Mediterranean pine forests: Comparison of fluxes and tree rings of Pinus pinaster Aiton and Pinus pinea L

• Two species of Pinus have been compared, using fluxes and tree ring measurements. • The more drought-tolerant species had lower productivity and lower water efficiency. • Fluxes and tree rings are in agreement, but the correlation is weak. Flux-monitoring networks nowadays integrate several stations, with more than 20 years of records. However, many questions are still open, and this work focused on the following two: (i) what is known about carbon and energy exchanges and efficiencies of evergreen conifer forests in the hot and dry Mediterranean climate zone, and (ii) what can be learned by combining information derived from flux and tree ring measurements in this context? These issues are addressed in this study using micrometeorological and dendrochronological measurements from a Mediterranean pine forest, San Rossore, in central Italy. The forest’s flux time series exceeds 20 years, merging two consecutive (not overlapping) time series of flux measurements above two evergreen conifer species: Pinus pinaster Aiton, 1789 (maritime pine) and Pinus pinea L., 1753 (stone pine). Despite both conifer species having shown high rates of photosynthesis, substantial differences have been found between them in terms of annual productivity, hydrological balance and efficiencies of carbon and water use. The more drought-tolerant Pinus pinea L. has been less productive than the drought-avoiding Pinus pinaster Aiton: interannual net ecosystem productivity averages and standard errors have been estimated to be 441 ± 46 gC m –2 y –1 and 224 ± 35 gC m –2 y –1 for Pinus pinaster Aiton and Pinus pinea L., respectively, and the latter has been found to consume more water as a consequence of the sustained transpiration process during prolonged summer droughts. Annual carbon accumulation, based on eddy covariance flux measurements, and annual wood increments, based on tree ring widths, for the two species are in overall alignment in terms of absolute values; however, a weak correlation suggests the need for more detailed analyses to gain more information about the carbon dynamics of the ecosystem, because the two approaches do not quantify exactly the same processes.