An operational robotic pollen monitoring network based on automatic image recognition.

There is high demand for online, real-time and high-quality pollen data. To the moment pollen monitoring has been done manually by highly specialized experts. Here we evaluate the electronic Pollen Information Network (ePIN) comprising 8 automatic BAA500 pollen monitors in Bavaria, Germany. Automatic BAA500 and manual Hirst-type pollen traps were run simultaneously at the same locations for one pollen season. Classifications by BAA500 were checked by experts in pollen identification, which is traditionally considered to be the "gold standard" for pollen monitoring. BAA500 had a multiclass accuracy of over 90%. Correct identification of any individual pollen taxa was always >85%, except for Populus (73%) and Alnus (64%). The BAA500 was more precise than the manual method, with less discrepancies between determinations by pairs of automatic pollen monitors than between pairs of humans. The BAA500 was online for 97% of the time. There was a significant correlation of 0.84 between airborne pollen concentrations from the BAA500 and Hirst-type pollen traps. Due to the lack of calibration samples it is unknown which instrument gives the true concentration. The automatic BAA500 network delivered pollen data rapidly (3 h delay with real-time), reliably and online. We consider the ability to retrospectively check the accuracy of the reported classification essential for any automatic system.

Building an automatic pollen monitoring network (ePIN): Selection of optimal sites by clustering pollen stations.

Airborne pollen is a recognized biological indicator and its monitoring has multiple uses such as providing a tool for allergy diagnosis and prevention. There is a knowledge gap related to the distribution of pollen traps needed to achieve representative biomonitoring in a region. The aim of this manuscript is to suggest a method for setting up a pollen network (monitoring method, monitoring conditions, number and location of samplers etc.). As a case study, we describe the distribution of pollen across Bavaria and the design of the Bavarian pollen monitoring network (ePIN), the first operational automatic pollen network worldwide. We established and ran a dense pollen monitoring network of 27 manual Hirst-type pollen traps across Bavaria, Germany, during 2015. Hierarchical cluster analysis of the data was then performed to select the locations for the sites of the final pollen monitoring network. According to our method, Bavaria can be clustered into three large pollen regions with eight zones. Within each zone, pollen diversity and distribution among different locations does not vary significantly. Based on the pollen zones, we opted to place one automatic monitoring station per zone resulting in the ePIN network, serving 13 million inhabitants. The described method defines stations representative for a homogeneous aeropalynologically region, which reduces redundancy within the network and subsequent costs (in the study case from 27 to 8 locations). Following this method, resources in pollen monitoring networks can be optimized and allergic citizens can then be informed in a timely and effective way, even in larger geographical areas.