealth cropSystem cropFieldSize rainAmount trapsCount trapsConfirmedFAW trapsSuspectedFAW trapsOtherSpecies cropRotation scoutingPlantsChecked scoutingPlantsFAW scoutingPercentageFAW fawCropDamage fawCobDamage fawNaturalEnemies fawDeadLarvae fawControlUndertaken fawControlChemicalPesticideName fawControlChemicalPesticideLitres scoutingStageFAW fawCurrentDamage fawPreviousDamage fawControlLocalTypes fawControlBiopesticideName fawControlBiopesticideLitres fawLarvaeKilledByNaturalEnemies rotationIntercroppingCrop 4. Missing data codes: N/A 5. Specialized formats or other abbreviations used: Excel