In-situ high resolution aircraft measurements of cloud microphysical properties were made in coordination with ground based remote sensing observations of Radar and Lidar as part of the Aerosol Properties, PRocesses And InfluenceS on the Earth's climate (APPRAISE) project. A narrow but extensive line (~100 km long) of shallow convective clouds over the southern UK was studied. Cloud top temperatures were observed to be higher than ~&minus;8 °C, but the clouds were seen to consist of supercooled droplets and varying concentrations of ice particles. No ice particles were observed to be falling into the cloud tops from above. Current parameterisations of ice nuclei (IN) numbers predict too few particles will be active as ice nuclei to account for ice particle concentrations at the observed near cloud top temperatures (~&minus;7 °C). The role of biological particles, consistent with concentrations observed near the surface, acting as potential efficient high temperature IN is considered important in this case. It was found that very high concentrations of ice particles (up to 100 L^−1) could be produced by powerful secondary ice particle production emphasising the importance of understanding primary ice formation in slightly supercooled clouds. <br><br> Aircraft penetrations at −3.5 °C, showed peak ice crystal concentrations of up to 100 L^−1 which together with the characteristic ice crystal habits observed (generally rimed ice particles and columns) suggested secondary ice production had occurred. To investigate whether the Hallett-Mossop (HM) secondary ice production process could account for these observations, ice splinter production rates were calculated. These calculated rates and observations could only be reconciled provided the constraint that only droplets >24 μm in diameter could lead to splinter production, was relaxed slightly by 2 μm. <br><br> Model simulations of the case study were also performed with the WRF (Weather, Research and Forecasting) model and ACPIM (Aerosol Cloud and Precipitation Interactions Model) to investigate the likely origins of the ice phase in these slightly supercooled clouds and to assess the role played by the HM process in this and in controlling precipitation formation under these conditions. <br><br> WRF results showed that while HM does act to increase the mass and number concentration of ice particles produced in the model simulations, in the absence of HM, the ice number concentration arising from primary ice nucleation alone (several L^−1) was apparently sufficient to sustain precipitation although the distribution of the precipitation was changed. Thus in the WRF model the HM process was shown to be non-critical for the formation of precipitation in this particular case. However, this result is seen to be subject to an important caveat concerning the simulation of the cloud macrostructure. The model was unable to capture a sharp temperature inversion seen in the radiosonde profiles at 2 km, and consequently the cloud top temperature in the model was able to reach lower values than observed in-situ or obtained from satellite data. ACPIM simulations confirmed the HM process to be a very powerful mechanism for producing the observed high ice concentrations, provided that primary nucleation occured to initiate the ice formation, and large droplets were present which then fell collecting the primary ice particles to form instant rimer particles. However, the time to generate the observed peak ice concentrations was found to be dependant on the number of primary IN present (decreasing with increasing IN number). This became realistic (around 20 min) only when the temperature input to the existing IN parameterisation was 6 °C lower than observed at cloud top, highlighting the requirement to improve basic knowledge of the number and type of IN active at these high temperatures. In simulations where cloud droplet numbers were realistic the precipitation rate was found to be unaffected by HM, with warm rain processes dominating precipitation development in this instance.