Climate variability and human impact in South America during the last 2000 years: synthesis and perspectives from pollen records

. An improved understanding of present-day climate variability and change relies on high-quality data sets from the past 2 millennia. Global efforts to model regional climate modes are in the process of being validated against, and integrated with, records of past vegetation change. For South America, however, the full potential of vegetation records for evaluating and improving climate models has hitherto not been sufﬁciently acknowledged due to an absence of information on the spatial and temporal coverage of study sites. This paper therefore serves as a guide to high-quality pollen records that capture environmental variability during the last 2 millennia. We identify 60 vegetation (pollen) records from across South America which satisfy geochrono-logical requirements set out for climate modelling, and we discuss their sensitivity to the spatial signature of climate modes throughout the continent. Diverse patterns of vegetation response to climate change are observed, with more similar patterns of change in the lowlands and varying intensity and direction of responses in the highlands. Pollen records display local-scale responses to climate modes; thus, it is necessary to understand how vegetation–climate interactions might diverge under variable settings. We provide a qualitative translation from pollen metrics to climate variables. Additionally, pollen is an excellent indicator of human impact through time. We discuss evidence for human land use in pollen records and provide an overview considered useful


Introduction
Accurately simulating the complexity of Earth's climate system is still a major challenge for even the most advanced Earth system models. One major obstacle for evaluating model performance in historical runs is the lack of long and reliable climate records from some regions of the Earth. Given the scarcity of instrumental records in many regions, alternative, proxy-based climate reconstructions therefore provide an excellent data set against which to test models and their ability to accurately simulate longer-term features of climate change. Proxy data sets from sedimentary records (in particular pollen, charcoal, and tephra from lake sediments and peat bogs) have been particularly underutilized in this regard.
Increasingly studies have demonstrated the integration of multiple proxies (Li et al., 2010) in climate reconstructions, with a special focus on the last 2 millennia. This period could be considered a baseline to current conditions, as climate has been very similar to the present. This integration is still in its infancy in South America (SA), especially in the tropics. Since 2009, regional climate reconstructions from SA have gained momentum from compilations of multiple data sets and from fine-tuning of model reconstruction methods . However, an improved understanding of the spatial distribution of proxy data sets has been identified as necessary to make further progress Flantua et al., 2015). Tree ring studies constitute a widely distributed and frequently used high-resolution climate archive that has fortunately recently expanded its spatial coverage (Boninsegna et al., 2009;Villalba et al., 2009). However, the tree ring records are limited compared to the spatial and temporal coverage provided by records obtained from sedimentary archives (e.g. pollen records). The newly updated inventory of palynological research in SA documents the extensive spatial and temporal coverage of pollen-based research available throughout the continent (Flantua et al., 2015). However, to integrate records from different sedimentary archives across SA, a standard chronological framework is required. To this end an alternative recalibrated age model and evaluation of chronologies have been undertaken to facilitate the integration of multi-proxy records in SA (Flantua et al., 2016). However, multi-proxy climate reconstructions from the last 2 kyr have hitherto been focused mainly on southern SA (PAGES-2k Consortium, 2013), omitting input from the northern two thirds of the continent. Furthermore, palynological research has been underrepresented in most reconstructions of climate variability Neukom et al., 2010;Neukom and Gergis, 2012). The lack of an adequate overview of available pollen records from the continent has been an impediment to the advancement of its use and inclusion in climate studies.
As a result, we identified the need to review and discuss pollen records in SA that can fulfil requirements for inclusion in 2 ka palaeoclimate reconstructions, within the framework of LOng-Term multi-proxy climate REconstructions and Dynamics in South America (LOTRED-SA, this Special Issue) and the PAGES-2k Network (http://www.pages-igbp.org/ini/ wg/2k-network/intro). This paper is structured following an assessment for individual regions in SA within the context of current climate modes. These modes are characterized by their precipitation and temperature fingerprint over SA and used as a baseline framework to identify past climatic changes from pollen records. Certain zones are more prone to particular climate signals; therefore, comparison between the spatial expression of climate modes and highly correlated records from different regions strengthens the interpretation of palaeoecological findings. To use pollen as a palaeoclimate proxy, the degree of human impact on the vegetation needs to be considered as being minimal or absent over the last 2 kyr. Therefore, drivers of vegetation change, both natural and anthropogenic, are discussed within the different regions to describe the general settings required for palaeoecological research in the last millennia. Records that identify significant human impact are identified and excluded from the proposed data set for PAGES-2k when the climate signal is lost but are considered useful within the regional purposes of LOTRED-SA (this Special Issue). We provide a qualitative translation from pollen metrics to climate variables based on expert knowledge. We finish by discussing the potential of including pollen-inferred climate information in 2 ka climate model validation and emphasize the importance of multi-proxy working groups such as LOTRED-SA.

Continental overview climate zones and modes
We begin with an overview of the main climate "zones" of SA to provide the climatological context for a discussion of pollen records covering the past 2000 calibrated years before present (2 cal yr BP). Climate zones are regions of coherent seasonality and mean climate (intra-annual climate regime), while climate "modes" are based on ocean-atmosphere interactions, with often oscillatory behaviour affecting the interannual to multidecadal climate variability in a region. The Clim. Past, 12, 483-523, 2016 www.clim-past.net/12/483/2016/ spatial influence of climate modes is assessed by documenting their role in driving interannual precipitation and temperature variability. Continental SA extends from the tropics (12 • N) to midlatitudes (55 • S). Three major noticeable climate zones can be distinguished: tropical, subtropical, and extratropical SA. Atmospheric circulation and climate in all three zones is highly modulated and constrained by the orography of the Andes, the shape of the continent, and interactions with the underlying land surface, vegetation, and soil moisture; furthermore, ocean currents, such as the cold Humboldt Current affecting coastal climate along the South American west coast, also affect climate (Wang and Fu, 2002;Li and Fu, 2006).
The climate of tropical SA is dominated by the seasonal migration of the Intertropical Convergence Zone (ITCZ) over the Atlantic and Pacific, and the seasonal development of convective activity associated with the South American summer monsoon (SASM) over the interior of the continent (Fig. 1). The seasonal migration of the ITCZ primarily affects coastal areas and northernmost SA as it is characterized by a fairly well-constrained narrow band of low-level wind convergence over the equatorial oceans. The SASM is a seasonal phenomenon that develops between September and April and primarily affects the SA tropics and subtropics south of the Equator (Garreaud et al., 2009). During the austral spring (September to November, SON) moisture influx from the ITCZ contributes to the development of this monsoon system (Zhou and Lau, 2001;). This monsoonal system reaches its mature phase (maximum development) during December to February (DJF) and is characterized by heavy rainfall advancing southward from tropical to subtropical latitudes. To the east of the tropical Andes a strong low-level wind, the Andean low-level jet (ALLJ), transports moisture in a southeasterly direction from the tropics to the subtropical plains (Cheng et al., 2013), feeding the South Atlantic convergence zone (SACZ), extending from the southeastern (SE) Amazon basin toward the southeast out over the South (S) Atlantic. The extratropical region is characterized by a quasi-permanent westerly circulation embedded in between the subtropical anticyclones located over the subtropical Pacific and Atlantic to the north and the circumpolar trough of low pressure to the south. Frequent northward propagation of extratropical cold air incursions east of the Andes provide for continued atmospheric interaction and heat exchange between mid-and low latitudes over the subtropical continent. The latitudinal extension of the westerlies over land displays limited variations across the year and covers southern and central (C) Argentina and Chile. Additional information is presented in the Supplement.
Both precipitation and temperature exhibit significant variability on interannual to interdecadal timescales in all three climate zones of SA (e.g. Garreaud et al., 2009). This variability is mainly caused by ocean-atmosphere interactions (Vuille and Garreaud, 2012) that lead to a reorganization of the large-scale circulation over SA and the neighbouring oceans. To quantify the influence and relative importance of these ocean-atmosphere coupled modes on the interannual precipitation and temperature variability over SA, spatial correlation and regression coefficients are calculated.
Gridded precipitation and temperature data were derived from the UDelaware data set V2.01 (Legates and Willmott, 1990) at 0.5 • resolution. We limit our assessment to the six most relevant climate modes (Table 1). Other modes analysed were either largely redundant or showed a much weaker influence over the SA continent. The resulting correlation maps indicate the correlation coefficient on interannual timescales between the mode in question and the local temperature and precipitation at each grid cell. Conversely, the regression maps indicate the local anomaly (in physical units of millimetres or degrees Celsius) at each location that corresponds to a unit (1 standard deviation) anomaly in the climate mode. The Southern Annular Mode (SAM) and all three Atlantic modes (Atlantic Multidecadal Oscillation -AMO; tropical North and South Atlantic sea surface temperature -TSA and TNA; Table 1) were detrended prior to analysis to ensure that correlation and regression coefficients account for co-variability on interannual timescales only and do not result from spurious common trends. More information on the methodology can be found in the Supplement.
In all correlation maps (Figs. 2 and 4) we show correlations in excess of ±0.2 only, which approximately corresponds to the 95% significance level. For the regression maps (Figs. 3 and 5), we used thresholds of ±0.12 • C and ±50 mm, respectively. The correlation maps can help inform www.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016  Enfield et al. (1999) us whether a certain temperature or precipitation anomaly in the regression map is statistically significant. In our discussion we focus primarily on the impact of the positive phase from each of these modes, as these are the fingerprints presented in Figs. 2-5. Since this is a linear analysis, the negative phase of these modes would lead to the same changes in temperature and precipitation but with the sign reversed. In general these outcomes are consistent with earlier analyses reported by Garreaud et al. (2009)

Temperature
The largest and most significant influence on interannual temperature variability in SA is exerted by ENSO, with above average temperatures during El Niño and reduced temperature during La Niña (Figs. 2 and 3). A 1 standard de-viation departure in the Niño3.4 index is associated with a change in temperature of up to 0.8 • C along the Pacific coast of SA. In the Andes of Colombia, the correlation between temperature and the Niño3.4 index is > 0.8, indicating that more than two thirds of the temperature variability on interannual scales can be explained by ENSO. The largest increase in temperature is observed during austral summer (DJF, not shown), linked to the peak phase of ENSO, which tends to occur at the end of the calendar year. Compared to ENSO, the IPO has a similar, albeit slightly weaker, fingerprint over SA, which is not surprising given that the Pacific decadal and multidecadal variability is often described as "ENSO-like" (e.g. Garreaud and Battisti, 1999). The IPO impact extends further south along the west coast of SA than ENSO, however, with a somewhat stronger influence on temperature in northern (N) to C Chile. It is noteworthy that the IPO impact over SA is almost identical to the influence of the Pacific Decadal Oscillation as described in Garreaud et al. (2009).
The N Atlantic modes, AMO and TNA are also quite similar, both featuring warming over tropical SA during periods when sea surface temperature (SST) in the N Atlantic domain is above average, most notably so over the southern C Amazon basin (Figs. 2 and 3). In fact the warming associated with a unit variation in the AMO or TNA index is larger over most of the Amazon basin than the warming associated with ENSO. The region of largest warming is colocated with an area of strong precipitation reduction during the warm phase of the TNA and the AMO (Figs. 4 and 5). This suggests that much of the warming is caused by cloud cover and soil moisture feedbacks associated with reductions in precipitation (reduced cloud cover leading to enhanced solar radiation and reduced soil moisture limiting evaporative cooling). The S Atlantic counterpart, the TSA, is associated with a temperature dipole over subtropical SA, characterized by warming along a zonal band extending from the S-C Brazilian coast westward to Bolivia, while C Argentina contemporaneously experiences cooling (Figs. 2 and 3). The warming in the subtropical region coincides with a region of reduced precipitation during the TSA positive phase (Fig. 4), suggesting that the warming is at least in part caused by changes in the hydrological cycle (cloud cover and/or soil moisture feedbacks).
The SAM is positively correlated with temperature over Patagonia (Fig. 2) and also shows a weak negative temperature departure over western tropical SA during its positive phase (Fig. 3). The warming over Patagonia is strongest during austral summer (Garreaud et al., 2009; not shown) and High negative values indicate that the positive (negative) phase of a mode leads to a decrease (increase) in temperature at the grid cell. Gridded temperature fields are from the University of Delaware (1958-2008). results from enhanced heat advection, combined with higher solar radiation receipts due to cloud-free conditions (Gupta and England, 2006).

Precipitation
Given that ENSO is the source of the strongest interannual variability on Earth, it is not surprising that it also leads to the strongest modern precipitation anomalies over SA (Fig. 5). In general in the tropics, El Niño events lead to significant precipitation reductions over much of tropical SA, with the strongest signal seen in N Brazil along the Atlantic coast and in the Andes of Colombia. Over northeast (NE) Brazil the precipitation reduction is the result of El Niño events inducing a delayed anomalous warming of the tropical N Atlantic in boreal spring (March-May) (e.g. Curtis and Hastenrath, 1995;Giannini et al., 2001). Hence, the ENSO influence in this region strongly projects onto the TNA pattern (Fig. 4). Over the N Amazon basin the precipitation reduction is the result of a shifted Walker circulation, enhanced subsidence, and reduced convective activity (e.g. Liebmann and Marengo, 2001;Ronchail et al., 2002). In the subtropics, on the other hand, precipitation is enhanced during El Niño events, in particular over southeastern SA (see also Grimm et al., 2000). The only tropical location that sees an increase www.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 in precipitation during El Niño is along the Pacific coast of Ecuador and northern Peru, where flooding is a common occurrence during these events (e.g. Takahashi, 2004). During La Niña events these precipitation anomalies are essentially reversed. The correlations are weaker in our annual analysis over some regions where the ENSO influence is highly seasonal, such as the precipitation reduction over the C Andean "Altiplano" (high plain) region in DJF  or the enhanced precipitation during El Niño in C Chile in June to August (JJA; Montecinos and Aceituno, 2003). The largest change in the IPO in the period analysed is related to the Pacific climate shift of 1976-1977, when the tropical Pacific switched from its cold to its warm phase. Since El Niño events also became more frequent and stronger over this period (including the two extreme events of 1982-1983 and 1997-1998), it is no surprise that the observed changes in precipitation associated with the IPO are similar to the ENSO footprint, albeit somewhat weaker. Indeed the low-frequency modulation by the IPO may strengthen El Niño events during its positive phase and weaken La Niña events, while the opposite is the case during the IPO negative phase, a phenomenon known as "constructive interference" (e.g. Andreoli and Kayano, 2005). Espinoza Villar et al. (2009) documented the influence of Pacific interdecadal variability on precipitation over the Amazon basin and showed that its positive phase is related to a decrease in precipitation over the basin since 1975, consistent with our results.
Precipitation is reduced in the southernmost part of SA during the positive phase of the SAM (Fig. 4). This reduction extends north into the subtropics along both the Atlantic and Pacific coast to approximately 30 • S (Silvestri and Vera, 2003;Gillett et al., 2006). Most of this precipitation reduction is associated with reduced westerly moisture flux and moisture convergence from the Pacific (Garreaud et al., 2013). The correlation (Fig. 4) and regression (Fig. 5) maps also suggest a significant influence of the SAM on precipitation in parts of the tropics. This signal, however, is not well documented and its physical mechanism is unclear. It may to some extent be related to teleconnections and an anticorrelation between ENSO and the SAM (e.g. Carvalho et al., 2005), which is supported by the fact that the Niño3.4 index and the SAM correlation maps are almost mirror images of one another (Fig. 4).
The AMO and the TNA have a similar fingerprint on the hydrologic cycle of SA (Fig. 5). Both modes are characterized by a significant reduction in precipitation over much of the Amazon basin during their positive phase, with the amplitude of the changes associated with TNA forcing being slightly larger. This negative precipitation anomaly is as-Clim. Past, 12, 483-523, 2016 www.clim-past.net/12/483/2016/ Records longer than 1 kyr must include a minimum of one additional age midway between the other two.
Records longer than 1 kyr must include minimum of one additional age midway between the other two.
1000_MID3 sociated with the northward displacement of convective activity in the ITCZ region due to warmer SST in the tropical North Atlantic and Caribbean during the positive phase of the TNA (and to a lesser extent also the AMO). This directly affects precipitation amounts over NE Brazil (e.g. Hastenrath and Greischar, 1993;Nobre and Shukla, 1996), while the northward shift in the core region of convection also leads to anomalous subsidence, located over the Amazon basin. In fact the recent droughts in 2005 and 2010 in the Amazon basin were both associated with such anomalously warm SST in the tropical N Atlantic (Marengo et al., 2008;Lewis et al., 2011). The only region where precipitation is enhanced is in the northwestern part of the Amazon, belonging to Venezuela, Colombia, and Peru (Fig. 4). An anomalously warm tropical S Atlantic (positive phase of the TSA) leads to the exact opposite conditions, with the ITCZ displaced anomalously far south, causing copious rainfall over NE Brazil, with weaker positive anomalies extending inland as far as the Peruvian border (Fig. 5). Another region of enhanced precipitation is located in S Brazil, associated with a southerly movement of the SACZ ( Fig. 1; e.g. Doyle and Barros, 2002).

Selection of pollen records for 2 ka
Within the working groups of PAGES, the "2k Network" was initially established in 2008 to improve current understanding of temperature variability across the Earth during the last 2 kyr. To collate records across the Earth for this time period systematically a set of criteria that defined the suitability of individual records was required. The principle of the criteria was to ensure, as far as possible, consistency (and therefore comparability) in the chronological control and sam-pling resolution of fossil pollen records (Table 2). Of the six PAGES-2k criteria within this paper, we regarded criterion A (peer-reviewed publication) as the base line criterion (all sites considered are from peer-reviewed studies). However, implementation of criterion B (resolution ≤ 50 years) was not possible for SA because such a criterion would leave only a handful of pollen records to discuss. The sparsity of samples that meet the stringent PAGES-2k resolution criterion occurs because sedimentary archives with long time spans (> 10 000 yr) are typically sampled at coarser temporal resolution. Furthermore, many lowland sites have slow sedimentation rates, which preclude high-resolution sampling. Therefore, we propose a more flexible temporal resolution, depending on the identified relevance of the case study.
From the newly updated Latin American Pollen Database (LAPD; Flantua et al., 2015), we initially selected all records that cover the last 2 kyr (Fig. 6). Good chronological control is required for PAGES-2k, but the youngest ages in pollen records are typically constrained by geochronological data. An assessment of the pollen records by the authors with expertise in each SA subregion has revealed 585 records with pollen samples within the last 2 kyr (Fig. 6), of which 337 and 182 records, respectively, contain one or more geochronological dates within that time period. In total, 182 studies were checked to confirm their suitability for palaeoclimate reconstruction as outlined by the PAGES-2k criteria. Records with a resolution of 200 to 300 yr are included in our discussion, while records along coastlines influenced by sea level changes were not included. Within the regional assessments, only records that fulfil more than three criteria are discussed, unless the records are considered particularly valuable for regional climate assessments.

Regional assessments
Pollen records are discussed according to their regional and geographical settings ( Fig. 6): A -Venezuelan Guayana highlands and uplands (for the purposes of this study, we have defined this region as "Guayana" (different from "Guiana") following Huber's criteria (1995a,b) in the reference botanical guide to the "Flora of the Venezuelan Guayana"); B -Northern Andes; C -Central Andes; D -lowland Amazon basin; E -southern and southeastern Brazil; F -pampean plain; G -Southern Andes and Patagonia. The references to all records discussed are presented in Table 3.

Climate-vegetation interaction in the Venezuelan Guayana highlands and uplands
The study area, also known as the Gran Sabana (GS), is located in SE Venezuela between the Orinoco and Amazon basins ( Fig. 6 Box A; Huber and Febres, 2000). Huber (1995a) recognized three main elevational levels on the Venezuelan Guayana: lowlands (0-500 m above sea level, m a.s.l.), uplands (500-1500 m a.s.l.), and highlands (1500-3000 m a.s.l.). Lowlands are absent in the GS, which is mainly characterized by a continuous upland peneplain spiked with isolated highlands (table mountains, "tepuis"). The GS highlands are part of the so-called Pantepui phytogeographical province, which is characterized by unique biodiversity and endemism patterns, encompassing all the tepui summits above 1500 m a.s.l. (Huber, 1994;Berry et al., 1995). The tepuian vegetation is characterized by a mosaic of bare rock, pioneer vegetation, tepuian forests, herbaceous formations, and shrublands (Huber, 1995b). Additional background information is provided in the Supplement. In the GS, 22 pollen records cover the last 2 kyr. There are four records with a chronology based on one control point and an additional 10 records for which most, or all, control points are older than 2 ka. Three potentially suitable records originate from the highlands -Eruoda PATAM6-A07, Churí Chim-2 and Apakará PATAM9-A07 -and only one is found in the uplands: Laguna Encantada PATAM4-D07 peatland ( Fig. 7a; Table 3). Of the three records of the highlands, just Eruoda provides sufficiently high resolution to explore the objectives proposed here. However, only Churí Chim-2 and Apakará contain several age control points within the last 2 kyr, and Laguna Encantada presents a relatively low sampling resolution of 200 to 300 yr.
The criteria for chronological control excluded some of the most relevant work for the research questions posed by this paper. For example, the vegetation at the Eruoda summit has persisted unchanged during the last ∼ 2.5 kyr. This constancy can be extended to all the tepuian summits studied so far for the last 6 kyr (except Churí). Equally of high importance is the Urué record in the uplands, which does not meet the dating control constraints, but the sampling resolution is high enough to provide important insights into the vegetation-climate dynamics during the last 2 kyr, and will be therefore be presented here.
The Eruoda summit represents an important reference to which almost all the tepuian summit vegetation dynamics can be compared (Fig. 7b). Based on the absence of human activities in these summits, it can be assumed that the vegetation dynamics observed in the fossil records are fully climate driven and therefore valuable for LOTRED-SA. In general, these summits are insensitive to temperature change (for 2 ka), whereas moisture variations potentially may cause small internal reorganizations of plant associations although these shifts are considered to be of minor ecological significance. Shifting river courses are considered to influence local vegetation patterns through the lateral movement of gallery forests in landscapes (Rull, 2005a, b  tions transitional to savanna. Savannas were fully established around 0.9 ka, coinciding with the beginning of a phase of lower charcoal values, and continued as the dominant plant association until the present day. Savannas were accompanied by Mauritia flexuosa palm swamps ("morichales"), which established a phase that was likely more humid. These palm swamps varied greatly in extent through time, showing a parallel between the lowest palm abundance and the occurrence of two drought intervals. These two drought intervals were centred on 0.65-0.55 and 0.15-0.05 ka, coeval with the Little Ice Age (LIA) signal observed in the Venezuelan Andes (Rull et al., 1987;Rull and Schubert, 1989;Polissar et al., 2006). Generally, the vegetation dynamics recorded so far in the Venezuelan Guayana uplands have shown a higher sensitivity to changes in the available moisture than to potential shifts in the average temperatures. The last 2 kyr have been mainly characterized by vegetation change on a local scale.

Climate-vegetation interaction in the Northern Andes
The region of the N Andes consists in political terms of Colombia, Ecuador, and Venezuela and includes a wide range of different ecoregions ( Fig. 6 Box B). Sharing both the Caribbean and the Pacific coastline and various climate influences, Colombia has a unique pattern of different ecosystems shared with neighbouring countries. Pollen records are found throughout a wide range of biomes and elevations (Flantua et al., 2015), from the tropical rainforest and mangroves along the coast to the high Andean "páramos". The complex formation of the Andes with the three mountain ridges characterizes this region with numerous valleys and watersheds. A total of 64 records are available that present pollen data within the last 2 kyr. Unfortunately, 14 were presented in publications without a peer-review procedure or presented only as a summary diagram (7 records with 4 positive criwww.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 ). An additional five records, which fulfilled all criteria, suggested human presence from before 2 ka and were therefore excluded for climate reconstructions. From the remaining records, only four lakes lack human interference during the last 2 kyr. The others describe human indicators over limited periods of time and are considered valuable for PAGES-2k purposes (Table 3). Lake Valencia (Figs. 6 Box B and 8a) is represented by three cores with varying quality in chronology and resolution. The last 2 kyr are characterized by a decline in forest cover, which attains the lowest values of the Holocene and gives way to savannas. Aquatic proxies indicate declining lake levels and increasing nutrient input, a trend that accelerated during the last 0.5 kyr, when human activities were more intense around the lake. Considering the entire Lateglacial-Holocene record, the Lake Valencia catchment has been shown to be more sensitive to moisture variations than to temperature, as known from tropical lowlands.
In the Andean region, changes in the altitudinal position of the upper forest line (UFL) are instrumental in reconstructing temperature changes. This ecotone is defined as the highest elevation contour of continuous forest and marks the boundary between the forest and the high Andean páramo biome (Moscol-Olivera and Hooghiemstra, 2010; Groot et al., 2013). The Andean sites in Venezuela and Colombia show indications of colder climates by decreased arboreal pollen at higher elevations. In the Venezuelan Andes, the only available pollen record is Piedras Blancas. There is no indication of human activity; hence, changes should be attributed mostly to climatic shifts, notably temperature and moisture. The expansion of superpáramo vegetation suggests a response to the warm and moist Medieval Climate Anomaly (MCA; ∼ 1.15-0.65 ka), while a period of scarce vegetation might be related to the LIA (∼ 0.6-0.1 ka) (Ledru et al., 2013a). The absence of tree pollen in several samples indicates significantly depressed UFL in comparison to today. Along the transitional zone between savanna and tropical rainforest in the east (E) Colombian savannas, three pollen records fulfil at least three criteria. Since 2 ka, a gradual increase in savanna vegetation is observed, suggesting a period of progressively drier conditions, e.g. at Loma Linda and Las Margaritas). However, the expanding Mauritia palm forest observed in several records is considered to reflect increased local water availability and precipitation (Fig. 8b), and/or human impact (Behling andHooghiemstra, 1998, 1999;. Along the N Andean Pacific slopes, Jotaordó, El Caimito, Guandal, and Piusbi document vegetation changes related to the precipitation regime in the C and S Chocó biogeographic region. Settings differ, as the first is located in a broad river valley with a meandering drainage system, while El Caimito and Guandal are located in the coastal plain receiving signals from shifting mangrove forests. These shifts were considered not to be climate related but explained by tectonic events in the region and/or dynamic shifts of the river deposition patterns. Frequent erosion events, various seismic shifts and disturbance indicators of mixed origin during the last 2 kyr hinder consistent conclusions for the region. Changes in vegetation composition around 0.65 ka were assigned in El Caimito to reduced flooding and possible human intervention, while similar changes at Jotaordó were ascribed to endogenous dynamics. Only the multi-proxy approach of El Caimito suggests a possible relationship between periods of higher riverine dynamics and the frequency of long-term ENSO variabilwww.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 ity. Within this region, Cecropia is used as natural disturbance indicator due to fluvial-marine dynamics, while in the other Colombian regions this fast-growing species is considered characteristic of human interference; both settings have disturbance as a common factor. In the Colombian Andes there are no undisturbed pollen records during the last 2 kyr suitable for climate reconstructions. Before the human disturbances, the La Cocha-1 record in the far south of Colombia ( Fig. 8b) indicated generally wetter conditions similar to the N Ecuadorian pollen records of Guandera-G15 and Guandera-G8. A different kind of index to highlight vegetation-climate interaction was used in the E Ecuadorian Andes at Papallacta PA1-08. Established to characterize the SASM and ENSO, the index interprets cloud transported arboreal pollen grains and Poaceae as a proxy for upslope cloud convection. Supported by a high resolution (∼ 15 yr), a high frequency of dry and humid episodes is detected during the last 1.1 kyr. In this alternation of con-vective activity, the MCA, LIA, and current warm period are considered detectable.
In S Ecuador four pollen records suitable for PAGES-2k purposes are found within a relatively small subregion. Tres Lagunas suggests a cold phase, possibly the LIA, as one of several warm and cold phases detected during the last 2 kyr (Fig. 8b). At Laguna Zurita, the decrease in Isoetes was considered an indication of increased precipitation after ∼ 1.2 ka, observed similarly in other fossil pollen records in the C Peruvian Andes. On the other hand, chemical analyses from the same core suggested drier conditions during the last millennium, confirmed by a different set of palaeoclimatic records. Unknown human interference in the last millennium could be related to these divergent patterns, as the nearby ECSF Refugio and Laguna Daniel Álvarez detected Zea mays around 0.8 and 1.4 ka, respectively.

Climate-vegetation interaction in the Central Andes
The C Andes includes the high-elevation plateau of the Altiplano, above 3000 m a.s.l, in S Peru, Bolivia, and N Chile ( Fig. 6 Box C). The Altiplano is an area of internal drainage within the Andes, which contains multiple peaks over 5000 m a.s.l.. The vegetation of the Altiplano is characterized by different grassland types, collectively known as "puna" (Kuentz et al., 2007). Within the grassland matrix are patches of woodland dominated by trees of the genus Polylepis (Fjeldså and Kessler, 1996). To the east and west of the Altiplano are the steep flanks of the Andes. In total, 57 pollen records covering the last 2 kyr were identified from the Altiplano in Peru and Bolivia. Only four of the Altiplano records met all PAGES-2k criteria: (i) Cerro Llamoca, (ii) Marcacocha, (iii) Chicha-Soras, and (iv) Pacucha ( Fig. 9a; Table 3). From the surrounding regions two additional records are also considered here because of their importance and fit to the PAGES-2k criteria: (i) Consuelo on the E Andean flank, at mid-elevation (1370 m a.s.l.), and (ii) Urpi Kocha on the Pacific coast at sea level (within the archaeological site of Pachacmac). Of the seven sites considered in this review only two records (Cerro Llamoca and Consuelo) show no human interference, while the others indicate human impact during different periods of time throughout the last 2 kyr.
Discerning a climate signal from the pollen records of the last 2 kyr in the C Andes is a challenge due to the long legacy of human occupation and landscape modification (Bennett, 1946;Dillehay et al., 2005;Silverman, 2008). However, some idea of vegetation-climate relationships can be gained from modern pollen studies within the puna; for example, Kuentz et al. (2007) use the ratio of Poaceae : Asteraceae (Nevado Coropuna) and Schittek et al. (2015) focus on the abundance of Poaceae (Cerro Llamoca) as an indicator of moisture availability. In the other records, where there is no direct relationship between vegetation and climate discernible, some authors look at the relationship between the pollen records and other indicators, such as independent evidence of farming activity (e.g. oribatid mites) or an association with archaeological evidence of abandonment or occupation (Chepstow-Lusty, 2011), to disentangle climate-and human-induced vegetation change. The two records considered here that are purported to have no local human impact (Cerro Llamoca and Consuelo) provide the best opportunity of obtaining a clear insight into past climatic change in the C Andes during the last 2 kyr. The record from Cerro Llamoca indicates a succession of dry and moist episodes (Fig. 9b). After 0.5 ka, sediments are composed of redeposited and eroded material, and consequently the interpretation of the latter half of the record is difficult. In contrast, little compositional change is evident in the Consuelo record, with the most significant variance during the last 2 kyr being a rise in Cecropia sp. pollen after 1 ka. Cecropia pollen is typically interpreted as an indicator of disturbance (Bush and Rivera, 2001), and therefore, in the absence of humans signal, the rise in Cecropia could be interpreted as an elevated level of natural disturbance. The switch to very dry conditions at Cerro Llamoca in the western Andean cordillera and the rise in Cecropia at Consuelo on the E Andean flank are broadly coincident (∼ 0.85 ka); however, it is not possible to say if this pattern results from a common climatic mechanism.
Archaeological evidence from Chicha-Soras does not show any evidence of human occupation of the valley between ∼ 1.9 and ∼ 1.4 ka. Between 1.4 and 1 ka and between 1 and 0.65 ka, a high abundance of Chenopodiaceae or Amaranthaceae (Chen-Am) could be interpreted as either indicating arid conditions or expansion of quinoa crops (Ledru et al., 2013b). However, a drop in charcoal fragments (fire activity) coupled with the absence of archaeological evidence (∼ 1.9-1.4 ka) suggests that people abandoned the val- ley during 1.5-0.5 ka and, consequently, that the aridity signal from the pollen could be interpreted as a climatic one.
Some climate information has been inferred from the four remaining sites (Marcacocha, Pacucha, Nevado Coropuna, and Urpi Kocha) despite the strong human influence on the vegetation. At Nevado Coropuna humid conditions persisted until a short dry episode occurred at 0.97-0.82 ka (Fig. 9b). During the last 2 kyr at Marcacocha, successive peaks in Cyperaceae pollen have been interpreted as indicative of three periods of elevated aridity, while more abundant elevated Plantago at ∼ 1.9 ka is suggested to indicate cooler conditions, and Alnus at ∼ 1-0.5 ka could indicate warmer and drier conditions, although discerning the climate signal related to Alnus is difficult due to its utilization in agroforestry practices (Chepstow-Lusty and Jonsson, 2000). At Pacucha and Urpi Kocha, significant changes to the pollen assemblage in the last 2 kyr are attributed to human activity rather than climate. Although the pollen records are likely to be somewhat obscured by the agricultural activities and irrigation of the crops, all high-elevation records with a moisture balance signal suggest generally drier conditions occurred in the C Andes between 1.2 and 0.7 ka when compared with the rest of the last 2 kyr.
Generally, the pollen records from the Altiplano tend to show a greater sensitivity to precipitation rather than to temperature. The greater sensitivity to precipitation is because moisture availability is the limiting factor in most areas for both vegetation and human occupation. However, human occupation indicates changes in temperature: (i) at Marcacocha, when the sudden stop in agricultural activities is attributed to colder temperatures and (ii) at Nevado Coropuna, when the increase in human occupation (expansion of Inca culture) at higher elevation shows that there was no glacier but there were warmer temperatures.

Climate-vegetation interaction in the lowland Amazon basin
For the purpose of this review, the lowland Amazon basin constitutes those regions of the Amazon drainage below www.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 500 m a.s.l. and extends to the lowland Guayanas ( Fig. 6 Box D). This encompasses the evergreen rainforest, which covers most of Amazonia, as well as the southern transitional or seasonally dry tropical forests located in NE Bolivia and S Rondônia, N Mato Grosso, and N Para State, Brazil. It also includes the Llanos de Moxos savannas of NE Bolivia, and the ecotonal rainforest-savanna areas of N Roraima State, Brazil, and extends to the coastal swamps or grasslands of N Brazil and French Guiana. In total, 42 published pollen records that cover the last 2 kyr were identified from the lowland Amazon basin. By applying the dating constraints of the PAGES-2k criteria, the majority of pollen records from the Amazon basin are discounted from any analysis of climate-vegetation interaction for the past 2 kyr. Only 5 records complied with all four of the criteria and 11 records met with three criteria (Fig. 10a; Table 3). One of these records, Lake La Gaiba, is situated just outside the Amazon basin, in the Pantanal region of central Brazil and SE Bolivia. However, the record and its hydrological catchment reflect Holocene precipitation in the S Amazon basin (Whitney et al., 2011), and it therefore was included as part of this review.
Lake Quistococha in the NE Peruvian Amazon is surrounded by a Mauritia flexuosa-dominated palm swamp.
Vegetation has undergone several significant species compositional changes over the past 2 kyr. The broad pattern of vegetation change was from Cecropia-dominated riverine forest at ∼ 2.2 ka to abundant Cyperaceae and floating grasses or ferns and the commencement of peat formation ∼ 2.1 ka to seasonally inundated riverine forest with abundant Moraceae and Myrtaceae from ∼ 1.9 ka and, finally, the development of closed-canopy, Mauritia-dominated swamp from ∼ 1 ka until the present. Superimposed on this broad pattern of change were rapid, centennial-scale shifts in forest composition and degree of openness. However, these rapid shifts were attributed by the authors to hydrological dynamics rather than to climate change or human impact.
Lake Werth belongs to a collection of sites (including Gentry, Vargas, and Parker) in the "Madre de Díos" region of the SE Peruvian Amazon. The lake formed at ∼ 3.4 ka and records continuous evergreen rainforest throughout, with little evidence of burning. The records from the surrounding three lakes concur, suggesting that, regionally, rainforest (and climate) has been stable over the last 2 kyr.
Laguna Granja is located on the edge of the Pre-Cambrian Shield in NE Bolivia. The earliest date for the record is 6 ka and indicates that savanna characterized the landscape from 6 ka. This is in agreement with a regional-scale reconstruc-Clim. Past, 12, 483-523, 2016 www.clim-past.net/12/483/2016/ tion from the much larger Lake Orícore (not shown; Carson et al., 2014), which is located < 20 km away from Laguna Granja and shows climate-driven expansion of evergreen rainforest in this region between ∼ 2 and 1.7 ka. However, forest expansion does not occur on the Granja site until 0.5 ka. The distribution of forest vs. savanna around Laguna Granja was shown to be heavily influenced by human land use between 2.5 and 0.5 ka (Carson et al., 2014(Carson et al., , 2015; therefore, it is not suitable for analysis of naturally driven vegetation dynamics. The Fazenda Cigana record is in the savanna-galleryforest mosaic landscape in the N Brazilian Amazon. The core was taken as one of a pair, along with the Terra Indígena Aningal record, which was cored from the same Mauritia swamp. The pollen records are dominated by Mauri-tia throughout, attributed to continuously wet climate in this region in the late Holocene. There are however centennialscale periods of gallery forest reduction and grassland expansion, accompanied by increased charcoal concentrations. Da Silva Meneses et al. (2013) inferred that these periods of high burning were anthropogenic in origin and compared them to modern-day prescribed burning practices used by indigenous people in the northern Amazon to maintain an open savanna landscape. Despite the potential human interference, these records demonstrate natural stability of the forest-savanna ecotone over the last 1.5 kyr in this particular part of the N Amazon.
The French Guiana K-VIII record was taken within a landscape of pre-Columbian mounded agricultural fields, with the principal aim of investigating ancient human land use www.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 associated with these earthworks on a local scale. From the earliest part of this record, the fossil pollen spectra indicate seasonally inundated savanna, dominated by Cyperaceae and Marantaceae until 0.8 ka, when human inference is detected.
In the post-European period after ∼ 0.5 ka, charcoal abundance increases, probably reflecting more intensive use of fire by colonial populations.

Climate-vegetation interaction in southern and southeastern Brazil
The landscape in S and SE Brazil is diverse, from lowlands to high mountains, from subtropical regions with frost to tropical regions. Due to this heterogeneity distinct vegetation types occur throughout the region. The vegetation in S-SE Brazil includes forest ecosystems such as the tropical Atlantic rainforest, Araucaria forest, semi-deciduous forest, "Cerrado" (savanna woodland), and different grassland ecosystems such as "Campos" and "Campos de Altitude" (high-elevation grassland) ( Fig. 6 Box E). There is a gradient from no or short dry seasons in the coastal lowland to 6-month dry seasons in the hinterland (northernmost part of the highland in SE Brazil), marking the vegetational gradient from moist Atlantic rainforest to semi-deciduous forest and to Cerrado. Additional background information is provided in the Supplement.
There are approximately 50 pollen records known from S-SE Brazil, but many sites have not been published in peerreviewed journals and were therefore not considered. Unfortunately, the two records that agree with all criteria show human interference (Table 3). Therefore, a general overview of climate-vegetation interaction from the region is presented, considering seven records that fulfil some of the criteria (Table 3, Fig. 11a).
In S Brazil pollen records indicate vegetational changes that reflect a change from a relatively dry climate during Clim. Past, 12, 483-523, 2016 www.clim-past.net/12/483/2016/ Not all records are suitable to derive both a moisture and a temperature signal. Climate and human presence are shown overlapping when the pollen record is not conclusive regarding the derived signal. Bars are shaded grey when the climate signal is obscured by human interference. Records fulfilling one or two criteria indicated by star.
the early and mid-Holocene to wetter conditions after about 4.3 ka, and in particular after 1.1 ka (Fig. 11b). Increasing moisture is clearly indicated on the S Brazilian highlands by the expansion of Araucaria forests in the form of gallery forests along rivers and a pronounced expansion of Araucaria forest into the Campos after about 1.1 ka (e.g. Cambará do Sul and Rincão das Cabritas). The expansion of gallery forests at similar time periods (5.2 and 1.6 ka) is also recorded in the southernmost lowland in S Brazil by the São Francisco de Assis record. Study sites that reflect changes in the Atlantic rainforest area indicate an expansion during the Holocene where overall wetter conditions prevailed compared to highland and southernmost lowland areas, e.g. Ciama 2 (Fig. 11b).
In contrast to other sites and regions, a relatively humid and warm phase during the LIA is inferred from the highresolution Cambará do Sul record as an expansion of Wein-mannia in the Araucaria forest is observed. In SE Brazil the Lago do Pires and Lagoa Nova record indicate that a dense and closed semi-deciduous forest existed in the region only during the late Holocene period under the current climatic conditions with a ∼ 3-month dry season. In the mountains of SE Brazil (e.g. Serra dos Órgãos record), a reduction in Campos de Altitude occurred at 0.9 ka, indicating a change to wetter conditions that is broadly coeval with a similar trend in the Lago do Pires record (Fig. 11b).

Climate-vegetation interaction in the pampean plain
The pampean plain extends east of the Andes, between 30 and 40 • S (Fig. 6 Box F), and is characterized by aeolian landforms marking the climatic gradient of the landscape. The natural vegetation of the pampas is a treeless grassland, dominated by Poaceae in terms of both species number and www.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 abundance. Asteraceae shrubs (e.g. Baccharis and Eupatorium) are present locally in the S pampas, Cyperaceae characterize aquatic and wet-ground communities of temporarily flooded depressions and shallow lakes mainly from the E pampas, and Chenopodiaceae characterize edaphic communities such as salt marshes and alkaline flat areas (Tonello and Prieto, 2008). Additional background information is provided in the Supplement. In total, nine pollen records were assessed for the last 2 kyr (Fig. 12a). All four dating criteria were met in one record only (Lake Lonkoy), and three criteria were matched at Sauce Grande (Table 3). The pollen record of the Hinojales-San Leoncio site does not fulfil the four dating criteria; however, the record shows important hydrological signals for the last 2 kyr and is therefore briefly discussed. Aquatic ecosystems are considered sensitive to climatic and/or hydrological variations and exhibit frequent fluctuations in their water level and extension, leaving flooded and/or exposed plains. Pollen together with non-pollen palynomorphs and plant macrofossil analysis present similar trends in the SE pampas that support the idea that climate is a regional trigger of change (Stutz et al., 2015). From 2 to 0.7-0.4 ka, an unstable regional environment with drier climatic conditions than at present is inferred for the region (Fig. 12b), based on halophyte plant communities (Chenopo-diaceae) surrounding the lakes, whereas Chara and other aquatic plants (e.g. Myriophyllum, Potamogeton) characterized the water bodies. Towards ∼ 0.5 ka, vegetation changed to Cyperaceae dominance and aquatic plant composition similar to modern associations. Thus, turbid conditions with a higher water level and/or an extension of surface lakes under more stable environmental conditions are inferred. These support humid conditions similar to the present with a noticeable increase in precipitation after 0.4 ka, indicated by high Cyperaceae abundances. However, an integrative multiproxy approach allows inferring stable conditions and higher salinity values between 1.9 and 0.9 ka and periods of water level fluctuations after 0.9 ka, with high water levels between 0.66 and 0.27 ka. These changes may have been caused by fluctuations in precipitation (Fontana, 2005).

Climate-vegetation interaction in the Southern Andes and Patagonia
The study area comprises the S Andes, which includes subtropical and temperate regions (22-56 • S) on both sides of the Andes, including Patagonia (40-56 • S), which extends eastwards from the Andes to the Atlantic Ocean ( Fig. 6 Box G). The region has different geomorphological settings associated with glacial, volcanic, and tectonic activities. Vegetation associations reflect the west-east precipitation gradient  from the wet Nothofagus forest to the dry grass and shrub steppe towards the Atlantic coast. The south to north gradient along the Andes ranges from the Nothofagus temperate forest in the austral region to the Nothofagus-Austrocedrus forest, sclerophyllous forest, and xerophytic woodland in the central region. At the northernmost end of the latitudinal gradient, the vegetation is adapted to extremely arid conditions characterized by small and dwarf shrubs and scarce cover (see the Supplement for additional descriptions). In this region, there are 48 pollen records that cover the last 2 kyr with at least one chronological control point during this period. Of these, the 19 records that fulfil PAGES-2k criteria are mostly concentrated in the temperate forests, while only few originate from xerophytic shrub steppe (1 record), subtropical-sclerophyllous forest (2 records), and grass steppe (4 records) (Table 3; Fig. 13a).
There are three sites in the far south of Patagonia: the Tierra del Fuego's Onamonte mire (54 • S) located at the Nothofagus forest-grass-steppe ecotone shows a gradual Nothofagus forest development between 1.5 and 0.5 ka followed by a major forest development up to the present, reflecting increased precipitation (Fig. 13b). Puerto Harberton (55 • S) in the mixed Nothofagus betuloides-N. pumilio forest shows Nothofagus dominance during the last 2 kyr, whereas the Ericaceae increase during the last 1 kyr suggests a local decrease in the water table. Similarly, at Valle de Andorra (54 • S) in a Nothofagus pumilio forest, Empetrum-Ericaceae fluctuations reflect changing water tables.
In S Patagonia (52-51 • S) along the E Andes, there are several sites at or near the forest-steppe ecotone. Of these ecotonal sites, Río Rubens (52 • S) shows a closed Nothofagus forest until 0.4 ka when European impact starts (Fig. 13b). Similarly, Lago Cipreses (51 • S) and Lago Guanaco (51 • S) show dominance of Nothofagus forest but with an increase in non-arboreal pollen (and a decrease in Nothofagus) associated with a reduction in precipitation induced by the southern westerly wind belt (SWWB) and the SAM phases. Furthermore, changes associated with dry or warm climate conditions appear to synchronize with northern hemispheric events, such as the Industrial Revolution, MCA, Roman Warm Period and Late Bronze Age Warm Period , that alternate with wet and cool phases. Cerro Frías (50 • S) shows open forest from 2.0 to 0.9 ka, followed by a period in which grass steppe that is punctuated by an increase in Nothofagus at 0.016 ka is prevalent. Estimates of annual precipitation suggest similar or higher values than in the modern period between 2 and 1 ka and lower values between 0.9 and 0.015 ka, followed by similar-to-modern precipitation in the last 0.015 ka. Currently located in mixed deciduous Nothofagus forest, the Peninsula Avellaneda Bajo (50 • S) records an open forest from 2 ka, of which large expanses were replaced by grass steppe between 0.4 and 0.2 ka, associated with a decline in precipitation.
In C Patagonia (47-44 • S) pollen records are located in the east of Andes (Fig. 13a). At Parque Nacional Perito Moreno (47 • S), a shrub-steppe expansion (Asteraceae and Embothrium dominance) suggests lower precipitation values between 1.2 and 0.25 ka compared to previous values, after which an increase in grass steppe occurs due to higher moisture availability (Fig. 13b). However, the Mallín Pollux (45 • S) record indicates an open canopy prior to 1.5 ka followed by a Nothofagus forest expansion associated with a precipitation increase. Mallín El Embudo (44 • S), in a Nothofagus deciduous forest, shows unvarying forest composition during the last 2 kyr. Located in the same valley, the Lago Shaman (44 • S) record (Nothofagus forest-steppe ecotone) shows a more diverse pattern throughout the last 2 kyr, with a forest retraction at ∼ 1.7 ka followed by an expansion around 1.5-1.3 ka and a major forest development around 0.5 ka. The forest decrease during the last 0.2 kyr is associated with human intervention.
In N Patagonia (44-38 • S), Lago Mosquito (42 • S) is the only record in the E Andes, and it is located at the transition between Austrocedrus woodland and shrubland steppe. The record shows an open Nothofagus-Austrocedrus forest with steppe and grassland elements between 2 and 1.4 ka, changing to higher Nothofagus forest dominance, which is attributed to wetter conditions (Fig. 13b). From 0.225 ka to the present, Nothofagus shows a sharp decrease and Cupressaceae increases, together with a rise in introduced species, e.g. Rumex and Pinus. At the same latitude, Lago Lepué (42 • S), located on the Isla Grande de Chiloé and surrounded by evergreen rain forest, shows dominance of Nothofagus during the last 6 kyr with an important reversal between 2 and 0.8 ka. This suggests lower precipitation than before and after 0.8 ka, shown by an increase in Weinmannia and Isoetes. The Lago Pichilafquen (41 • S) record, under the domain of the SWWB and influenced by the subtropical Pacific anticyclone in summer, shows a series of warm and dry and cold and wet phases for the last 2 kyr (Fig. 13b). These phases are inferred by the varying abundances of Nothofagus and Eucryphia or Caldcluvia and Poaceae. The last centuries are characterized by human intervention. At the temperate-subtropical transition, the Lago San Pedro (38 • S) record shows drywarm phases, which were associated with the MCA period. Cold and wet conditions, inferred by the relation between Nothofagus and Poaceae and changes in the depositional time, prevailed during the LIA, possibly related to El Niño and La Niña influencing these wet and dry phases, respectively (Fig. 5).
To the north (west of the Andes), the Laguna de Aculeo record (34 • S) shows dominance of Poaceae, suggesting relatively steady conditions during the last 2 kyr, with the exception of the last 0.1 kyr, when a trend towards warmer conditions or human disturbance is reflected by an increase in Chenopodiaceae (Fig. 13b). Interestingly, the sedimentary record shows a series of turbidite layers associated with major ENSO frequency between 1.8 and 1.3 and between 0.7 and 0.3 ka (Jenny et al., 2002). The Palo Colorado (32 • S) record shows dominance of Myrtaceae associated with wet conditions during the last 2 kyr alternating with several dry pulses. A major dry peak at 0.4 ka may be related to climate and/or human activity. Similarly in the E Andes, Abra del Infiernillo (26 • S) shows an increase in moisture between 2 and 0.75 ka inferred from Juncaceae, Poaceae, Cyperaceae pollen, and fern spores and a change to dry climatic conditions similar to modern conditions from 0.75 ka on.
Laguna Potrok Aike and Laguna Azul (both 52 • S) show a dominance of Poaceae from 2 ka onwards, with pollen of Nothofagus that was transported over a long distance. At Laguna Potrok Aike, reconstructed annual precipitation based on a transfer function indicates rising values during the last 2 kyr (Fig. 13b). Cabo Vírgenes (52 • S), located in the SE Patagonian grass steppe, shows a shrubland community between 1.2 and 0.7 ka, associated with drier conditions than at present. An increase in moisture after ∼ 0.7 ka is indicated by Poaceae and Juncaginaceae pollen. Cabo Vírgenes CV22 shows a similar trend, with dry grass-shrub steppe between 1.05 and 0.6 ka, followed by a grass-dominated steppe suggesting higher moisture availability.

Indicators of human land use in 2 kyr pollen records
In any past environmental change reconstruction concerning the last 2 kyr, human land use must be considered as a potentially important agent of environmental change. However, where there is no direct evidence of human land use, such as cultigen pollen, distinguishing natural from anthropogenically induced burning and vegetation change can be difficult. In some cases anthropogenic deforestation and decreased moisture may result in similar signals in the pollen record, and therefore complementary proxies of past environmental change can be used to support interpretations, such as chironomids (Matthews-Bird et al., 2015;Williams et al., 2012) and geochemical records from speleothems.
There are six key aspects of fossil records (pollen and charcoal) that can be seen as indicators of past human activity. These are (i) a decrease in forest taxa (degraded forest) and/or forest composition, (ii) the presence of crops, e.g. Zea mays, Manihot esculenta, and Phaseolus and Ipomoea, (iii) the presence of crop-related herbs, e.g. Rumex, (iv) an increase in grasses or herbs, e.g. Poaceae, Cyperaceae, and Asteraceae subf. Cichorioideae, (v) an increase in dis-turbance indicators, e.g. Chen-Am, Cecropia, Vismia, ferns, and palms (including Mauritia and Euterpe or Geonoma), and (vi) an elevated amount of charcoal due to anthropogenic fire (Fig. 14). These indicators of human activity can be split into two classes: those that directly indicate human presence and those from which it is indirectly inferred. Manihot esculenta and other crops, such as Zea mays, are considered direct indicators of human influence and provide clear evidence of land use. Indirect indicators, such as a change in forest composition (e.g. due to deforestation) or the appearance of species known as possible disturbance indicators (e.g. Cecropia or Mauritia), need further evidence from other proxies to support any inference of past human activity. Only by looking at changes in pollen spectra in the context of other evidence (e.g. from charcoal, limnological, sedimentological, or archaeological data sets) can the most probable driver of any change be suggested.
In this paper, ambiguous records with few proxies were not immediately discarded but were considered within the context of the other records from their wider region. Based on this, an assessment could be made as to whether an anthropogenic signal may have obscured the natural vegetation change trajectory. The moisture balance and temperature summaries for each region (Figs. 7-13) clearly indicates when human interference obscures the climate assessment and when both climate and/or humans may have influenced the pollen record.
To date, major human impact in the Venezuelan Guayana uplands has been suggested for the last 2 kyr and inferred from the charcoal record, without any evidence from crops. Compared to the highlands (1500-3000 m a.s.l.), the situation in the uplands (500-1500 m a.s.l.) differs substantially as fire is the factor most responsible for vegetation change during the last 2 kyr. The Urué record shows the consequence of repeated burning for the vegetation, preventing the recovery of pre-existing forests and allowing the appearance of "helechal" (fern-dominated vegetation; Huber and Riina, 1997) and the establishment of the savanna. The occurrence of frequent fires during the last 2 kyr is a common feature of most of the upland records analysed so far, regardless of the plant association present at each location. Synchronous with this increase in fire regime, the records that nowadays are characterized by Mauritia palm swamps, showed parallel a sudden appearance and establishment of Mauritia. Human activities have been proposed as the likely cause of this high abundance of fires and thereby of the consequences that appeared in the landscape. In this sense, the repeated use of fires would have promoted the reduction in forests and expansion of the savanna, favouring the establishment of Mauritia swamps after clearing. Two records are particularly relevant regarding the human influence on the Venezuelan Guayana uplands. The Lake Chonita sequence ( Table 3) was one of the earliest Mauritia establishments, coeval with a significant increase in the fire regime during a likely local wet period around 2 ka. In the southernmost boundary of the Venezuelan  Table 3. Guayana, El Paují (Table 3) was interpreted as potentially reflecting human activities since the mid-Holocene. This location is characterized today by treeless savanna surrounded by dense rainforests that established themselves at ∼ 1.4 ka as shown by the high abundance in the record of algal remains (local wet conditions) and charcoal particles (fire regime). The establishment of the present-day landscape was interpreted as mainly anthropogenically driven, with the arrival of the current inhabitants. The occurrence of a previous secondary dry forest was interpreted as the result of climatehuman interplay, linking land abandonment and likely drier climate as the main factor responsible for the vegetation shift. From the Colombian savannas, human occupation is attested since the mid-Holocene . At site Loma Linda a plausible signal of human interference in the last 2 kyr is the increased savanna, although precipitation increase during the same period (Behling and Hooghiemstra, www.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016Past, 12, 483-523, 1998Past, 12, 483-523, , 1999Marchant et al., 2001Marchant et al., , 2002 could be interfering with that signal. The human history in the N Andean region goes back to the Lateglacial period ( Van der Hammen and Correal Urrego, 1978). The high plains of the Colombian Cordilleras have provided suitable conditions for human settlements since the start of the Holocene. Increasing human occupation became evident in pollen records after ∼ 3 ka, such as at Fúquene-2 and Pantano de Genagra. In several Andean pollen records, Rumex acetosella marked the arrival of Europeans from 0.4 ka onwards (Bellwood, 2004;Bakker et al., 2008). Before these dates, indigenous populations were scarce and their practices negligible in terms of impact, especially at high-elevations sites such as Piedras Blancas in Venezuela.
In the tropical lowlands along the Pacific coast, increases in the presence of palms (mainly Euterpe or Geonoma) are commonly interpreted as a result of more intensive forest use, e.g. Lake Piusbi. Pollen grains from crops like Zea mays, Phaseolus, and Ipomoea are found in Piagua (Vélez et al., 2001). Human disturbance to the forest is considered indicated by high percentages of abundance of Cecropia, ferns, and palms. Decreases in human impact during the last 2 kyr have been described by sites like Pitalito basin (Bakker, 1990;Wille et al, 2001), Timbio (Wille et al., 2000), Pantano de Genagra (Behling et al., 1998), Quilichao, and La Teta , as grassy vegetation (Poaceae) and Zea mays disappeared and forest started to recover. This vegetation change could be related to the first arrival of the Spanish "conquistadores" (González-Carranza et al., 2012) or by a set of different causes (Wille and Hooghiemstra, 2000).
In the C Andes a high level of human activity, spatially variable in intensity, has been shaping the landscape for the last 2 kyr. Chen-Am and Zea mays generally appear in all the records in the Central Andes after 4 ka, e.g. Pacucha, Marcacocha, Chicha-Soras, and Urpi Kotcha. After 2 ka, Alnus and agroforestry practices are observed (Marcacocha, Pacucha). When irrigation started to be developed at sites without a nearby lake, as for instance ∼ 1 ka at Nevado Coropuna, Ambrosia may have been used as a terrace consolidator. Evidence of afforestation at two sites with high human influence (Marcacocha and Pacucha) is observed. Indeed, Alnus acuminata is a tree planted by the Inca to stabilize landscapes (Chepstow-Lusty, 2011). At lower elevations, in the Andean forest, the last 2 kyr pollen data indicate little change in woodland cover, which remains high on the E Andean flank (Consuelo) and low in the west (Urpi Kocha).
Of the 42 pollen records identified from the lowland Amazon basin, 15 show evidence of pre-and post-European land use within the last millennia. Human land use is inferred from these records from cultigen pollen grains, charcoal, and forest clearance (Table 3). In some cases there is also archaeological and archaeobotanical evidence for human land use. At many of the sites occupied by native Amazonians, evidence of decreased land use shows as a decline in burning by or before 0.5 ka, probably in relation to first European contact. However, some sites, such as French Guiana VII and Laguna Granja, show evidence of continued post-European land use.
In SE-S Brazil, the modern vegetation is strongly affected by the logging of forests and different agricultural land-use practices. During the last few decades, large-scale afforestation of grassland by Pinus is seen on the highlands. Similar to SE-S Brazil, the pampas region has a relatively short farming history, since most of the area remained native grassland until the end of the 19th and the beginning of the 20th century (Viglizzo and Frank, 2006). Today, only around 30 % of the region is covered by natural or seminatural grassland. Pampas vegetation does not show evidence of human impact prior to European settlement at 0.4 ka. Europeans introduced several tree species (e.g. Eucalyptus, Pinus), as well as cattle (Bow taurus and Equus) and crops (Triticum aestivum, Helianthus annuus), but the intensive agricultural activities only began 0.05 ka (Ghersa and León, 2001). The palaeoenvironmental history of shallow lakes shows a change to more productive systems (higher mass of phytoplankton and organic matter content) during the last 0.1-0.08 kyr, probably due to agricultural activities. On the other hand, pollen records show an increase in pollen types associated with overgrazing (Plantago and/or Asteraceae Asteroideae) and exotic trees during the last 0.1 kyr.
In the S Andes and Patagonia, anthropogenic activities during the last century have caused a range of disturbances (e.g. fire, forest clearance, grazing, agriculture) and major vegetation changes in forest and steppe areas have occurred. There is no conclusive evidence of native human activities in the pollen records and native fire disturbance has been long discussed. Charcoal records from the E Andes flank have not revealed fire activity associated with native populations. A probable explanation for this lack of evidence is a low density of populations associated with a sporadic impact on forest (Iglesias and Whitlock, 2014). In general, human activities indicators are forest decrease, the presence of exotic pollen types (e.g. Rumex), and an increase of some pollen types (e.g. Asteraceae subf. Cichorioideae, Chenopodiaceae), associated with a European presence in the region. The time of colonization varied among S Andes and Patagonian sites, but ∼ 0.1 ka can be considered as the start of European activities in Patagonia. Differences in timing of the first appearance of human indicators in pollen records could reflect European settlement dynamics, with an earlier presence at more northerly sites and later in more isolated areas (in the south of continent). The first human indicator is recorded at Río Rubens (52 • S) with the appearance of the European weed pollen Rumex acetosella in the early European era (∼ 0.3 ka).

General observations for 2 ka pollen compilations
This review reveals that those records with better dating resolution in the late Holocene are often from cores that span a shorter time period, while longer temporal records have less well-resolved Holocene chronologies. This likely reflects (i) the need to spread limited numbers of radiocarbon dates in order to provide robust age models for these deeper time records, (ii) the greater interest of previous researchers in potential large-scale palaeovegetation changes, driven by glacial-interglacial climate cycles, and other significant periods of climatic change, such as the early-to-mid Holocene drought, and (iii) the low sedimentation rate during the last millennia in certain regions, e.g. lowland Amazonia. Furthermore, strong anthropogenic interference during the last 2 kyr complicates the interpretation of many records from a palaeoclimate perspective, but with expert knowledge, climate signals can be filtered. Additional difficulties arise from the "one topic focus" of many studies, and authors do not often present the full range of data in their publications that are required for a comprehensive reconstruction of vegetation, climate, and human impacts over the last 2 kyr.

Venezuelan Guayana highlands and uplands
For the Venezuelan Guayana region, we here discuss the highland and upland areas separately due to the significant differences in physiographical, climatic, and ecological features, as well as in the intensity of human pressure on their respective ecosystems.
The highlands are virtually pristine and, according to the palaeoecological records, they have remained in this state at least since the early Holocene. Therefore, climate has been the main driver of change. Palaeoecological records for the last 2 kyr are scarce and generally of low resolution, but a common trait is the ecological stability as expressed in the vegetation constancy. The following hypotheses have been suggested to explain these observations: (i) environmental changes were insufficient to affect the highland vegetation, (ii) the high precipitation and relative humidity of the Chimantá summits (Briceño et al., 1990) have buffered climatic changes, and (iii) the study sites are unsuitable for recording significant vegetation changes because there are no vegetation ecotones nearby (Rull, 2015). Further work is needed to test these hypotheses. So far, palaeoecological fieldwork atop the tepuis has been carried out in an exploratory, nonsystematic manner due to the remoteness of the tepuis and the logistic and administrative constraints (Rull et al., 2008). In the LOTRED-SA framework, the issue of vegetation constancy emerges as a priority and should be addressed properly by finding suitable coring sites to be analysed with high-resolution multiproxy tools. The use of physical-chemical proxies independent of pollen and spores is essential to record climatic shifts. Lake sediments would be excellent for this purpose but, unfortunately, lakes are absent on tepui summits. The only permanent lake known so far is Lake Gladys atop the Roraima tepui, the age and origin of which remain unknown (Safont et al., 2014). At present, the analysis of the Apakará PATAM9-A07 core, which meets the PAGES-2k criteria, is in progress. The preliminary study of this core showed the main Holocene vegetation trends at millennial resolution , and the current analysis is being performed at multidecadal resolution. A new core obtained at the Uei summit (PATAM8-A07; not included in the Chimantá massif) and containing a decadal record for the last 2 kyr is also currently being analysed (Safont, et al., 2016).
In the GS uplands, the situation is very different and the main driver of ecological change is fire caused by humans. This does not mean that climatic shifts have been absent or that they have not affected the vegetation, but the action of anthropogenic fires is more apparent than the action of climate and obscures it . So far, regional palaeoclimatic trends, based on independent data obtained from the Cariaco basin (∼ 680 km to the north; González et al., 2008), have been used as a reference for past climate change on the GS uplands (Rull et al., 2013). Unfortunately, a more local independent palaeoclimatic record for the GS uplands is still lacking, not only for the last 2 kyr but also for the entire Holocene. Another limitation is that most palaeoecological records available for the GS uplands are from their southern sector, which is the lowermost part of the peneplains and has a different climate and vegetation regime as compared to the northern sector. Some records from the northern sector are available that fit with the chronological PAGES-2k requirements (Leal et al., 2011), but only summary diagrams are provided in peer-review publications and therefore they cannot be used in this reconstruction. The decadal to multidecadal analysis of a new core obtained in Kamoirán (PATAM10-A07), in the northern GS uplands, is in progress.
It should be stressed that the last 2 kyr seem to have been critical for the ecological history of the GS uplands. Detailed knowledge of this period may be crucial to understand the origin of the present-day landscape, which is intimately linked to the temporal patterns of human impact from using fire. The date of arrival of the current indigenous culture (Pemón) at GS is still unknown. Based mainly on historical documents, it has been postulated that this culture settled in GS at ∼ 0.6 to 0.3 ka, coming from Guayana or Brazil (Thomas, 1982, Colson, 1985Huber, 1995a). However, these dates could be considered minimal ages, as recent palaeoecological studies suggest that human groups with landscape management practices similar to the Pemón people would have been present in the GS since ∼ 2 ka Montoya et al., 2011a). Before that time, the GS landscape was different from the present, including larger extents of forested areas since the Lateglacial period www.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 (22-11.7 ka) and the absence of Mauritia palm swamps until ∼ 2 ka. The same time period seems to have been a landmark event in neotropical history for similar reasons, as  showed a generalized increase in Mauritia pollen abundances in northern South America during the last 2 kyr. Given the northern position of the Venezuelan Guayana, the vegetation responses studied have normally been related to ENSO and ITCZ movements. These two main drivers are represented by the Niño 3.4, AMO, IPO, and TNA modes, which do indeed exert the main influence in the area, as shown in Figs. 2-5 (especially with respect to temperature). The lack of a significant influence of AMO on precipitation in the region is surprising. It is worthwhile to compare the climatic inferences made through fossil pollen records with the climate modes' effect on the area. Fossil pollen records have suggested available moisture (or a precipitation / evapotranspiration ratio: P / E) as the main climatic driver to take into account for vegetation responses. However, these inferences are based on very local spatial scale proxies (e.g. algal remains). P / E is complex and higher temperature drives evapotranspiration, finally leading to lower water levels under unchanged precipitation regimes (Van Boxel et al., 2013). Interpretation of the fossil record is therefore complex and sometimes ambiguous. On the other hand, both Pacific and Atlantic climate modes appear to have a potentially large effect on both temperature and precipitation in the region. Such findings suggest that the variations in P / E inferred from the fossil record could be caused by either of these two factors, or by both. Additional higherresolution multi-proxy analyses should shed a light on previously undetected modes in the region as well as disentangle the combined effect of several forcing factors. Nevertheless, upland records have been interpreted as primarily humandriven vegetation responses, so for the last 2 kyr the climatic conclusions are constrained. Highland records have been described as an example of constancy, even insensitive to temperature change during the last 2 kyr, which could confirm that the temperature variability related to climate modes in this region has been of a lesser magnitude than those required to cross the vegetation tolerance ranges. Alternatively the intrinsic characteristics of the sites studied so far have inhibited detecting any change.

Northern Andes
Study sites without human presence have been not identified with certainty within the northern Andean region, inhibiting the detection of a clear signal of climate tendencies in the last 2 kyr. Drier conditions prevailed in Colombian savanna lowlands, although the increased presence of Mauritia suggests either increased humidity and/or human influence. Along the Pacific coast, generally wetter conditions prevailed (Fig. 8b), but tectonic events might be masking clear patterns. The interpretation of some records should be undertaken with care due to the noisiness of the data. Furthermore, due to the geomorphological complexity of the landscape and its latitudinal characteristics, this region is prone to a combination of strongly overlapping climate signals within and between years (Figs. 2-5; Marchant et al., 2001).
For the Northern Andes the position of the ITCZ and the ENSO phenomenon are most important in driving changes in precipitation, as clearly illustrated in the La Cocha-1 record Figs. 4 and 5). Most significantly, the altitudinal gradient in temperature is modulated by ENSO and the TNA. This is shown by the increased temperature variability around 5 ka when the ENSO signal starts (Figs. 2 and 3). The Papallacacta record highlights the two modes which affect precipitation variability in this region, namely the eastern equatorial Pacific and the tropical Atlantic. SST anomalies in both basins have been related to climate variability in the N Andes until 0.45 ka, with interdecadal variability dominating during the last 0.5 ka. Pallcacocha in S Ecuador also shows a close match with ENSO, recording its strength during the last 15 kyr. Similarly associated with ENSO are the changes in the plant assemblages detected in the high-resolution record of El Junco on the Galápagos Islands.
Comparing vegetation-climate signals between the Colombian lowlands and E Venezuela and NE Brazil has shown opposite climate conditions. Dry conditions identified in the Colombian savannas (suggesting an ENSO-La Niña) agree with similar conditions in the Bolivian pollen records. During an El Niño setting, when Bolivian savannas indicated wet conditions, the signal from Lake Valencia in Venezuela reflected dry conditions (Martin et al., 1997;Wille et al., 2003). Lowland sites generally show similar patterns of climate change during the last 2 kyr and apparently synchronous events are observed over a larger spatial scale. This climate-sensitive transition zone is thought to reflect precession-forced changes in seasonality, latitudinal migration of the ITCZ, and changes in the ENSO (Figs. 3  and 4). The sites in the Andean region, on the other hand, are much more influenced by local geographical variability, causing a more variable response mechanism.

Central Andes
The records from the Central Andean Altiplano suggest an oscillation in moisture availability (precipitation) on a multicentennial timescale during the last 2 kyr (Fig. 9b). These oscillations are probably due to differences in the strength of the summer precipitation. The timing of wet and dry events is not uniform between sites, probably due to local microclimates and differences in vegetation sensitivity to climate change, i.e. the high-elevation grassland (puna) versus midelevation Andean forest. The high-elevation peatland site of Cerro Llamoca is the only Altiplano site with no discernible local human impact and is the most robustly dated record used in this study, with 33 radiocarbon ages in the last 2000 years. The Cerro Llamoca record therefore likely represents Clim. Past, 12, 483-523, 2016 www.clim-past.net/12/483/2016/ the clearest palaeoclimate signal for the C Andean region. For example, records of glacial advance and retreat, and associated vegetation changes, from the Altiplano associated with the LIA are not discernible in any record, apart from Cerro Llamoca, because they are masked by changes associated with the arrival of Europeans, i.e. the abandonment of the sites and/or changes in agricultural practices. The interpretation of the climate signal from the C Andes fossil pollen records suggests that during the last 2 kyr, precipitation, rather than temperature, was the key natural driver of vegetation change. Nonetheless, the increase in temperature observed at Nevado Coropuna during the Inca period, after 0.85 ka, could correspond to El Niño or IPO forcing. Furthermore, the decrease in temperature observed at Marcacocha between 1.85 and 0.85 ka could be related to La Niña. The Pacific modes (Figs. 2 and 3) show a strong influence along the coast, which is in agreement with the results of the coastal pollen record (Urpi Kocha), where ENSO is considered responsible for extreme flooding events.
The greater sensitivity to precipitation seen in the pollen records is probably because moisture availability is the limiting factor for both vegetation and human settlement in most areas. On the Altiplano, variations in the SASM have been identified as a major driver of changes in moisture balance at Cerro Llamoca, Nevado Coropuna, and Pacucha through altering the summer precipitation. The SASM is also responsible for precipitation variations along the E Andean flank, as seen at Consuelo. As noted earlier, the highly seasonal precipitation in the C Andean region leads to rather weak correlations with ENSO and the IPO on an annual scale (Figs. 4 and 5). Notwithstanding this ENSO has been shown to have a significant influence in the C Andean region (both on temperature and precipitation) in numerous studies. It should also be noted that ENSO and IPO influence the intensity of the SASM Vuille and Werner, 2005), thereby affecting moisture delivery to the Altiplano region, but because both ENSO and monsoon rainfall tend to peak during a fairly short period between November and February, this connection is not clearly expressed in Figs. 4 and 5.

Lowland Amazon basin
The lowland Amazon basin shows a high spatial complexity in relation to the various modelled climate modes . ENSO and IPO, for example, both indicate strong negative relationships with precipitation in the northeastern quarter of Amazonia, where they induce drying over this region during their positive phase. Conversely, TSA shows a positive relationship with precipitation over the NE Amazon. Precipitation in the NE Amazon region is clearly strongly linked to tropical SSTs and ENSO variability. There are two pollen records in this region (Les Nouragues NO9203 and French Guiana VII), both of which display more local-scale forest dynamics with additional human interference. There-fore, these records are not considered suitable to investigate the effect of these modes on vegetation over the last millennia. New pollen-based reconstructions should be prioritized in this region to uncover the long-term drying effect of dominant ENSO-IPO or TSA modes on tropical lowland vegetation in the northeast. The most significant late Holocene vegetation changes are observed in records from the ecotonal areas of the S Amazon (lakes Chaplin, Bella Vista, Orícore, Carajás), where rainforest vegetation is located near the edge of its climatic range. Therefore, vegetation response to precipitation change is most likely to be observed. This rainforest expansion during the mid-to-late Holocene resulted from increasing insolation over the S tropics and the strengthening or migration of the SASM, a complex component of the climate system that is influenced by several dominant modes. Figures 4-5 show a weak negative precipitation anomaly across the lowland Amazon associated with the TNA mode. It is thought that higher sea surface temperatures in the tropical North Atlantic cause a reduction in Atlantic moisture reaching the Amazon during austral winter, thus extending the length or severity of the dry season, especially in S and southwestern (SW) Amazonia (Lewis et al., 2011). The influence of the TNA mode may therefore be important to consider in Amazonian pollen records, given the known sensitivity of vegetation in these ecotonal areas to seasonal rainfall.
Most modes in Figs. 2-3 show high correlation and regression coefficients with temperature anomalies over the lowland Amazon. Temperature anomalies can play a role, but rainforest vegetation is unlikely to have shown sensitivity to temperature changes of < 1 • C (Punyasena, 2008;Punyasena et al., 2008) but would show greater sensitivity to reductions in minimum annual temperature (i.e. frost).
Better-resolved late Holocene records originate from small lake basins (e.g. oxbows like Maxus-1, Laguna El Cerrito and Laguna Frontera), which have small pollen catchment areas. This means that they reflect predominantly local-scale changes and are, therefore, more susceptible to having their record of past environmental change dominated by signals of ancient human land use and local hydrology (e.g. savanna gallery forest) rather than regional climate. Many of these records from small basins were specifically selected in the original study to investigate local-scale human impacts around known occupation sites (Iriarte et al., 2012;Whitney et al., 2014;Carson et al., 2014Carson et al., , 2015. Examples of continuous anthropogenic signals during the last 2 kyr are Laguna El Cerrito, Laguna Frontera, and Laguna San José (Fig. 10b).
In order to address these complicating factors of pollen catchment area and the anthropogenic signal, any future effort to obtain better-resolved Holocene pollen records in the lowland Amazon should carefully consider the sampling methodology employed. Carson et al. (2014) demonstrated that sampling a combination of small and large lake basins from within the same catchment allows a distinction to be made between local-scale, anthropogenic impact and regional-scale, climate-induced vegetation changes. In rewww.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 gions such as the C Amazon, where lakes are predominantly limited to small oxbows, a sampling approach might be to analyse cores from multiple records within the same locality and to compare those records in order to identify any regionally significant pattern of palaeovegetation change Whitney et al., 2014). Oxbow lakes build potentially dynamic archives and so require careful interpretation. However, their higher sedimentation rate means that they have the potential to provide high temporal-resolution palaeovegetation records of the late Holocene, which currently are largely absent from the Amazon lowlands.
Considering the large area of the Amazon basin, the number of pollen records is very small, and by applying the PAGES-2k criteria, those numbers are further reduced. Furthermore, the records which are excluded from the analysis by these criteria include some of the most important records of climate-driven vegetation change in the Amazon basin, e.g. lakes Orícore (Carson et al., 2014), Carajás (Hermanowski et al., 2012), and lakes Bella Vista and Chaplin (Mayle et al., 2000).
In order to avoid a "black hole" situation over the Amazon lowlands in any regional synthesis, one approach may be to apply a lower threshold of dating criteria. If the selection criteria are relaxed to allow for those records that are > 500 years old and have at least two chronological control points within the last 2000 years, a further 14 records are added to the list of qualifying records. Also, if the criteria are stretched further to allow records with a lower date which is earlier than, but close to 2 ka, the Chaplin and Gentry records would also be included. Considering these records would provide coverage from the central Amazon River region, the N Brazilian Amazon, the E and NE coastal Amazon, and the southeastern and southwestern basin. However, even with these relaxed criteria, a number of key records would still be excluded, e.g. Pata D'Apolito et al., 2013), La Gaiba (Whitney et al., 2011), and Bella Vista (Mayle et al., 2000).
Any future investigation of late Holocene climatevegetation interaction may require new dating efforts to improve the age models of these key records. A Holoceneaged record from Lake La Gaiba produced by McGlue et al. (2012) has produced a better-resolved age model than the longer record from Whitney et al. (2011), which would meet the PAGES-2k criteria. However, McGlue et al. (2012) analysed the geochemical properties of sediments from a new core taken after the Whitney et al. (2011) study and did not include any pollen data. No attempt has been made subsequently to correlate the chronologies of the two records.
Although the dating resolution in the late Holocene is poor in many lowland Amazonian pollen records, it should be noted that the majority of pollen records also shows little variation in vegetation over the past ∼ 1 or 2 kyr. Whether this reflects genuine ecosystem (and climate) stability over the late Holocene or is a product of low sampling resolution within these long records is unclear. Most of these deep temporal pollen records, as they are published now, likely have subsample intervals of insufficient resolution to be able to discern high-frequency events, such as vegetation changes associated with ENSO variability. However, in some cases, such as Bella Vista (Burbridge et al., 2004) and Orícore (Carson et al., 2014), the potential for such fine-resolution temporal reconstructions may be limited by the low sedimentation rate of the basins. Often these records come from short sediment cores, in which the Holocene time interval is contained within a short depth range (i.e. < 1 m). A number of shorter records, spanning Holocene time periods, exist in the E coastal Amazon and could potentially provide high temporal-resolution reconstructions over the last 2 kyr. However, most do not currently meet the PAGES-2k dating criteria.

Southern and southeastern Brazil
The limited number of pollen records from S-SE Brazil for LOTRED-SA-2k has several reasons besides the insufficiently dated cores: (i) many archives, in particular peat bogs, have very low sedimentation rates, i.e. often 100 cm of peat deposit encompasses the complete Holocene (last 11.7 kyr), and (ii) the upper part of peat archives contains actively growing roots and is therefore difficult to date.
Despite the limited number of study sites available, general vegetation changes in S-SE Brazil can be established (Fig. 11). Pollen assemblage shifts suggest a change toward wetter conditions over the last 2 kyr, in particular due the reduction in the dry-season length. The increased moisture availability is generally thought to commence in SE Brazil between 6 and 4 ka but is particularly pronounced in S-SE Brazil during the last ∼ 1 kyr. Sites located in the highlands of S-SE Brazil along the transition zone between the subtropics and tropics are sensitive to both temperature and precipitation, but in the lower elevations, the length of the dry season plays a more important role than temperature. This dry-season length is modulated by the interplay between SASM and SACZ, which bring abundant rainfall to SE Brazil during the summer months (October-March) and the South Atlantic Anticyclone, a semipermanent high-pressure system located over the subtropical South Atlantic. The main impetus for rainfall on interannual timescales is ENSO, as El Niño events tend to bring copious rainfall to the region (Figs. 4 and 5).
According to the pollen records, the intra-annual variability plays an important role in SE and S Brazil. The generally long annual dry period during the early and mid-Holocene limited the expansion of different forest ecosystems, while a much shorter annual dry period during the late Holocene allowed a strong expansion of forests, in particular of the Araucaria forest in S Brazil. Interannual variability, influenced by the ENSO frequency, which increased during the late Holocene, may also have a certain effect on the vegetation in the region. El Niño events cause high rainfall rates in Clim. Past, 12, 483-523, 2016 www.clim-past.net/12/483/2016/ S-SE Brazil (Garreaud et al., 2009). This is consistent with results in Fig. 4, which show a positive correlation between precipitation in the region, on the one hand, and Nino3.4 and the IPO, and to a lesser extent also the TSA, on the other hand. The effect of the slightly increasing precipitation in S Brazil may be rather small, however, as rainfall is already relatively high, as inferred from the records of past environmental change from S Brazil. According to Fig. 2, the correlation of annual mean temperature over SA with the climate modes Nino3.4, IPO, AMO, and TNA suggest a slight warming associated with the positive phase of these modes (Fig. 3). Increasing temperatures sustained over a long period may cause a slight shift of tropical plant populations to higher elevations on the mountains in SE Brazil and a slight expansion of tropical plants on the southern highlands.

The pampean plain
There are several pollen records in the pampean plain that span Holocene times, but few of them have well-resolved chronologies for the last 2 kyr. Just one site fulfills all PAGES-2k criteria. Previously, pollen analyses in the pampean plain region have been carried out on alluvial sequences, or archaeological sites, which usually contain sedimentological discontinuities that impede a good chronological control. These pollen records show regional vegetation changes and climate inferences related to precipitation changes (humid, dry, or arid conditions) or sea level fluctuations, mainly on a millennial or centennial scale. Until today, few studies have focused on elucidating palaeoenvironmental changes at a high temporal resolution during the last 2 kyr. Furthermore, the pampean plain has a high number of potential sites: shallow lakes characterized by a continuous sedimentation that would provide robust age models and highquality pollen records. Conversely, the current pollen records do not have the necessary resolution to identify vegetationhuman interaction during the last 0.3 kyr, and therefore improved chronological control and higher resolution is necessary.
General climatic tendencies in the region can be inferred although few accurately dated pollen records are available. While individual palaeoecological studies reveal local developments, general patterns emerge when information from several sites is combined, such as Lonkoy and Hinojales-San Leoncio (Fig. 12b). A multi-proxy approach, including pollen analyses, shows synchronous changes in these shallow lakes from the SE pampas that are mainly a response to precipitation variations. Thus, between 2 and 0.5 ka, drier conditions than at present are inferred. Then, a transition phase towards more humid conditions is observed, which stabilizes between ∼ 0.3 and 0.1 ka, with values close to modern values (Stutz et al., 2014). These climatic inferences are valid for the southeastern region but do not extend to the entire pampean plain. In the S pampean plain, multi-proxy inter-pretation at Sauce Grande (Fontana, 2005) shows a similar change to more humid conditions at 0.66 ka and similar conditions to the present day after 0.27 ka, but pollen composition shows a low responsiveness to change (Fig. 12a). New palaeoenvironmental reconstructions based on pollen records are needed to disentangle the intrinsic ecosystem variability from climate and to elucidate whether climatic events such as the MCA or LIA had different expressions in the pampean plain.
As seen in Figs. 2-5, these plains fall outside the areas that are strongly influenced by the investigated climate modes. Nonetheless, Figs. 2 and 3 indicate that the positive phase of the TSA (a warm tropical South Atlantic) leads to a cooling over the region, while a slight warming is associated with the positive phase of the IPO. In Fig. 5, a weak positive correlation between rainfall in the region and the Niño 3.4 and IPO modes is observed, which was previously also discussed by Barros et al. (2006). The SAM on the other hand is negatively correlated with precipitation in the region (Fig. 4), consistent with results by Silvestri and Vera (2003), although this relationship has not yet been explored in pollen records as a possible influence in the region. Hitherto, studies from the pampean plain have only discussed dry or humid conditions associated with reduced or increased precipitation, but no attempt to link these observations to large-scale climate variability has been made. The situation in this region is further complicated by the fact that the moisture supply to the region stems from two distinct sources, the South Atlantic (Labraga et al., 2002) during austral winter and the SA monsoon system (Vera et al., 2006) during the austral summer. Hence, pollen-based precipitation reconstructions also need to consider changes in the seasonality of precipitation and not just climate variability associated with external influences from ENSO or the SAM. Seasonally stratified analyses of the influence of ENSO or the SAM could therefore provide additional insights into the climate-vegetation interpretation as focusing on annual mean values may mask strong seasonal signals in the same way as discussed above for the C Andes.

Southern Andes and Patagonia
Even though a large number of pollen records are available in the S Andes and Patagonia region, just 19 (between 32 and 54 • S) fulfil the PAGES-2k criteria. In Patagonia most pollen studies have been carried out with a focus on vegetation and climate change over different or longer timescales, i.e. the Pleistocene-Holocene transition (ca. 11.7 ka) or the entire Holocene (last 11.7 kyr). The pollen records are considered to mainly reflect changes in the SWWB and hence to be indicative of the polarity of the SAM. Southern records receive precipitation related to the SWWB, whereas those located to the north (40-32 • S) are also influenced by the subtropical Pacific anticyclone (SPA) that blocks winter precipitation along a latitudinal gradient (decreasing precipitation during JJA in the southern part to scarce precipitation during DJF in www.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 the northern part). Furthermore, the Andean ridge provides for a fundamental climatic divide with stronger westerlies leading to enhanced precipitation to the west of the divide, while sites located in Patagonia east of the Andean divide receive enhanced precipitation associated with winds from the east (Garreaud et al., 2013). In addition to this east-west asymmetry, the comparison between northern and southern records could also shed light on the expansion or retraction and/or latitudinal shifts of the SWWB or a differential influence of the SPA. For example, records south of 46 • S show relatively dry conditions between ∼ 1 and 0.5 ka, whereas drought occurs between 2 and 1.5 ka at sites north of this latitude (Fig. 13b). Differences in seasonality are another key feature distinguishing precipitation records in N Patagonia (summer rainfall, e.g. Lago San Pedro) from records further north in central Chile (winter rainfall, e.g. Laguna de Aculeo and Palo Colorado). Due to the regional complexity of the climate, the region cannot easily be characterized by a single climate mode. Different patterns are distinguished (Fig. 13b), due to their geographical position, latitude, position east or west of the Andes, and the intrinsic sensitivity of each record to climatic variability. Superimposed on the seasonally changing SWWB and SPA dynamics are the interannual influences of the SAM or Antarctic Oscillation and ENSO (Figs. 2-5). The positive phase of the latter (El Niño) is associated with humid winters in subtropical Chile and with dry summers in northwest (NW) Patagonia (Montecinos and Aceituno, 2003). Sites in N Patagonia and C Chile therefore might be suitable to study this asynchronous behaviour with regard to ENSO activity (e.g. Lago San Pedro and Laguna de Aculeo).
The strongest influence in the region on interannual timescales, however, is exerted by the SAM . Figures 2-5 show a highly inverse correlation with precipitation and a positive correlation with temperature over the southern tip of South America (especially south of 40 • S). The strong influence of the SAM on Patagonian climate, with drier and warmer than average conditions associated with its positive phase, is well known and consistent with previous analyses by Gillet et al. (2006) and Garreaud et al. (2009). Southernmost Patagonia therefore appears as a key area to study climate-vegetation variability associated with the SAM (e.g. Lago Cipreces). LIA and MCA chronozones are well recorded both in S and N Patagonia (e.g. Lago Cipreses, Peninsula Avellaneda Bajo, Lago San Pedro) but not in central Chile.

Conclusions
Through this review and analysis, ca. 180 fossil pollen records that fulfill at least two of the PAGES-2k criteria for robust climate reconstruction were identified for SA. Although this is still a relatively small number, compared to the total number of fossil pollen records available from SA (ca. 1400; Flantua et al., 2015), we expect that the number of high-quality sites for reconstruction of climate over the last 2 kyr is likely to increase rapidly as new work is produced. To conduct a review on this scale, it was necessary to divide SA into seven subregions. Firstly, we summarize the finding from each region and then draw broad conclusions regarding the patterns across the whole of SA.

Conclusions by region
The following are the findings for the Venezuelan Guayana highlands and uplands (seven study sites reviewed, Fig. 7): -Moisture balance and temperature: records show a higher sensitivity to moisture than to temperature. Two drought intervals were detected coeval with the Little Ice Age (LIA) in the Venezuelan Andes. Wet conditions prevailed on the tepuian summits during the last 1 kyr.
-Humans: impact has been inferred from the charcoal record, without any evidence of crops (four of seven records). The use of fires can favour the reduction in forests and expansion of the savanna, favouring the establishment of Mauritia swamps after clearing. Earliest Mauritia establishment was observed around 2 ka, but humans might have been present since the mid-Holocene leaving their signature on the present-day landscape.
-Climate modes (Table 1): both Pacific and Atlantic climate modes (Niño 3.4, AMO, IPO, and TNA modes) are predicted to have a large effect on Venezuelan Guayana, especially with regard to temperature. However, the fossil pollen records from the highlands show no responses to temperature variability, suggesting that tolerance ranges were not surpassed and that temperature thus did not produce vegetation shifts. The precipitation / evapotranspiration ratio may play an additional important role not yet studied.
Findings from the Northern Andes region (21 study sites reviewed; Fig. 8): -Moisture balance and temperature: fossil pollen records are both moisture balance and temperature sensitive, with tropical lowlands more sensitive to moisture and Andean areas more sensitive to temperature. Overall wetter conditions with warm and cold episodes are seen during the last 2 kyr. These shifting temperatures are displayed asynchronously in the records, and changes in ENSO frequency have been detected in multiple records.
-Humans: Andean records without human impact are rare (just 4 of the 21 records) and a wide range of indicators for human activity is found. These include deforestation (loss of tree taxa) and the appearance of introduced taxa, e.g. palms, crops, and disturbance taxa. The high level of evidence of humans in this region is not surprising given that the history of the human occupation of the Andes goes back to the Lateglacial period (ca. 10 ka; Van der Hammen and Correal Urrego, 1978).
-Climate modes: the altitudinal gradient in temperature is most importantly modulated by Pacific modes (Niño3.4) and the TNA. Records show a close match with precipitation variability trigged by ENSO that displays a highly diverse spatial pattern throughout the region (Fig. 4).
Findings from the Central Andes (seven study sites reviewed; Fig. 9): -Moisture balance and temperature: fossil pollen records are more sensitive to changes in moisture balance than temperature. The records on the E Andean flank (Amazon flank) suggest overall moist conditions during the last 2 kyr, while the W Andean flank (valleys and Pacific flank) shows a succession of dry and moist episodes. Generally drier conductions occurred in the C Andes between 1.2 and 0.7 ka when compared with the rest of the last 2 kyr. Findings from the lowland Amazon basin (19 study sites reviewed; Fig. 10): -Moisture balance and temperature: fossil pollen records from the lowland Amazon basin are moisture sensitive and indicate continuously wet climate throughout the last 2 kyr; however, centennial-scale shifts are observed in terms of forest composition attributed to hydrological change.
-Humans: human activity has been detected in most records (15 of 19 sites), evidenced by fire (charcoal abundances), forest clearance, and crops, e.g. Zea mays and Manihot esculenta. After European contact, land use decreases as shown by a decline in burning around 0.5 ka.
-Climate modes: precipitation in the NE Amazon region is strongly linked to tropical sea surface temperatures and ENSO variability. ENSO and IPO induce drying in the NE Amazonia during their positive phase, while TSA induces precipitation. Both the Pacific and the Atlantic modes show high correlation and regression coefficients with temperature anomalies over the lowland Amazon.
Findings from southern and southeastern Brazil (seven study site reviewed; Fig. 11): -Moisture balance and temperature: Records are moisture sensitive and indicate continuously wet climate throughout the last 2 kyr. Changes in forest composition suggest a relatively humid and warm phase during the LIA, in contrast to other regions.
-Humans: most human impact occurred during the last 0.4 kyr as indicated by the increased use of fire. Furthermore, in the southern part of Brazil, human modification of ecosystems is indicated by the appearance of introduced taxa such as Pinus and Eucalyptus.
-Climate modes: Nino3.4, IPO, AMO, and TNA suggest a slight warming associated with the positive phase of these modes. There is a positive correlation between precipitation in the region, on the one hand, and Nino3.4 and the IPO, and to a lesser extent also the TSA, on the other hand. The ENSO frequency influences the interannual variability in precipitation and may affect the vegetation in the region where the duration of the dry season is more important than temperature.
Findings from the pampean plain (three study sites reviewed; Fig. 12): -Moisture balance and temperature: fossil pollen records are moisture sensitive and do not detect temperature shifts. From 2 to 0.7-0.4 ka, drier climatic conditions than at present are inferred, while after 0.3 ka a noticeable increase in precipitation occurred (more positive moisture balance).
-Humans: all records have human impact but this widespread impact only occurs during the last 0.1 kyr, and is a consequence of the introduction of exotic tree species such as Eucalyptus and Pinus.
-Climate modes: models suggest that the climate modes explored here exert only weak influences over the pampean region. Precipitation seasonality probably plays a more important role as moisture supply stems from distinct sources during the year.
Findings from the Southern Andes and Patagonia (23 study sites reviewed; Fig. 13): -Moisture balance and temperature: fossil pollen records are both a moisture and a temperature sensitive, showing a highly diverse pattern of alternating phases during the last 2 kyr. One record displays a major ENSO frequency between 1.8-1.3 and 0.7-0.3 ka.
-Humans: impact is present in most records (17 out of 23). Only the last centuries show clear human intervention associated with European arrival through the occurrence of Plantago (indicator of overgrazing), increased grasses, introduced taxa (Pinus), and crop-related herbs (Rumex). European colonization followed a clear north to south migration pattern, while evidence for the presence of earlier human populations in the region is not conclusive from palaeoecological records.
-Climate modes: the strongest influence in the region is exerted by the SAM for both temperature and precipitation. The pollen records are considered to mainly reflect changes in the southern westerly wind belt and hence to be indicative of the SAM. ENSO influences mostly precipitation.

General conclusions
On the basis of the region-by-region assessments from SA we conclude more generally that Clim. Past, 12, 483-523, 2016 www.clim-past.net/12/483/2016/ -the low number of SA records that fulfill all the PAGES-2k criteria (only 44) is a consequence of the age and quantity of the sediments recovered (which place fundamental limits on the duration and resolution of any study) and of the focus of the original research. Many SA records have been excluded because their long time span (> 10 ka) coupled with a relatively slow sedimentation allows only low temporal resolution of sampling; furthermore, slow sedimentation rates mean that many records do not have radiocarbon ages from within the last 2 kyr.
-pollen records in SA can detect long-distance (between sites) synchronicity (differences and similarities) in vegetation changes as an indication of regional precipitation and temperature variability; however, they can also detect the local-scale change or variability, which needs to be understood to determine whether a long-distance signal is present. This interaction between long-distance and local-scale signal has long been a problem for palynologists (e.g. Jacobson and Bradshaw, 1981), but interestingly in SA it seems that the degree of variation in signal varies between regions, i.e. in lowland regions there seems to be less between-site variability (consistent long-distance signal) compared with Andean sites (high local-site-specific variability). This variation between lowland and Andean sites is probably a function of topographic complexity and hence lowland pollen records provide a relatively cleaner long-distance signal, from which large-scale atmospheric circulation (climate) change can be assessed. However, we show that fossil pollen records from all regions of SA can be compared to help better understand past changes in the intensity and area of influence of different climate modes, such as ENSO or the AMO.
-throughout SA a number of overlapping climate modes operate. We assess the correlation and regression coefficients of the six most relevant climate modes to identify the modes with the most significant influence on interannual temperature and precipitation variability. Every single pollen record most likely captures the signal of various climate modes (Figs. 2-5), although they do not all operate in the same frequency bands and modes interact with one another through constructive interference. The causes of ambiguous climate-vegetation responses observed in pollen records can therefore probably be ascribed to the degree of climate mode interaction at a given location and to the degree of climate mode interaction interaction with non-climate variables (e.g. ecological interactions and natural hazards).
-the geographical location (latitude, longitude, and altitude) of a record naturally affects the sensitivity of a study site to temperature-or precipitation-related forcing (Figs. 7-13). The baseline for understanding climate-driven changes in vegetation is related to either of these variables, but interpreting pollen records in terms of a response to large-scale climatic forcing may yield further insights as it allows an attribution of temperature-and/or precipitation-driven changes to forcing from climate modes originating in either the Atlantic or Pacific Ocean.

Recommendations
Below we list a few specific recommendations for future engagements between climate-and pollen-related studies.
1. Quantitative translation from pollen metrics to climate variables: assembling a meaningful multisite and multiproxy data set is hampered by the current gap between the palynological and the climate dynamics and modelling community, both in terms of interpretation and quantitative translation of pollen data into climate indicators. This gap can be narrowed when pollen studies provide their own temperature or precipitation approximations if the data are suitable for that purpose. There are only a few pollen studies that provide a quantitative interpretation of their pollen data in terms of a climate variable. In the Andes, La Cocha-1 (González et al., 2012) and Papallacta PA1-08 (Ledru et al., 2013a) provide such estimates of climatological changes. In both cases the percentage of arboreal pollen was used as a measurement of moisture or temperature changes. Similarly Punyasena et al. (2008) and Whitney et al. (2011) present innovative methodologies for climate reconstructions in the lowland tropics, as do Markgraf et al. (2002), Tonello and Prieto (2008), Tonello et al. (2009Tonello et al. ( , 2010, and Schäbitz et al. (2013) in southern SA. Providing additional climate estimates is not a common feature in palynological studies, and this missing link becomes more obvious when the palynological community is being engaged in a multidisciplinary effort such as LOTRED-SA and PAGES-2k.
2. Multi-proxy based research should become a mandatory goal for all further investigations. Caution should be exercized when interpreting apparently contradictory records provided by different groups for the same region; the interpretation of climatic and anthropogenic signals in each record may be based on very different (indirect) proxies. Hence, the apparent asynchronies or contradictory interpretations could simply occur as a result of methodological artefacts (e.g. by not including charcoal records, non-pollen palynomorphs, geochemical analyses). On the other hand, multi-proxy based research is especially relevant for those areas where human impact has been found for the last 2 kyr, but a climatic interpretation is the aim of the study. Developing proxies suitable for generating independent climate reconstructions from lake sediments in SA inwww.clim-past.net/12/483/2016/ Clim. Past, 12, 483-523, 2016 cludes considering chironomids (Matthews-Bird et al., 2015;Williams et al., 2012), while indications of humans can come from non-pollen palynomorphs, such as the dung fungus Sporormiella (Williams et al., 2011).
3. For the stated purposes of the current and future PAGES initiatives, researchers should be motivated to further improve chronologies for existing sites. There is a need to increase efforts in high-resolution studies with an accurate chronology for the last 2 kyr. At the same time, the PAGES-2k criteria should be adjusted for pollen records, especially by applying a lower threshold of dating criteria. A region such as the lowland Amazon is notorious for its paucity of records with good dating (e.g. Ledru et al., 1998). Therefore, additional valuable sites available should be considered for the overall purpose of studying vegetation-climate linkages.
4. Further advances in understanding climate-human relationships are also likely to be made by the integration of palaeoecological and archaeological data (e.g. Mayle and Iriarte, 2014) through conceptual modelling, which can provide a framework for identifying patterns and trajectories of change (e.g. Gosling and Williams, 2013).

5.
Multi-proxy studies should compare data between different regions and records that are comparable in terms of chronology and resolution. Comparisons may yield insight into anti-phased climate variability resulting from certain dominant climate modes (e.g. a comparison between the coast of Colombia and NE Brazil-Guayanas, on the one hand, and S Brazil and E Argentina, on the other).
6. All Andean zones are quite active from tectonic and volcanic points of view, and those drivers will have had significant impacts on the vegetation and maybe in the fossil pollen records as well. However, this aspect was only discussed for the southern region of the Andes. A chronology database focused on tephra control points could support current chronology constraints and improve comparison between records. The recent geochronological database within the LAPD can support such a multi-proxy approach for palaeoecological integration (Flantua et al., 2016).
7. In this paper we focused less on the seasonal contrasts throughout the continent, but in southern SA the seasonal component is extremely important, as precipitation shifts latitudinally over the course of the year. Precipitation in this region is the limiting factor for vegetation growth and pollen production. Key issues that need further study include (a) a better understanding of the relationship between winter and summer rainfall, (b) the question of whether this relationship has remained stationary over the last 2 kyr, and (c) the question of whether changes in the intensity or location (latitudinal shift) of rainfall have occurred.
8. High-resolution time series should be explored with frequency analysis to find support for operating climate modes.
9. Optimal exploration of the presence of climate modes in pollen records requires a specific research design. Temporal resolution should be increased to below decadal scale, chronological control of the sediments should be optimized, and the main frequencies in the time series should be analysed and compared with a frequency spectrum to be developed that shows characteristics of the climate modes.
The Supplement related to this article is available online at doi:10.5194/cp-12-483-2016-supplement. Ricardo Villalba, José Ignacio Martínez, Catalina González, and Thorsten Kiefer for organizing the LOTRED-SA Special Issue "Climate change and human impact in Central and South America over the last 2000 years" and for allowing us to submit our work to it ((http://www.clim-past.net/special_issue84.html).
Edited by: R. Villalba