2.1 GLOBE Observer Mosquito Habitat Mapper App Description

The Global Learning and Observations to Benefit the Environment (GLOBE) Program is an international science and education program engaging students, teachers, community members and scientists in Earth system monitoring and research (www.globe.gov). Since its inception 25 years ago, GLOBE has supported authentic inquiry-based science in K-12 classrooms in 124 countries though a student environmental monitoring program that applies scientist-developed research protocols and data reporting procedures. In 2016, The GLOBE Program launched the GLOBE Observer (GO) mobile application (app), expanding its mission to include the participation of citizen scientists at large, both broadening the reach of NASA science and increasing the spatial and temporal density of environmental observations reported to the database (Amos & Andersen, 2019).

Citizen scientists using the GO app provide in-situ, ground reference data that can be used by scientists to analyze and interpret digital data derived by sensors on remote platforms. Terrestrial, hydrologic and atmospheric observations are collected and reported using four-in app protocols: Clouds, Land Cover, Tree Height, and Mosquito Habitat Mapper. In the field, coincident observations using two or more protocols are encouraged because this provides citizen scientists with an opportunity to observe and document firsthand some of the complex Earth system interactions that are responsible for the changes we see taking place on our planet.

Internet connectivity is not required to make and record observations using the app: Data are stored in the mobile device until sufficient bandwidth is available for uploading data to the GLOBE database. GLOBE data are publicly available and accessible in a variety of ways. Data can be downloaded as KML or CSV files using the GLOBE Visualization Tool (GLOBE Vis), the Advanced Data Analysis Tool (ADAT) or through an Application Programming Interface (API) from the GLOBE website (https://www.globe.gov/globe-data). Descriptions of GLOBE data visualization and access procedures, metadata descriptions, and quality assurance practices are detailed in Amos and Andersen (2019) and GLOBE (2019).

The GLOBE Mosquito Habitat Mapper tool was released in May 2017 to provide in-situ ground based observations to support scientific research using environmental data obtained from satellites. Scientists who develop risk models for mosquito borne diseases frequently employ remotely sensed data, but the mosquito vectors themselves obviously can't be seen from remote sensing platforms in space. Mosquitoes are arthropods whose immature forms are aquatic, and they respond sensitively to changes in environmental conditions such as precipitation, temperature and humidity. Along with weather and climate, land cover data are frequently analyzed by scientists who conduct mosquito vector borne disease research (see systematic review by Sallam et al., 2017). There is a pressing need for fine-grained in-situ data describing relationship of mosquito larval habitats to the land cover classes identified in satellite products (Lorenz et al., 2020) and thus in-situ observations of mosquito larval habitats are needed, especially in regions where variable conditions are pronounced or where change is occurring rapidly. Citizen scientists, working in conjunction with local public health and mosquito control authorities, are situated to fill this data gap. To support community-level engagement, the GO app enables groups to establish a team, a feature that enables data from a local mosquito surveillance group, campaign or event to be readily associated together and analyzed.

As a built-in incentive for citizen scientists who collect data needed by the scientific community, their actions performed as part of the GLOBE Mosquito Habitat Mapper protocol simultaneously reduce the risk of mosquito vector borne disease in and around their homes and neighborhoods. By reporting and mitigating mosquito larval habitats, citizen scientists eliminate thousands of immature mosquitoes that otherwise would become biting adults and potential vectors of human disease.

By design, GLOBE Mosquito Habitat Mapper citizen scientists can participate in data collection at both casual and dedicated levels of engagement. To accommodate a citizen scientist's interest, time constraints, access to simple equipment, and/or ability, the app is designed to allow the volunteer to terminate their observation at each of the 4 steps in the app. This opens participation to individuals in a broad range of ages (13+) and skill levels, from novice to expert amateur to professional scientist. The steps in the GLOBE Mosquito Habitat Mapper protocol are described below.

Step 1: Mosquito larval habitat documentation. The first data collection step involves identifying and photo documenting potential mosquito larval habitats on the landscape. Users are prompted to look for standing water sources-natural or artificial-where an adult mosquito might lay her eggs. For example, users might find discarded items such as cans, bottles or old tires, or purposeful water storage containers, such as cisterns or rain barrels that serve as oviposition sites. Natural standing water sources include ponds, puddles, marshes and estuaries. GLOBE Observers classify the mosquito larval habitat they observe into one of 32 categories (including “other”) by making selections from images in a graphic interface. Next, they are instructed to document their observation with up to nine photos of the mosquito habitat and surrounding area. The app prompts them to report whether mosquito larvae are visible in the standing water source. A series of yes/no questions follow where users report whether they observe larvae, pupae, adults or eggs at their larval habitat: Reference images are provided for reference by novice users. Once the larval site has been documented, citizen scientists can either terminate observations or continue to the next step.

Step 2: Determine if immature mosquitoes are present in the habitat. Using a macropipette or mosquito dipper, citizen scientists are asked to take a sample of the water and enter separate counts for larvae and pupae observed. If no immature specimens are detected, users are instructed to enter “zero” as the larvae and/or pupae counts. If larvae are observed, users are prompted to photograph representative specimens, using the zoom feature of the camera in their mobile device. Interested citizen scientists are encouraged to obtain an inexpensive (less than $10 US) 60x macrolens that can be clipped onto their mobile camera lens to capture images of sufficient resolution to identify to genera. To support expert identification, voucher photos of larvae are stored at their original image resolution in the GLOBE database. Users are then given the option to terminate observations at this step or continue to specimen identification.

Step 3: Taxon identification. Users are prompted to capture a full body photo of their specimen, as well as a close-up photo of the diagnostic features located on the terminal anal segments used for discriminating between taxa. Once a photo is taken, close-up larva photo is displayed at the top of each step in the key for convenient comparison with diagnostic images.

Citizen scientists employ an in-app pictorial dichotomous taxonomic key to determine whether their larva specimen belongs to one of three medically important genera found worldwide: Aedes, Culex or Anopheles (WHO, 2020). The app's information system collects user data at each step in the key, so that if users end their identification task before identifying the specimen to genus, the last step attempted is recorded. This feature provides a record of the precision of citizen scientist identifications (Lukyanenko et al., 2019). At several points in the identification process, citizen scientists have the option of selecting “I'm not sure,” a feature of the information system that reduces false reports and enhances data quality (Torre et al., 2019). If citizen scientists would like to identify different genera than the three targeted ones, they can use a local mosquito identification key and enter the genus or species into the app where they write comments.

Step 4: Source reduction. Regardless of the number of steps previously completed, all users end their observation by indicating if they were able to mitigate the mosquito larval habitat site by removing, dumping out, or covering a water container. This action, known within the mosquito control community as source reduction, is a is an important behavioral practice which has a measurable impact on the transmission of dengue, malaria and other diseases (Forsyth et al., 2020; Gu et al., 2006).

Once the observation is completed, the user is prompted to make a coincident observation using the Land Cover tool, also found on the GO app. Then the user is prompted to either upload their data to the GLOBE database, or alternatively, store the data in the device until an internet connection is available. Each citizen scientist is able to access a collection of their own data in app. Personally reported Land Cover observations can also be reviewed in. app via a map interface.

In the following section, we describe the data obtained using the GLOBE Mosquito Habitat Mapper for the three years since its release on the GO platform, from May 29, 2017 through May 28, 2020.

After describing the data as a whole, we compare three world regions and present a limited case study of data reported by citizen scientists from Senegal.