During the AIDA campaign in Ny-Ålesund, Svalbard, in spring 2024, near-surface particle and heat fluxes were measured using the gradient method. The gradient system measured the number of particles, temperature and wind speed at five heights close to the surface. We categorised the surface conditions as snow, snow/vegetation or vegetation. To measure particle concentration near the surface, we used a 1.5 m high linear actuator with a sensor board. This included a heat wire anemometer, an inlet connected to a condensation particle counter (CPC), and a distance sensor. Data about solar radiation, precipitation and wind direction and speed was, among others, provided by a weather station located at the top of the linear actuator. The gradient system was set up to approach six different height levels in sequence. Depending on the height level, the distance to the surface from the inlet varied, with measurements taken at 10 cm, 14 cm, 22 cm, 38 cm, 69 cm and 134 cm. We calculated the sensible heat flux and the particle flux using the flux-gradient relationships method. Monte Carlo simulation was utilized to estimate the uncertainty of the flux estimate due to the error of the measuring instruments.

During the AIDA campaign in Ny-Ålesund, Svalbard, in spring 2025, the gradient method was used to measure near-surface particle and heat fluxes. This system measured the number of particles, temperature and wind speed at five heights close to the surface. We categorised the surface conditions as snow, snow/vegetation, or vegetation. This dataset contains preprocessed particle, temperature and wind profile data, as well as calculated particle and sensible heat fluxes. To measure near-surface particle concentration profiles, a 1.50 m high linear actuator with a sensor board was used. This included a heat wire anemometer, an inlet connected to a condensation particle counter (CPC), and a distance sensor. The temporal resolution of the wind sensor was 1 s, while that of the distance sensor was 0.5 s. The CPC measured the total particle number concentration with an lower cut-off diameter of 10 nm and a temporal resolution of 1 s, with a sample volume flow rate of 0.7 l/min.The gradient system was set up to approach six different height levels consecutively. The fourth height level was recorded twice: once during ascent and once during descent. This provided an additional reference for changes in background particle number concentration. The system remained at each height for 50 seconds, resulting in a complete profile cycle of approximately six minutes (350 seconds). As the CPC has a maximum response time of 9 s and adjusting the inlet height requires a maximum of 10 s, only the last 30 s of measured values per height level were used for data analysis. Discarding the first 20 seconds of each measurement interval at each height takes into account the response times of all the sensors, as well as the travel time of the gradient system to the newly set height. Depending on the height level, the distance to the surface measured from the inlet was 10 cm, 14 cm, 22 cm, 38 cm, 69 cm and 134 cm. A profile was defined by at least five different height levels.However, evaluating the gradients from ascending stepped profiles could introduce bias if the particle number concentration, wind speed and temperature change significantly over the course of a full profile cycle of 350 seconds. Therefore, the particle and sensible heat flux were evaluated as mean trends over 20-minute intervals. To this end, 20-minute intervals were created. For each interval, the median time point was determined, and it was verified that at least five of the six heights within the 20 minutes had been correctly measured by the gradient system. Linear interpolation based on time was used to determine the corresponding value at the median time point for the three measurements of particle number concentration, wind speed and temperature.The measured particle number concentrations were adjusted for losses of particles inside the tubing and inlet. The penetrating fraction was calculated according to the method described by Gormley and Kennedy (1949). To determine the penetrating fraction in relation to size distribution, SMPS and NAIS data measured at the Gruvebadet atmospheric observatory were used. It was assumed that this particle size distribution was also representative for the gradient system measurements.The sensible heat flux and particle flux were calculated using the flux-gradient relationship method (Foken, T., 2017). Monte Carlo simulation was used to estimate the uncertainty in the flux estimate caused by errors in the measuring instruments. For each profile, 10,000 iterations were generated using Gaussian-distributed noise. The assumed standard deviations were 20% of the particle concentration, 5% of the temperature, and 0.5 m/s for wind speed. Fluxes were recalculated for each ensemble, and the 90th percentile was taken as the uncertainty range.In the case of stable stratification, strong temperature gradients can occur very close to the surface. To exclude periods with weak turbulence and thus a high probability of vertical decoupling, the calculated fluxes can be classified according to u. One possible filter would be u ≥ 0.15 m/s. Another quality criterion is the variable R² of the linear regressions for particle concentration, wind speed and temperature, which can also be used as a filter.