Effect of sampling effort on the regional chronology statistics and climate–growth relationships estimation

Regional chronology and climate–growth relationships assessment are known to be sensitive to sampling effort. To disentangle the respective benefits of increasing whether the number of plots or the number of trees per plot when investigating climate–growth relationships under temperate conditions, we propose to simulate samples from a set of 84 plots established in plantations of Corsican pine (Pinus nigra Arnold ssp. laricio Poiret var. Corsicana), within which 10 dominant trees were cored in 1992. The effect of sampling effort was investigated through 20 modalities of sampled plots (from 3 to 84) and 10 modalities of sampled trees per plot (from 1 to 10). Regional chronology was studied using the mean effective correlation and the expressed population signal, while climate–growth relationships were evaluated through correlation functions. The calculation of the correlation functions also allowed testing the effect of sampling effort on various climatic regressors presenting different climate–growth correlation strengths. The accuracy of the dendroecological investigations increased with increasing sample size: estimating climate sensitivity with a regional chronology built from a small sample led to a general under-estimation of the climate–growth correlations. Decreasing sample size also increased the risk of estimating (i) “false” non-significant correlations for the most influencing climatic regressors, and (ii) “false” significant correlations for the less influencing ones. Increasing the number of plots was found of a greater interest than increasing the number of trees per plot to improve the climate–growth relationships assessment. Finally, the analysis revealed that the improvement of the expressed population signal with increasing sampling effort did not linearly mirror the bettering in the climate–growth correlations assessment.