Revealing climatic variability of the last three millennia in northwestern Iberia using pollen influx data

Abstract

Climatic variability of the last 3 millennia in NW Iberia has been documented using high-resolution pollen analysis of Vir-18 core, retrieved from the Rı́a de Vigo (42°14.07′N, 8°47.37′W). The depth–age model is based on two accelerator mass spectrometry ¹⁴C dates and three historically dated botanical events in Galicia: the expansion of Juglans and Pinus, as well as the introduction of Eucalyptus. During the last 3000 years, the relative pollen record demonstrates the occurrence of an open deciduous oak forest, indicating a humid and temperate climate in northwestern Iberia. Two-step forest reduction since 975 cal BC suggests climate as the main cause rather than major socio-economic changes documented in historical archives. Absolute pollen influx has been compared with instrumental summer and winter temperatures and tentatively used as a proxy of short (decadal-scale) and low-amplitude (∼1°C) temperature variations. This new approach allows us to detect for the first time in NW Iberia the millennial-scale climatic cyclicity suggested by North Atlantic records, challenging the apparent climatic stability reflected by the relative pollen record. The Little Ice Age is recorded as low pollen influx values between 1400 and 1860 cal AD, with a cold maximum at 1700 cal AD (Maunder Minimum). The Roman and Medieval Warm Periods are detected through high pollen influx values at 250 cal BC–450 cal AD and 950–1400 cal AD, respectively.

Introduction

Recent studies on marine records from the North Atlantic Ocean [1], [2], [3], [4], [5] have revealed that the Holocene was characterised by a millennial-scale climatic variability. This variability was of weak amplitude, estimated at ca. 2°C in the sea surface temperatures [1]. Changes in ocean circulation and solar radiative budget are the two mechanisms frequently quoted to explain this climatic variability [3], [4], [6], [7], [8]. A focus on the climate of the last millennium (for example [9], [10]) shows a relatively warm period, the Medieval Warm Period (MWP), centred on the 11th century and a relatively cold period, the Little Ice Age (LIA) between the 16th and 19th centuries. The worldwide nature of these events and their chronology are still discussed.

Little is known, however, about the impact of this millennial-scale climatic variability on the European continent and, in particular, the Iberian Peninsula. In this region, Martinez-Cortizas et al. [11] identified several major climatic changes during the last 4000 yr using the analysis of mercury in the peat-bog sequence of Penido Vello, Galicia. However, they neither discuss the chronology of these events nor the frequency of such climatic variability. Valero-Garcés et al. [12] detected in the Salada Mediana and Salada Chiprana records (Ebro valley, NE Iberia) a rise in lake levels after the 14th–15th centuries, coinciding with the end of the MWP. Recently, the high-resolution analysis of the upper part of the Lake Sanabria core (NW Iberia) [13] revealed a sedimentary episode interpreted as the LIA. Historical archives, in turn, only provide information about the impact of the Maunder Minimum on Iberian temperature and precipitation [14], [15].

The examination of pollen records from Iberia shows that most (for example [16], [17], [18], [19], [20], [21]) lack sufficient temporal resolution to detect millennial-scale climatic changes. Vegetational shifts recorded in high-resolution pollen diagrams (20–300 yr between samples) are ambiguously attributed to human activity and/or climatic variations [22], [23], [24], [25], [26] because the criteria are missing to disentangle these two factors [27], [28]. Other authors consider that human activity is the only factor accounting for the vegetational changes detected over the late Holocene [29], [30], [31]. In any case, most agree that human impact in pollen records is clearly observable since 1000 cal AD. In the early Holocene, when anthropogenic action was less significant, the millennial-scale climatic fluctuations detected in the North Atlantic marine sequences are not recorded by the pollen percentage diagrams. This may be explained by the fact that such weak changes affect the vegetation composition of southwestern Europe only slightly. However, is the traditional use of pollen percentage diagrams the appropriate approach to detect rapid (less than 100 yr) climatic changes of small amplitude such as those punctuating the Holocene period?

The goal of this work is to show that absolute pollen influx in sedimentary sequences (referred to as pollen influx in the text) can be used to identify rapid and weak amplitude climatic changes on the continent. For this, we use a well-dated high-resolution core, Vir-18, retrieved from the Rı́a de Vigo (NW Iberia), covering the last 3000 yr. The Vir-18 core is especially suitable for this kind of study because: (a) it comes from a fluvial system and it has been demonstrated that pollen assemblages in river water will vary seasonally with the production of pollen and spores in the source basin [32], (b) its geographical location in the northwestern Iberian Peninsula should mean its climate is affected by North Atlantic sea surface conditions, (c) its fluviatile pollen input provides an integrated image of the regional vegetation [32], [33] directly related to climatic parameters or to large-scale human impact.

Previous studies on Vir-18 core (benthic and planktonic foraminiferal assemblages, stable oxygen isotopes, molecular biomarkers and SST-alkenone derived) established the hydrographic evolution of the ria during the last 3000 yr [34]. This work identifies only one major hydrographic change at around 1000 yr AD, the MWP and the LIA. Our complementary work will reveal that a millennial-scale climatic cyclicity characterised the last 3 millennia in northwestern Iberia.