Search

Citizen science, scientific work and data collection conducted by or with non-experts, is rapidly growing. Although the potential of citizen science activities to generate enormous amounts of data otherwise not feasible is widely recognized, the obtained data are often treated with caution and skepticism. Their quality and reliability is not fully trusted since they are obtained by non-experts using low-cost instruments or scientifically non-verified methods. In this study, we evaluate the performance of Parrot's Flower Power soil moisture sensor used within the European citizen science project, the GROW Observatory (GROW; https://growobservatory.org). The aim of GROW is to enable scientists to validate satellite-based soil moisture products at an unprecedented high spatial resolution through crowdsourced data. To this end, it has mobilized thousands of citizens across Europe in science and climate actions, including hundreds who have been empowered to monitor soil moisture and other environmental variables within twenty four high-density clusters around Europe covering different climate and soil conditions. Clearly, to serve as reference dataset, the quality of ground observations is crucial, especially if obtained from low-cost sensors. To investigate the accuracy of such measurements, the Flower Power sensors were evaluated in the lab and field. For the field trials, they were installed alongside professional soil moisture probes in the Hydrological Open Air Laboratory (HOAL) in Petzenkirchen, Austria. We assessed the skill of the low cost sensors against the professional probes using various methods. Apart from common statistical metrics like correlation, bias and root-mean-square difference, we investigated and compared the temporal stability, soil moisture memory, and the flagging statistics based on the International Soil Moisture Network (ISMN) quality indicators. We found a low inter-sensor variation in the lab and a high temporal agreement with the professional sensors in the field. The results of soil moisture memory and the ISMN quality flags analysis are in a comparable range for the low-cost and professional probes, only the temporal stability analysis shows a contrasting outcome. We demonstrate that low-cost sensors can be used to generate a dataset valuable for environmental monitoring and satellite validation and thus provide the basis for citizen-based soil moisture science. [ABSTRACT FROM AUTHOR]

The involvement of non-experts in scientific work and data collection, namely citizen science, is rapidly increasing. Although the potential of citizen science activities and projects to generate huge amounts of data otherwise not feasible is widely recognized, the obtained data and observations are often treated with caution and skepticism. Their quality and reliability is not fully trusted since they are obtained by non-experts using low-cost instruments and/or simplified methods. In this study, we evaluate the performance of the low-cost soil moisture sensor used within the European citizen science project GROW Observatory (GROW; https://growobservatory.org/). The aim of GROW is to empower hundreds of motivated citizens to monitor soil moisture and other environmental variables within nine high-density clusters around Europe covering different climate and soil conditions. The citizen's contribution is laying the foundation for scientists to validate satellite-based soil moisture products at an unprecedented high spatial resolution. Clearly, to serve as reference dataset, the quality of ground observations is crucial, especially if obtained from low-cost sensors. To investigate the accuracy of such measurements, the low-cost sensors were installed alongside professional soil moisture probes in the Hydrological Open Air Laboratory (HOAL) in Petzenkirchen, Austria, where data has already been collected for more than a year. We assess the skill of the low cost sensors against the professional probes by the means of various methods. Apart from common statistical metrics like correlation, bias and rmsd, we investigate and compare the temporal stability as well as soil moisture memory. We will demonstrate that low-cost sensors can be used to generate a dataset valuable for environmental monitoring and satellite validation and thus provide the basis for citizen-based soil moisture science.This study is funded by the GROW Observatory project of the European Union's Horizon 2020 research and innovation programme (https://growobservatory.org/) and the International Soil Moisture Network under IDEAS+ of the European Space Agency (https://ismn.geo.tuwien.ac.at/). [ABSTRACT FROM AUTHOR]

Monitoring soil moisture at both high spatial and temporal resolution is necessary for a number of applications, ranging from crop irrigation to landslide prediction. Microwave remote sensing has been successfully used to estimate soil moisture globally. Radiometers and scatterometers typically observe the same location with a daily frequency, thus providing adequate temporal coverage. However, most of the soil moisture products currently available are characterized by a coarse spatial resolution (~ 25-50 km). Clearly, such products lack the ability to monitor the variability occurring at much finer scales (~ 1-100 m), which is mainly controlled by topography, soil texture and vegetation. The aim of this study is to estimate soil moisture at sub-field resolution by combining coarse scale remotely sensed data and ancillary information. We employ data collected from dozens of low-cost sensors measuring soil moisture and incoming solar radiation, the latter being then converted to fAPAR. The sensors are installed in the Hydrological Open Air Laboratory (HOAL), a small agricultural catchment (66 ha) characterized by complex topography and different land cover, located in Petzenkirchen (Austria). Additional ancillary data, i.e. soil texture and a Digital Elevation Model (DEM), are available for the study area. We use a Random Forest regression model to estimate high-resolution soil moisture using the following input features: soil texture, topographic indices (derived from the DEM), fAPAR, and the average catchment soil moisture computed from the low-cost sensor measurements. The latter, which represents the ideal scenario where satellite-derived soil moisture perfectly agrees with ground observations, is then replaced with a coarse scale remotely sensed product (ASCAT SM, 25 km spatial resolution), and results compared. The accuracy of the model(s) is evaluated with a cross-validation.Results show the overall good accuracy of the estimated high-resolution soil moisture as compared to in-situ measurements. Clearly, using the catchment average as input provides better results than using coarse scale satellite-derived data. In fact, the latter might not be fully representative for the study area due to the large spatial mismatch between satellite footprint and catchment size. The inclusion of a proxy of vegetation dynamics (i.e. fAPAR) in all experiment setups improves the accuracy of the estimated soil moisture; indeed fAPAR is the second most important variable for the model, following average catchment or remotely sensed soil moisture. Overall, the increasing availability of low-cost sensors providing reliable measurements allows to develop catchment-specific models to estimate soil moisture at sub-field resolution. These can provide valuable information both in the spatial and temporal domain (e.g. for locations and periods where sensors are not installed, respectively). Further analysis will assess the suitability of using remotely sensed dataset for monitoring vegetation development (e.g. LAI, NDVI, etc.), and the possibility to employ such model for similar catchments. [ABSTRACT FROM AUTHOR]