Late-Holocene maize cultivation,fire,and forest change at Lake Ayauch i,Amazonian Ecuador

Abstract

A high-resolution paleoecological record provides a 2690 year-long fossil pollen and charcoal history from Lake Ayauchⁱ, Ecuador, in lowland Amazonia. The record begins with a landscape that is already partially deforested and in which maize is being grown. Dated charcoal fragments from local soils coincide with fire events and peaks of land clearance seen in the lake sediment record. After c. AD 550 grass pollen becomes less abundant, as a broad array of forest types show small increases in abundance. Between c. AD 750 and 1280, Zea mays pollen was at its most abundant. Although maize cultivation continued until the AD 1700s, forest pollen abundance showed a significant increase at c. AD 1260. Another transition at c. AD 1420, which saw a transition from dominance by early successional taxa and an increase in mid-successional elements, suggests the onset of reduced human activity at the site. Fossil maize is found in a lower proportion of samples, disappearing altogether for a century in the late 1700s. Forest taxa increase in abundance and charcoal disappears from the record at c. AD 1790. These data suggest a complex social history prior to and following European arrival with phases of forest clearing and episodes of apparent regrowth at c. AD 500, 950, and 1260. Increased forest pollen after c. AD 1260 and a reduction in maize pollen abundance suggests some abandonment, with a second, relatively late, depopulation following European Conquest (c. AD 1790). Evidence is not found supporting reforestation associated with European arrival.

Keywords

Amazonia cultivation European colonization fossil charcoal fossil pollen maize

Introduction

That pre-Columbian people exerted an influence on some Amazonian forests is now well accepted (e.g. Balée, 1989; Brugger et al., 2016; Bush et al., 1989, 2015a; Erickson, 2006; Glaser and Birk, 2012; Levis et al., 2017; Neves et al., 2004; Pärssinen et al., 2009; Roosevelt, 1991; Watling et al., 2017). Little is known, however, about how site use affects long-term ecological trajectories, and whether apparent changes in land use were driven by internal social dynamics, invasions from the outside, or influences such as climate change.

Human population growth in Amazonia is thought to have increased persistently, but slowly, from the early Holocene until c. 5000 years ago, with an exponential population growth thereafter and an increased pre-Columbian site frequency after c. 2500 years ago (e.g. Bush et al., 2015a; Erickson, 2006; Goldberg et al., 2016; McMichael and Bush, 2019; McMichael et al., 2017). The Ecuadorian Amazon has a long history of human occupation but sparse archeological records. Evans and Meggers (1968) conducted the first systematic archeological survey of this region along the Napo River. Based on pottery designs, they identified four distinct cultural phases and provide ages for them: the Yasuní (50 BC), Tivacundo (AD 510), Napo (AD 1168–1480), and the Cotacocha from the European era (after AD 1500) (Evans and Meggers, 1968). The best-known regional archeological record is that of the Upano valley (Figure 1; Supplemental Materials Upano valley occupation, available online) (Rostain, 2012; Salazar, 2008). Here, at an elevation of between 1000 and 1500 m, there was long term occupation and site modification through cultivation and mound-building. The Upano valley site was occupied by different cultural groups in the last 4750 years, including the Upano from c. 500 BC to AD 400–600 and the Huapula from AD 800 to 1200 (Rostain, 2012; Salazar, 2008). Sometime between AD 400 and 600, occupation of the site was interrupted until AD 800 by an eruption of the Sangay volcano (Figure 1).

Figure 1.

Map of the study area showing (a) the study site Lake Ayauchⁱ (1, red circle) and the discussed sites Lake Kumpak^a (1, overprinted red circle at this scale; Åkesson et al., 2021; Liu and Colinvaux, 1988), Lake Maxus 4 (2, white circle; Weng et al., 2002), Lake Pumacocha (3, white upside-down triangle; Bird et al., 2011), Palestina cave (4, white upside-down triangle; Apaéstegui et al., 2014), Sangay volcano (5, white triangle), and Upano valley (6, red square; Rostain, 2012, Salazar, 2008), (b) Google Earth image of Lake Ayauchⁱ, Santiago River, and the surrounding landscape; and (c) bathymetry data of Lake Ayauchⁱ showing coring location (red circle) (modified from Bush and Colinvaux, 1988). In (a), red circles denotes paleoecological records with signs of human occupation, white circles denotes paleoecological records without signs of human occupation, white upside-down triangles denotes paleoclimate records, white triangles denotes geographical landmarks, and red squares denotes archeological sites.

The naturally wet environment of western Amazonia makes it very unlikely that fire would spread without the assistance of people, hence the findings of charcoal in paleoecological records are strong indicators of human disturbance (Aragão et al., 2007; Bush et al., 2007; Nepstad et al., 2004). Published paleoecological records in Ecuadorian Amazonia showed strong contrasts in vegetational history according to whether a site was occupied and how it was used. At the Maxus 4 site (Figure 1), which lacked any signs in the paleorecord of human occupation, there was almost no charcoal (e.g. fire events) for thousands of years and all changes in the fossil pollen record could be related to a hydrarch succession (Weng et al., 2002). Similarly, at Lake Kumpak^a (Figure 1), much of the last 2500 years showed little fire, but there was periodic disturbance of the system with evidence of deforestation and crop cultivation (Åkesson et al., 2021; Liu and Colinvaux, 1988). Lake Ayauchⁱ (Figure 1), which lay within 25 km of Lake Kumpak^a, had a 6000-year history of maize cultivation with frequent fires (Bush and Colinvaux, 1988; Bush et al., 1989; McMichael et al., 2012c; Piperno and Pearsall, 1998). Terrestrial soil cores collected around Lake Ayauchⁱ and investigated for charcoal and phytoliths, further indicated the presence of localized burning and crop cultivation within 1 km of the lake for the last 6000 years (McMichael et al., 2012c). The original pollen analysis of the site (Bush and Colinvaux, 1988; Bush et al., 1989), however, lacked the temporal resolution to assess the ecological responses that followed these disturbance events.

No high-resolution western Amazonian climate record exists during periods of highest human occupation, that is, the last 2500 years, to assess whether and how climatic changes drove patterns of land use. The isotopic records closest to western Amazonia are Palestina cave (325 km distance from Lake Ayauchⁱ; Apaéstegui et al., 2014) in the Andean/Amazon foothills of northern Peru, which spans the last 1600 years, and the high Andean Pumacocha lake sediment record (870 km distance from Lake Ayauchⁱ; Bird et al., 2011), which provides a subdecadally-resolved isotopic history of the last 2700 years (Figure 1). From these data, it appears that the period between c. 700 BC and AD 500 showed rather minor changes in precipitation compared with those of the period from c. AD 500 to 1400 (Bird et al., 2011). Within this period of dynamic climate change lay the medieval climate anomaly (MCA) from c. AD 950 to 1200. The MCA probably peaked between AD 1000 and 1200, but other regional studies highlight the period from AD 800 to 1300 as being both culturally and climatically distinct (Binford et al., 1997; Thompson et al., 2013). This period was locally expressed by strong floods and droughts, ending with a transition to increasingly wet conditions between AD 1200 and 1400 (Bird et al., 2011). Culturally, this period appears to have been a time of retreat to more defensive positions amid escalating societal tensions (de Paula Moraes and Neves, 2012)

European arrival (AD 1492) is thought to have triggered a massive, perhaps 90%, reduction in New World human populations (Black, 1992; Cook, 1998; Dobyns, 1966). The subsequent abandonment of land following depopulation has been suggested to have resulted in substantial forest regrowth (Koch et al., 2019). The young, vigorously regrowing forest would have absorbed a large amount of CO₂ (Chazdon, 2003; Martin et al., 2013). It was hypothesized that the uptake of carbon in regenerating forests was so rapid that it was a major contributor to the Orbis Spike of AD 1620, a c. 7–10 ppm reduction in global atmospheric CO₂ concentrations (Koch et al., 2019; Lewis and Maslin, 2015). A recent analysis of 39 fossil pollen records from Amazonia cast doubt on any such widespread forest recovery, but rather pointed to a distinctly pre-European phase of forest recovery that took place between c. AD 750 and 1300 (Bush et al., 2021). This inference was based in part on new pollen data from Lake Ayauchⁱ.

Lake Ayauchⁱ, in lowland Ecuador (Figure 1), provided one of the first fossil pollen records from western Amazonia (Bush and Colinvaux, 1988). The fossil pollen sequence, originally cored in 1983 stood out for its length of pre-Columbian land use and early maize cultivation (Bush and Colinvaux, 1988; Bush et al., 1989). Four conventional ¹⁴C dates supported the chronology of the original Lake Ayauchⁱ record that spanned the last 7000 years, but the youngest of those ages had a calibrated age of c. 500 BC. Since that first study, significant improvements have been made in ¹⁴C dating, pollen identification, and metadata for environmental and human cultural change. We have adopted more accurate measures of quantifying charcoal and maize and, perhaps most importantly, have generated new research questions that demand higher sampling resolution. Here, we investigate the continuity of occupation, types of land management, trajectories of land use and forest recovery, effect of European arrival on these systems, and the relationships between people and climate with a new high-resolution record spanning the last 2690 years from Lake Ayauchⁱ. Specifically, we ask, “Was there a steady trajectory of increasing deforestation and use of fire prior to European arrival?” and “Were the largest changes in land use during the last 2700 years associated with European arrival?”.

Study area

Lake Ayauchⁱ (3°02′41″S, 78°02′05″W) is located in southeastern Ecuadorian Amazonia (Figure 1). The natural vegetation of the region is dense tropical rainforest, with a mean annual temperature of approximately 24°C. The mean annual precipitation of the region is around 2000–3000 mm, with a poorly defined dry season that lasts 0–2 months (Bush and Colinvaux, 1988).

Lake Ayauchⁱ lies within 1 km of the Santiago River, a tributary of the Marañón River (Figure 1). At 300 m above sea level, the basin of the lake is suggested to have been an explosion maar, which is a shallow and broad volcanic crater once formed by an eruption (Colinvaux et al., 1985). Lake Ayauchⁱ is roughly circular with a diameter of 380 m and has two sub-basins with maximum water depths of 45 and 25 m (Figure 1). The lake is strongly oligotrophic, unlike the majority of Amazonian lakes, which are generally eutrophic, and is also unusual for an Amazonian lake in that it appears blue from the air and has a secchi depth of 4.5 m (Colinvaux et al., 1985).

The indigenous community adjacent to Lake Ayauchⁱ belongs to the Shuar Nation. The pronunciation of Ayauchⁱ and nearby Lake Kumpak^a in the Shuar language is reflected in the elevated vowel at the end of its name (Colinvaux, 2007). The modern landscape surrounding Lake Ayauchⁱ is a patchwork of cultivated and fallowed land with roughly 25–30% of the area within 1 km of the lake currently deforested (Figure 1) (source: Google Earth imagery).

Materials and methods

At Lake Ayauchⁱ, a 1.7-m long core, which did not capture the upper c. 0.60 m of sediment, and a 0.83-m long surficial core were obtained from 20 m water depth in July 2007 and June 2016, respectively. The 2007 core was retrieved using a Colinvaux-Vohnout piston corer (Colinvaux et al., 1999) and the 2016 core was retrieved using a Universal percussion corer by Aquatic Research Instruments. The uppermost 30-cm of the 2016 core was extracted in 0.5-cm increments and bagged in the field. All other core sections were sealed in the field and transported to Florida Institute of Technology, where they were stored at 4°C. As bedrock was not reached, the sediment cores represent the limit of coring ability.

Radiocarbon dates for the 2007 core (n = 5, Table 1; McMichael et al., 2012c) and the 2016 core (n = 6, Table 1) were re-calibrated and calibrated, respectively, using the IntCal20 calibration curve (Reimer et al., 2020). The “rbacon” package version 2.5.7 (Blaauw and Christen, 2011) was used to construct individual age-depth models of each core. The Ayauchⁱ soil core charcoal samples (n = 6; McMichael et al., 2012c) were also re-calibrated (Table 2) using the Intcal20 calibration curve (Reimer et al., 2020) and the “rbacon” package version 2.5.7 (Blaauw and Christen, 2011). A southern Hemisphere correction was not required for the ¹⁴C dates as moisture arriving to Amazonia mainly originates from the subtropical Atlantic Ocean in the northern Hemisphere (Vuille et al., 2003). The two sediment cores were cross-correlated using changes in sediment stratigraphy and the ¹⁴C dates to produce a single chronology that overlapped by c. 150 years. Subsamples were taken from both the 2007 and the 2016 core at depths where ages overlapped, with results reported from both cores.

Table 1.

Radiocarbon dates and calibrated ages for the 2016 core and the 2007 core of Lake Ayauchⁱ, Ecuador.

Lab ID number	Core/Sample type	Depth (cm)	¹⁴C age ±1σ	Mean probability (cal. yr BP)	Mean probability (AD/BC)
OS-136403	2016/Bulk	17	160 ± 20	218	AD 1732
OS-142568	2016/Bulk	26	430 ± 15	454	AD 1496
OS-136467	2016/Bulk	31.5	460 ± 20	517	AD 1433
OS-142567	2016/Bulk	51	935 ± 15	840	AD 1110
OS-136404	2016/Bulk	62.5	1090 ± 20	1010	AD 940
OS-136405	2016/Bulk	80	1480 ± 20	1353	AD 597
OS-72629	2007/Bulk	13.5	1480 ± 30	1351	AD 599
OS-72694	2007/Bulk	90.25	1800 ± 30	1703	AD 247
OS-72695	2007/Bulk	111	1910 ± 30	1805	AD 145
OS-72696	2007/Bulk	145	2340 ± 35	2331	381 BC
OS-72696	2007/Bulk	178.5	2580 ± 35	2734	784 BC

Table 2.

Recalibrated ages from soil charcoal fragments from around Lake Ayauchⁱ (McMichael et al., 2012c).

Lab ID number	Core/Sample type	Depth (cm)	14C age ± 1σ	Mean probability (cal. yr BP)	Mean probability (AD/BC)
OS-73251	Soil core 13/Charcoal	13–25	1250 ± 25	1200	AD 750
OS-80262	Soil core 9/Charcoal	32–39.5	1670 ± 25	1560	AD 390
OS-80326	Soil core 11/Charcoal	19–26	1720 ± 35	1610	AD 340
OS-80325	Soil core 11/Charcoal	12–19	1940 ± 35	1860	AD 90
OS-80263	Soil core 9/Charcoal	32–39.5	2120 ± 25	2080	130 BC
OS-70865	Soil core 10/Charcoal	52–60	2300 ± 30	2330	380 BC

Subsamples (n = 105; volume = 0.25 or 0.5 cm³) for pollen analysis were taken at every 4–10-cm from the 2007 core and at every 1-cm from the 2016 core. Samples for pollen were treated according to standard methodology (Faegri and Iversen, 1989; Stockmarr, 1971), the samples were spiked with exotic Lycopodium spores, treated with 10% HCl, 10% KOH, 10% NaP₂O₇, and acetalized (9:1 ratio of (CH₃CO)₂O and H₂SO₄). In addition, sodium metatungstate heavy liquid flotation (density: 2.0–2.1 g/mL) was used to remove inorganic matter before the samples were mounted in glycerol. A minimum of 300 terrestrial pollen grains, excluding Cecropia, were counted in each sample using a Zeiss Axioskop photomicroscope at magnifications of ×400 and ×630. Cecropia was excluded from the terrestrial pollen sum because of its high abundance and to prevent the masking of other less common taxa. Cecropia pollen percentages were calculated based on the 300 terrestrial pollen sum. After the completion of the initial pollen counts, all pollen extracts were filtered using a 60 µm mesh and reanalyzed to search for maize grains (>75 µm) (Holst et al., 2007; Whitney et al., 2012). The Neotropical Pollen Database (Bush and Weng, 2007), standard text for the region (Colinvaux et al., 1999; Roubik and Moreno, 1991), and the reference collection at Florida Institute of Technology were used for identifications. As some wind-pollinated (anemophilous) tree pollen can have long dispersal distances (Faegri and van der Pijl, 1979), the non-anemophilous arboreal/woody pollen types (erring toward inclusion rather than exclusion of types that could be vines, shrubs, or trees) were calculated to provide an index of local forest pollen abundance (Supplemental Materials Table 1, available online).

A Detrended Correspondence Analysis (DCA) (Hill, 1979) on fossil pollen types comprising at least 0.5% of the pollen sum or occurring in five or more samples (Birks and Gordon, 1985), which reduced the dataset to 95 taxa, was performed using the “vegan” package (Oksanen et al., 2013). A CONISS constrained clustering analysis and the broken stick model were performed to aid with the zonation of fossil pollen assemblages using the “vegan” (Oksanen et al., 2013) and the “rioja” (Juggins, 2019) packages (Supplemental Materials Figures 1–2, available online). All chronologies and data analyses were performed in R (R Development Core Team, 2021).

The charcoal sediment data from the 2007 core, originally published by McMichael et al. (2012c) and analyzed at 1-cm intervals (n = 179; volume = 1 cm³), were re-plotted here using the updated IntCal20 core chronology. Charcoal fragments from the 2016 core were isolated from 165 subsamples (volume = 0.5 cm³) using standard laboratory techniques (McMichael et al., 2012a, 2012b; Whitlock and Larsen, 2002), deflocculated using <9% H₂O₂, sieved with a 180 µm mesh, and identified using a Zeiss Stemi-C stereoscope at magnifications of ×70 and ×100. ImageJ software (Rasband, 1997–2016) was used to calculate the surface area (mm²/cm³) of charcoal fragments from each core. Sediment charcoal influx (mm²/cm³/yr) and sediment accumulation (mm/yr) rates were also calculated.

Results

The 2007 and 2016 sediment cores raised from Ayauchⁱ represent a total of c. 2.5 m of sediment and were of a uniform, dark brown-black gyttja with no laminations. The laminated sections documented in the 1983 core raised from Ayauchⁱ, which were mostly older than 700 BC (2650 cal. yr BP), were not observed in the 2007 and 2016 cores. The updated chronology of the 2007 core provided a core-top age of c. AD 670 and a basal core age of c. 780 BC (Table 1). The Ayauchi 2016 core chronology was based on six ¹⁴C dates, all of which were accepted, and provided a basal core age of c. AD 510 (Table 1). Together, the two cores provided a robust chronology spanning the last 2690 years (Figure 2). Sedimentation rates for the 2007 and 2016 cores were 6–50 mm/yr and 3–9 mm/yr, respectively. Consequently, the pollen sampling interval in the last c. 1500 years is about 10–20 years, while between c. 680 BC and AD 450 the record has a resolution of 30–160 years.

Figure 2.

Age-depth models of the sediments from Lake Ayauchⁱ, Ecuador, consisting of core 2016 (left) and core 2007 (right; McMichael et al., 2012c), showing age (AD/BC) on the y-axes and depth (cm) on the x-axes. Sediment accumulation rates through time (mm/yr; solid black line) for both cores are plotted against age (AD/BC). The Markov Chain Monte Carlo iterations, prior (green curve) and posterior (gray histogram) distributions for the accumulation rate, and the memory for the cores Bayesian age-depth models (Blaauw and Christen, 2011) are also shown.

The pollen record of Lake Ayauchⁱ included 168 terrestrial pollen types, where the broken stick model indicated that three significant zones could be identified using the constrained hierarchical cluster analysis (CONISS) (Supplemental Materials Figure 2, available online) on the pollen assemblages (Figure 3): 680 BC–AD 530 (zone 1), AD 530–1420 (zone 2), and AD 1420 to modern time (zone 3).

Figure 3.

Percentage diagram for fossil pollen from Lake Ayauchⁱ, Ecuador, plotted against age (AD/BC) showing the most abundant taxa, pollen concentration (grains/cm³ × 1000), and scores of DCA Axes 1 and 2. Zea mays pollen are shown as presence (gray triangles = 75–80 µm, gold circles ⩾ 80 µm). Comb.-Mela. abbreviates Combretaceae and Melastomataceae. Online version shows grasses (gold), herbs (pink), non-anemophilous forest (blue), and anemophilous forest (green). Pink shaded bar denotes the timing of the medieval climate anomaly (MCA). Which core, 2016 or 2007, was used for which depths is also shown, where “overlap” indicates where the two cores overlapped for c. 150 years.

Lake Ayauchⁱ zone description

Zone 1 (680 BC–AD 530; 170–28 cm 2007-core; 83–82 cm 2016-core)

Pollen concentrations ranged from 228,100–1,727,000 grains/cm³ (Figure 3). Cecropia fluctuated between 60 and 130% (here and hereafter all % are of the non-Cecropia pollen sum). Poaceae pollen was abundant (7.5–21.5%) and 19 out of 20 samples contained Zea mays pollen grains. Disturbance indicators other than Cecropia and Poaceae, such as Iresine, Asteraceae, Acalypha, and Piperaceae, accounted for 13–28% of the pollen sum. Forest taxa, for example, Moraceae-Urticaceae, Alchornea, Apiaceae, Boraginaceae, Brosimum, Celtis, Combretaceae-Melastomataceae, Schefflera (ex. Didymopanax), Forsteronia, Sloanea, and Tapirira, had values of 55–82%. Local (non-anemophilous) forest types were generally 15–25% of the pollen sum. Charcoal fragments were consistently found throughout the zone (76 of 151 samples) with charcoal influx rates of 0–13.85 mm²/cm³/yr. The samples in zone 1 clustered toward the negative ends of the DCA Axes 1 and 2, mainly driven by grasses and the disturbance elements (Figure 4).

Figure 4.

The Detrended Correspondence Analysis (DCA) biplot of fossil pollen data from Lake Ayauchⁱ, Ecuador, with eigenvalues 0.09679 and 0.04801 on Axes 1 and 2, showing sample (left) and species (right) scores. Samples are separated in three pollen zones: zone 1 (680 BC to AD 530; blue squares), zone 2 (AD 530 to AD 1420; yellow circles), and zone 3 (AD 1420 to modern; green diamonds). A color gradient (dark to light) shows the oldest to youngest sample in each zone. Online version shows species categorized in four vegetational groups: grasses (gold), herbs (pink), non-anemophilous forest (blue), and anemophilous forest (green). C-M abbreviates Combretaceae-Melastomataceae and M-U abbreviates Moraceae-Urticaceae.

Zone 2 (AD 530–1420; 28–0 cm 2007-core; 82–32 cm 2016-core)

Pollen concentrations ranged from 190,600–1,317,800 grains/cm³ (Figure 3). Z. mays pollen grains were identified in 51 out of 52 samples. The zone was characterized by generally lower but variable Poaceae abundances, with peaks up to 13.5%, c. AD 760–840 and c. AD 1120–1340, but otherwise low abundances of <3%. These peaks in Poaceae matched declines in local forest abundance (Figure 3). Forest elements fluctuated between 52 and 93% in the zone but showed a strong increase beginning c. AD 1260 and reaching a peak of 42% in AD 1420. Cecropia had values around 30–100% throughout the zone, except for three major peaks around AD 815 (170%), AD 1010–1110 (120–320%), and AD 1289–1300 (200%). Charcoal fragments were found in 49 out of 79 samples throughout the zone with charcoal influx rates of 0–30 mm²/cm³/yr.

In the DCA, samples were scattered along Axis 1 (Figure 4). Samples containing an increased abundance of forest elements were at the positive end of Axis 1, whereas samples containing high abundances of open elements were at its negative end. Axis 2 further divided the samples with increased disturbance/openness, with samples rich in Cecropia having positive scores.

Zone 3 (AD 1420 to AD 2016; 32–0 cm 2016-core)

Pollen concentrations were significantly lower in zone 3 than in the other zones, but still relatively pollen rich with 78,700–335,100 grains/cm³ (Figure 3). Z. mays grains were identified in fewer samples (15 out of 32 samples), as Poaceae and disturbance elements decreased to <3%, and <15%, respectively. Forest elements had high values (>80%), with local forest declining toward the modern value of c. 25%. Cecropia increased steadily throughout the zone, reaching 200% at c. AD 1980. Charcoal fragments were routinely found throughout the zone (14 out of 32 samples) until c. AD 1790, with no charcoal fragments found thereafter. Charcoal influx rates were between 0 and 28.8 mm²/cm³/yr.

The zone 3 samples plotted toward the positive end of DCA Axis 1, due to the increased abundance of forest elements (Figure 4). DCA Axis 2 further divided the samples, with the older samples in zone 3 clustering at the negative end, and the later samples clustering toward the positive end. Within the last 200 years, the Cecropia-rich samples plotted toward the negative end of DCA Axis 1.

Discussion

We present a 2690 year-long fossil pollen and charcoal record from Lake Ayauchⁱ that reflects both climate change and modification of the environment by humans. Because we are using cores collected 9 years apart, it is important to establish that differences observed through time are not solely attributable to which core was being analyzed. The transition between the 2007 and the 2016 core occurred at c. AD 500. Importantly, the pollen assemblages and concentrations from the basal sample of the 2016 record were very similar to the uppermost samples of the 2007 core (Supplemental Materials Figure 3, available online). While this transition was temporally close to one of the major changes in the fossil pollen signatures (zones 1–2), it occurred within the 2016 core, not at the boundary between cores. Marking the transition, Poaceae abundances fell and forest abundances increased at c. AD 550. More intense sampling after AD 500, may have accounted for the observed increase in inter-sample variance, but as peaks and troughs in the pollen types were sustained for centuries, it was unlikely that similar variation was missed in the period from c. 400 BC–AD 500. Thus, we do not consider the merging of cores to provide a problem for interpretation.

The data presented here are far more detailed than those originally published for this site (Bush and Colinvaux, 1988). The earlier study covered 7000 years and provided a history of land use spanning that period. Sampling within the last 2690 years, which is the period of highest pre-Columbian human influence across the basin, was sparse in the original record, that is, 13 samples and no ¹⁴C dates in that period, compared with 105 samples and 11 ¹⁴C dates in the present study. Nevertheless, major trends, such as peaks of Poaceae that preceded a relatively recent rise in Alchornea from c. 5 to 20% after AD 1440 were clearly common to both datasets, as was the high abundance of Cecropia, and relatively steady abundances of total forest taxa. The new results, however, show a more consistent representation of maize pollen compared with the 1983 core. The 1983 record reported maize pollen and phytoliths from a minority of samples (n = 3) in the last 2690-years, whereas in this study they are found in the majority (n = 86). The reason for this apparent increase in maize occurrence was that techniques for maize extraction (Whitney et al., 2012) and identification (Holst et al., 2007) improved in the interim, and that through extended counts we now detect maize pollen presence at a much higher level than was possible in the 1980s. The re-analysis of Lake Ayauchⁱ sediments highlights the continuity of maize cultivation at the site from the onset of the new record until AD 1700, illustrating nearly continuous human impacts and disturbances. The amount of total forest pollen, however, remained near 80% in both analyses, suggesting that the landscape was a shifting mosaic of fire, cultivation, and forest opening amidst a matrix of closed canopy trees. The data presented here add to a growing body of evidence that humans were not clear-cutting but rather maintaining some closed canopy forests (Bush et al., 2021).

Evidence of occupation and land use

The strongest evidence of human activity in the Ayauchⁱ sedimentary record comes from the regular occurrences of maize pollen and charcoal (Figures 3 and 5). Maize pollen was found in most samples prior to c. AD 1420, but the samples where maize pollen was identified in the first 300 grains encountered, as opposed to being found in the extended counts were mostly between c. AD 750 and 1370. We infer that this period had the most extensive maize cultivation. Maize pollen are wind dispersed like all grass pollen but are large in size (>72 µm) and do not travel far from the source plant (Lane et al., 2010). The maize fields documented in the pollen record were likely located within a kilometer of Lake Ayauchⁱ. Maize phytoliths found within terrestrial soil cores collected around the lake (McMichael et al., 2012c) support this inference.

Figure 5.

A comparison of the land use and fire history of Lake Ayauchⁱ, Ecuador, with the isotopic records from Lake Pumacocha (Bird et al., 2011) and Palestina cave (Apaéstegui et al., 2014) showing from the top to the bottom: (a) δ¹⁸O isotopic record from Palestina cave (light blue line; Apaéstegui et al., 2014); (b) δ¹⁸O isotopic record from Lake Pumacocha (blue line; Bird et al., 2011); (c) percentage Cecropia pollen (dark yellow line); (d) percentage sum forest pollen (anemophilous + non-anemophilous forest pollen excluding Cecropia pollen; gray line), percentage local forest pollen (non-anemophilous forest pollen; green line), and percentage grass pollen (red line); (e) the proportion of samples with Zea mays pollen per century (yellow circles); (f) lake sediment charcoal influx (mm²/cm³yr; gray bars); and (g) macroscopic (>180 µm) lake sediment charcoal fragments (mm²/cm³; black bars). Lake sediment charcoal sample effort (black plus signs) and soil charcoal fragment ages (black open stars; McMichael et al., 2012c) are also shown. The timing of the medieval climate anomaly (MCA) is shown by the vertical pink bar.

One of the distinct changes in the record is a sudden increase in arboreal taxa, for example, Cecropia, Alchornea, Trema, and Combretaceae-Melastomataceae, that took place between c. AD 950 and 1000. The recovery of forest between c. AD 1260 and 1420 is evident in both the wind-pollinated taxa such as Cecropia, Moraceae-Urticaceae, and Alchornea and local forest (non-anemophilous) taxa (Figures 3 and 5). There was still some decline of forest cover notable in the record after c. AD 1260, but this transition marks a lasting change from samples containing c. 20% local forest pollen to c. 35–40%. Although some cultivation was still taking place from c. AD 1700 to 1800, the scale was much reduced compared with earlier times.

Charcoal occurred in many samples prior to the 1800s but did not show strong coherence with either maize occurrence or dry events inferred from the isotopic climate records (Figure 5). Charcoal at Ayauchⁱ was associated with human activity, evidenced by the lack of charcoal in the last 200 years. The lack of alignment of charcoal with climate records (Apaéstegui et al., 2014; Bird et al., 2011) and evidence of land use in the Ayauchⁱ pollen record suggested that people did not need prolonged (multi-annual) dry events to burn the forest. While soil charcoal aligned with peaks of Poaceae pollen (Figure 5), and some soil profiles contained maize phytoliths (McMichael et al., 2012c), the lack of a relationship between sediment charcoal and maize pollen occurrence suggested either these proxies were operating at different scales, or that people were burning for other reasons than growing maize.

Although the abundance of Poaceae pollen is not correlated with finding maize pollen, it likely is indicative of canopy openings. Poaceae abundances greater than 1–2% in an aseasonal Amazonian rainforest sequence indicate canopy openings and are often correlated with human disturbance (Bush, 2002; Whitney et al., 2019). Because the basin of Ayauchⁱ has a slightly raised lip (Colinvaux et al., 1985) and no inflowing streams, it does not receive overland flow. Consequently, the pollen inputs to Ayauchⁱ are almost exclusively wind-borne rather than water-borne. Indeed, as the lake edge is quite steep, there is no fringing marsh, and because the lake itself is highly oligotrophic, Poaceae pollen would most likely come from terrestrial rather than aquatic sources. If the modern proportion of grass pollen is taken as a baseline, the 2% corresponds to c. 21% clearance within 1 km and 7% cleared within 10 km of the lake. The high abundances of Poaceae at 8–20% prior to c. AD 550 highlights a period when the landscape was more open and deforested than it is today.

Most samples in this sedimentary record contained some charcoal fragments (>180 µm), but prior to c. AD 550 the fire events appear to be discrete, with decades to centuries separating them. The resulting peaks of sediment charcoal mostly aligned with dated macro charcoal fragments (>500 µm) recovered from soil samples collected within 1 km of the lake (McMichael et al., 2012c). During this period, spikes of sedimentary charcoal and soil charcoal appear to align with an abundance of Poaceae pollen (Figures 3 and 5), indicating that cleared land was being invaded by grasses. Forest regrowth was repeatedly disrupted by fires (Figure 5), especially after AD 550, suggesting that fire was actively used for land management at the site. The decline of forest elements following sediment charcoal peaks, however, never exceeded 10–25%, indicating that the majority of fires died out as they spread to mature forests (Cochrane and Laurance, 2008; Nepstad et al., 2001). Slash-and-burn cultivation was likely used at the site, at various intensities through time, as a tool for clearing land for crop growth, releasing nutrients into the soil, suppressing forest regrowth, and reducing vegetation density (Arroyo-Kalin, 2012; Guariguata and Ostertag, 2001; van Vliet et al., 2013). It should be noted that the soil charcoal (Figure 5) were sampled to represent the older fires in each soil core, and so the lack of soil charcoal ages in the upper portion of the record does not indicate a lack of local fire. Rather these soil charcoal ages confirm that fires were set close to the lake. These episodes, however, do not appear to be related to climatic change as they seem equally likely to occur during phases indicated to be wet or dry according to the isotopic records (Figure 5). The period from c. AD 550 to 1420, exhibited the most consistent burning, the highest density of samples containing maize, and largest oscillations in Cecropia abundance. Somewhat surprisingly, Poaceae was less abundant than in the preceding zone, perhaps associated with increased weeding at the site, but showed great variability in abundance as it alternated with peaks of Cecropia. The per-century-proportion of samples that contained maize pollen was generally 75% or greater until c. 1420 when its presence became more erratic, possibly linked to the Inka invasion of Ecuador AD 1463 (Cabello Balboa, 1945). The proportion of samples yielding maize pollen per century declined to its nadir (20–30%) in the 1700s and 1800s.

Evidence of forest regrowth

Total arboreal taxa includes both anemophilous (wind-pollinated) and non-anemophilous taxa (Figures 3 and 5). An exclusion of the potentially long-distance transported anemophilous taxa, for example, Cecropia, Alchornea, Moraceae-Urticaceae, and Celtis, is likely to provide a better index of local forest (non-anemophilous forest) cover at the site (Figures 3 and 5). Compared with modern, the local forest pollen showed some depletion throughout the early portion of the record, but forest recovery between c. AD 500 and 700 resulted in higher-than-modern local forest pollen representation. Further, multi-centennial cycles of forest clearance and recovery followed, with the nadir of local forest representation occurring within the MCA. Each of the forest recovery cycles was paralleled by a surge in Cecropia pollen. The local peak of forest cover occurred at c. AD 1450 and thereafter there was a gradual loss of these local forest indicators, but the anemophilous forest elements, such as Trema and Alchornea, showed marked increases in abundance (Figure 3). Post-1450, the intensity of cultivation decreased, indicated by, for example, fewer peaks of charcoal and rarer findings of maize pollen. This landscape appeared to have less open ground than before but an increasing cover of moderately disturbed forest, indicated by, for example, increased abundances of Cecropia and decreased abundances of herbaceous and grass taxa. The shift in vegetation and cultivation suggested less intense land use, but enough disturbance to maintain much of the area as a relatively early successional forest. Note that this transition, though extending into the period of European influence, was initiated at least a century before European arrival, possibly during the time of the Inka invasion of Ecuador in AD 1463 (Cabello Balboa, 1945).

Although disease may have spread a few years ahead of European invasion, that is, in the late 1520s in the high Andes (Black, 1992), the majority of enslavement and social dislocation occurred between c. AD 1550 and 1700 (Smith, 1994). In the 1600s, there is further evidence of reduced human alteration of the landscape with a decline in maize occurrence and a steady increase in forest elements, but this does not bear the signature of sudden abandonment. The site does appear to have been used minimally between in the 1700s and early 1900s, noted by the decreased presence of maize pollen and lack of charcoal fragments at the site (Figure 5), prior to a wave of renewed use in the uppermost sample associated with the modern Shuar community.

How much climatic or environmental factors played a role in Amazonian human history is actively debated. While dry times appear to have favored increased abundances of Cecropia at Ayauchⁱ (Figure 5), this pioneer genus appeared to rise and fall independently of charcoal, grass, or maize. Today, Cecropia is common in the disturbed lands near the site and has also been observed growing in secondary forest areas and along the edge of Lake Ayauchⁱ. Cecropia was likely taking advantage of fallowed fields, but the largest of all Cecropia peaks at c. AD 1100 does not align to a major phase of forest regrowth, but rather with a peak of drought during the MCA. Such extreme droughts in recent years have initiated substantial canopy mortality and gap formation (Brando et al., 2014; Laurance and Williamson, 2001; Meir et al., 2009), which would favor gap-filling species, such as Cecropia. The increase in Cecropia from <50% to 200% in modern time, however, was probably promoted by the renewed land use of the site by the Shuar Nation community. The transition in cultivational practices inferred at c. AD 550 may align with the onset of increased precipitation variance; as new data are gathered from other western Amazonian sites it would certainly be worth considering whether such a linkage occurred. Additionally, there may have been population replacements bringing novel cultivational techniques. The location of Ayauchⁱ is at the extreme range of penetration of the Arawakan diaspora (Heckenberger, 2002) and the timing correlates (within ¹⁴C error) with the change in cultivation practices c. AD 550 and with the AD 400–600 to 800 abandonment of the Upano valley (Rostain, 2012). Further archeological exploration of this site could test that hypothesis. The apparent forest regrowth, reduction in burning, and less frequent occurrence of maize after c. AD 1420 could relate to a wetter period that lasted from c. AD 1400 to 1650 (Figure 5), which fits with a broader Amazonian change in land use (Bush et al., 2021) that does not appear to have an underlying single climatic driver.

Times of cultural change in pre-Columbian Ecuador

Amazonian archeology indicates that the period from c. 1000 BC to AD 1000 was one of increasing population (Arroyo-Kalin and Riris, 2021), societal complexity (Heckenberger, 2009), reliance on cultivation (Piperno, 2006, 2011), and population migrations (Heckenberger, 2002; Iriarte et al., 2017). From our study, two potential cultural transitions stand out. The first was an apparent change in land use about AD 550, the second a sudden increase in forest cover that began c. AD 1260.

To find strong differences in apparent land use practices surrounding maize cultivation could indicate the choices of a single group changing through time. Alternatively, the changes could represent new occupants of the site. In the Upano valley near Macas (Figure 1), Ecuador, about 90 km from Ayauchⁱ, mound-builders of the Upano culture occupied the site for at least 900 years but abandoned the site c. AD 400–600 (Rostain, 2012; Sales, 2022). When the Huapula culture re-occupied the site c. AD 800 they were culturally and behaviorally different from the earlier inhabitants. Similarly, in the Napo an undated transition occurred from the Yasuní to the Tivacundo by c. AD 550 (Evans and Meggers, 1968), and is broadly coincident with the expansion of Arawakan peoples (Hill and Santos-Granero, 2010; Walker and Ribeiro, 2011) who brought new languages and technologies to Amazonia. On present evidence we cannot determine if the changes seen at c. AD 550 at our site were caused by a cultural replacement, but future interdisciplinary studies combining paleoecology with archeological excavations can explore this hypothesis.

The second transition was at c. AD 1260–1420 when the record of maize cultivation became more sporadic and forest taxa increase in abundance. This period was also marked by a shift to wetter conditions that lasted into the 1600s (Figure 5). This distinctly pre-European increase in forest pollen is similar to that observed in the mountains of northern Peru (Åkesson et al., 2020; Bush et al., 2015b), and in other regions of Amazonia (Bush et al., 2021). European influence is evident in the collapse of maize cultivation in the 1800s, but we find no evidence of a major forest expansion that could relate to the Orbis spike of AD 1620 (Koch et al., 2019; Lewis and Maslin, 2015; sensu Ruddiman, 2005).

Conclusions

A fossil pollen and charcoal record spanning the last 2690 years from Lake Ayauchⁱ confirmed a long history of human occupation, fire activity, forest modification, and maize cultivation. Fire in the record was associated with human activity and did not show an alignment with dry events inferred from regional climate records. Although the near-continuous presence of maize pollen suggests continual occupation of the site, it does not mean that the number of occupants was constant. A change in cultivational practices at c. AD 550 may be part of broader social re-organizations in both the Amazon and the Andes, and may also have been influenced by a shift to a more variable precipitation regime. Times of reduced disturbance may indicate reduced population density. During a time marked by a regional shift to wetter conditions, forest cover began to recover c. AD 1260 and cultivation was reduced by AD 1420. This onset of forest recovery clearly precedes European arrival by c. 400 years and is a more significant change in ecological trajectory than any experienced between AD 1420 and the near-modern period. The depopulation associated with European arrival is likely reflected in the lack of maize recorded from the site in the 1800s, but the overall effect on the pollen record of this probable abandonment was minor. European arrival may have caused the final abandonment of Lake Ayauchⁱ, but it was not responsible for initiating major ecological changes at this site.

Supplemental Material

sj-pdf-1-hol-10.1177_09596836231151833 – Supplemental material for Late-Holocene maize cultivation, fire, and forest change at Lake Ayauchi, Amazonian Ecuador

Supplemental material, sj-pdf-1-hol-10.1177_09596836231151833 for Late-Holocene maize cultivation, fire, and forest change at Lake Ayauchi, Amazonian Ecuador by Christine M Åkesson, Crystal NH McMichael, Susana León-Yánez and Mark B Bush in The Holocene

Footnotes

Acknowledgements

We thank the people and government of Ecuador, especially the Ayauchⁱ community, for their permission to access their lands. We also thank Eliane Bakker for assistance with charcoal analysis and Gabriela López for assistance with permissions and logistics. Bryan Valencia, Geovanny Byron Llivisaca, and Jacob Schiferl are thanked for assistance in conducting fieldwork.

Authors’ contribution

CMÅ conducted fieldwork, laboratory and statistical analyses, and wrote the manuscript. CNHM and MBB conceived the project, conducted fieldwork, and contributed to data analysis, interpretation, and writing the manuscript. SLY provided logistical support and contributed to writing the manuscript.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Ministerio del Ambiente del Ecuador (permit number 08-2017-IC-FLO/DNB/MA). Funding for this research was provided by the National Aeronautics and Space Administration (grant no. NNX14AD31G), the National Science Foundation (grant no. NSFEAR-1303831), and the Netherlands Organisation for Scientific Research (grant no. ALWOP.322).

Data archiving

The datasets generated in this study will be available through the NEOTOMA Paleoecological Database () upon publication.

ORCID iDs

Christine M Åkesson

Crystal NH McMichael

Mark B Bush

Supplemental material

Supplemental material for this article is available online.

References

Åkesson

Matthews-Bird

Bitting

, et al. (2020) 2,100 years of human adaptation to climate change in the high Andes. Nature Ecology & Evolution 4: 66–74.

Åkesson

McMichael

CNH

Raczka

, et al. (2021) Long-term ecological legacies in western Amazonia. Journal of Ecology 109(1): 432–446.

Apaéstegui

Cruz

Sifeddine

, et al. (2014) Hydroclimate variability of the northwestern Amazon basin near the Andean foothills of Peru related to the South American Monsoon system during the last 1600 years. Climate of the Past 10(6): 1967–1981.

Aragão

LEO

Malhi

Roman-Cuesta

, et al. (2007) Spatial patterns and fire response of recent Amazonian droughts. Geophysical Research Letters 34(7): L07701.

Arroyo-Kalin

(2012) Slash-burn-and-churn: Landscape history and crop cultivation in pre-Columbian Amazonia. Quaternary International 249: 4–18.

Arroyo-Kalin

Riris

(2021) Did pre-Columbian populations of the Amazonian biome reach carrying capacity during the late-Holocene? Philosophical Transactions of the Royal Society B 376(1816): 20190715.

Balée

(1989) The culture of Amazonian forests. In: Posey

Balée

(eds) Resource management in Amazonia: indigenous and folk strategies. New York: New York Botanical Garden, pp.1–21.

Binford

Kolata

Brenner

, et al. (1997) Climate variation and the rise and fall of an Andean civilization. Quaternary Research 47: 235–248.

Bird

Abbott

Rodbell

, et al. (2011) Holocene tropical South American hydroclimate revealed from a decadally resolved lake sediment δ¹⁸O record. Earth and Planetary Science Letters 310(3–4): 192–202.

10.

Birks

HJB

Gordon

(1985) Numerical Methods in Quaternary Pollen Analysis. London: Academic Press.

11.

Blaauw

Christen

(2011) Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Analysis 6: 457–474.

12.

Black

(1992) Why did they die? Science 258: 1739–1740.

13.

Brando

Balch

Nepstad

, et al. (2014) Abrupt increases in Amazonian tree mortality due to drought–fire interactions. Proceedings of the National Academy of Sciences 111(17): 6347–6352.

14.

Brugger

Gobet

van Leeuwen

JFN

, et al. (2016) Long-term man–environment interactions in the Bolivian Amazon: 8000 years of vegetation dynamics. Quaternary Science Reviews 132: 114–128.

15.

Bush

(2002) On the interpretation of fossil Poaceae pollen in the humid lowland neotropics. Palaeogeography Palaeoclimatology Palaeoecology 177: 5–17.

16.

Bush

McMichael

Piperno

, et al. (2015a) Anthropogenic influence on Amazonian forests in pre-history: An ecological perspective. Journal of Biogeography 42(12): 2277–2288.

17.

Bush

Mosblech

NAS

Church

(2015b) Climate change and the agricultural history of a mid-elevation Andean montane forest. The Holocene 25(9): 1522–1532.

18.

Bush

Colinvaux

(1988) A 7000-year pollen record from the Amazon lowlands, Ecuador. Vegetatio 76: 141–154.

19.

Bush

Nascimento

Åkesson

, et al. (2021) Widespread reforestation before European influence on Amazonia. Science 372(6541): 484–487.

20.

Bush

Piperno

Colinvaux

(1989) A 6000 year history of Amazonian maize cultivation. Nature 340: 303–305.

21.

Bush

Silman

Listopad

CMCS

(2007) A regional study of Holocene climate change and human occupation in Peruvian Amazonia. Journal of Biogeography 34(8): 1342–1356.

22.

Bush

Weng

(2007) Introducing a new (freeware) tool for palynology. Journal of Biogeography 34: 377–380.

23.

Cabello Balboa

(1945) Obras. Quito: Editorial Ecuatoriana.

24.

Chazdon

(2003) Tropical forest recovery: Legacies of human impact and natural disturbances. Perspectives in Plant Ecology Evolution and Systematics 6(1-2): 51–71.

25.

Cochrane

Laurance

(2008) Synergisms among fire, land use, and climate change in the Amazon. AMBIO: A Journal of the Human Environment 37(7): 522–527.

26.

Colinvaux

de Oliveira

Moreno

PJE

(1999) Amazon Pollen Manual and Atlas. Amsterdam: Harwood Academic Publishers.

27.

Colinvaux

(2007) Amazon Expeditions: My Quest for the Ice-Age Equator. New Haven and London: Yale University Press.

28.

Colinvaux

Miller

Liu

K-B

, et al. (1985) Discovery of permanent Amazon lakes and hydraulic disturbance in the upper Amazon basin. Nature 313: 42–45.

29.

Cook

(1998) Born to die: disease and New World conquest, 1492-1650. Cambridge: Cambridge University Press.

30.

de Paula Moraes

Neves

(2012) O ano 1000: Adensamento populacional, interação e conflito na Amazônia Central. Amazônica-Revista de Antropologia 4(1): 122–148.

31.

Dobyns

(1966) Estimating Aboriginal American population I: An appraisal of techniques with a new Hemispheric estimate. Current Anthropology 7(4): 395–416.

32.

Erickson

(2006) The domesticated landscapes of the Bolivian Amazon. In: Balée

Erickson

(eds) Time and Complexity in Historical Ecology. New York: Columbia University Press, pp.235–278.

33.

Evans

Meggers

(1968) Archaeological Investigations on the Rio Napo, Eastern Ecuador. Washington, DC: Smithsonian Institution Press.

34.

Faegri

Iversen

(1989) Textbook of Pollen Analysis. Chichester: Wiley.

35.

Faegri

van der Pijl

(1979) Principles of Pollination Ecology, 3rd edn. Oxford: Pergamon Press.

36.

Glaser

Birk

(2012) State of the scientific knowledge on properties and genesis of Anthropogenic Dark Earths in central Amazonia (terra preta de índio). Geochimica et Cosmochimica Acta 82: 39–51.

37.

Goldberg

Mychajliw

Hadly

(2016) Post-invasion demography of prehistoric humans in South America. Nature 532(7598): 232–235.

38.

Guariguata

Ostertag

(2001) Neotropical secondary forest succession: changes in structural and functional characteristics. Forest Ecology and Management 148(1–3): 185–206.

39.

Heckenberger

(2002) Rethinking the Arawakan diaspora: Hierarchy, regionality, and the Amazonian formative. In: Granero

Hill

(eds) Comparative Arawakan Histories: Rethinking Language Family and Culture Area in Amazonia. Champaign, IL: University of Illinois Press, pp.99–122.

40.

Heckenberger

(2009) Lost cities of the Amazon. Scientific American 301(4): 64–71.

41.

Hill

Santos-Granero

(2010) Comparative Arawakan Histories: Rethinking Language Family and Culture Area in Amazonia. Urbana and Chicago: University of Illinois Press.

42.

Hill

(1979) DECORANA - A FORTRAN Program for Detrended Correspondence Analysis and Reciprocal Averaging. Ecology and Systematics. New York: Cornell University.

43.

Holst

Moreno

Piperno

(2007) Identification of teosinte, maize, and tripsacum in Mesoamerica by using pollen, starch grains, and phytoliths. Proceedings of the National Academy of Sciences 104(45): 17608–17613.

44.

Iriarte

Smith

Gregorio

Souza

, et al. (2017) Out of Amazonia: Late-Holocene climate change and the Tupi–Guarani trans-continental expansion. The Holocene 27(7): 967–975.

45.

Juggins

(2019) Rioja: Analysis of Quaternary Science Data. R Package Version (0.9-21). Newcastle upon Tyne: University of Newcastle.

46.

Koch

Brierley

Maslin

, et al. (2019) Earth system impacts of the European arrival and great dying in the Americas after 1492. Quaternary Science Reviews 207: 13–36.

47.

Lane

Cummings

Clark

(2010) Maize pollen deposition in modern lake sediments: A case study from Northeastern Wisconsin. Review of Palaeobotany and Palynology 159: 177–187.

48.

Laurance

Williamson

(2001) Positive feedbacks among forest fragmentation, drought, and climate change in the Amazon. Conservation Biology 15: 1529–1535.

49.

Levis

Costa

Bongers

, et al. (2017) Persistent effects of pre-Columbian plant domestication on Amazonian forest composition. Science 355(6328): 925–931.

50.

Lewis

Maslin

(2015) Defining the Anthropocene. Nature 519(7542): 171–180.

51.

Liu

Colinvaux

(1988) A 5200-year history of Amazon Rain Forest. Journal of Biogeography 15: 231–248.

52.

Martin

Newton

Bullock

(2013) Carbon pools recover more quickly than plant biodiversity in tropical secondary forests. Proceedings of The Royal Society B Biological Sciences 280(1773): 20132236.

53.

McMichael

Correa-Metrio

Bush

(2012a) Pre-Columbian fire regimes in lowland tropical rainforests of southeastern Peru. Palaeogeography Palaeoclimatology Palaeoecology 342-343: 73–83.

54.

McMichael

Piperno

Bush

, et al. (2012b) Sparse pre-Columbian human habitation in western Amazonia. Science 336(6087): 1429–1431.

55.

McMichael

Bush

Piperno

, et al. (2012c) Spatial and temporal scales of pre-Columbian disturbance associated with western Amazonian lakes. The Holocene 22(2): 131–141.

56.

McMichael

CNH

Bush

(2019) Spatiotemporal patterns of pre-Columbian people in Amazonia. Quaternary Research 92(1): 53–69.

57.

McMichael

CNH

Matthews-Bird

Farfan-Rios

, et al. (2017) Ancient human disturbances may be skewing our understanding of Amazonian forests. Proceedings of the National Academy of Sciences 114(3): 522–527.

58.

Meir

Brando

Nepstad

, et al. (2009) The effects of drought on Amazonian rain forests. In: Keller

Bustamante

Gash

, et al. (eds) Amazonia and Global Change. Washington, DC: American Geophysical Union, p.565.

59.

Nepstad

Carvalho

Barros

, et al. (2001) Road paving, fire regime feedbacks, and the future of Amazon forests. Forest Ecology and Management 154(3): 395–407.

60.

Nepstad

Lefebvre

Lopes

Silva

, et al. (2004) Amazon drought and its implications for forest flammability and tree growth: A basin-wide analysis. Global Change Biology 10(5): 704–717.

61.

Neves

Petersen

Bartone

, et al. (2004) The timing of terra preta formation in the central Amazon: Archaeological data from three sites. In: Glaser

Woods

(eds) Amazonian Dark Earths: Explorations in Space and Time. Berlin, Heidelberg: Springer, pp.125–134.

62.

Oksanen

Blanchet

Kindt

, et al. (2013) Package ‘vegan’. Community ecology package, version 2(9): 1–295.

63.

Pärssinen

Schaan

Ranzi

(2009) Pre-Columbian geometric earthworks in the upper Purús: A complex society in western Amazonia. Antiquity 83(322): 1084–1095.

64.

Piperno

(2006) The origins of plant cultivation and domestication in the Neotropics. In: Kennett

Winterhalder

(eds) Foraging Theory and the Transition to Agriculture. Berkeley, CA: University of California Press, pp.137–166.

65.

Piperno

(2011) The origins of plant cultivation and domestication in the New World tropics: Patterns, process, and new developments. Current Anthropology 52(S4): S453–S470.

66.

Piperno

Pearsall

(1998) The Origins of Agriculture in the Lowland Neotropics. San Diego, CA: Academic Press.

67.

Rasband

(1997–2016) ImageJ. Bethesda, MD: U. S. National Institutes of Health.

68.

R Development Core Team (2021) R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. Available at: https://www.R-project.org/ (accessed 7 March 2022).

69.

Reimer

Austin

WEN

Bard

, et al. (2020) The IntCal20 Northern Hemisphere radiocarbon age calibration curve (0–55 cal kBP). Radiocarbon 62(4): 725–757.

70.

Roosevelt

(1991) Moundbuilders of the Amazon: Geophysical Archaeology on Marajó Island, Brazil. San Diego, CA: Academic Press.

71.

Rostain

(2012) Between Sierra and selva: Landscape transformations in upper Ecuadorian Amazonia. Quaternary International 249: 31–42.

72.

Roubik

Moreno

PJE

(1991) Pollen and spores of Barro Colorado Island. Missouri Botanical Garden: Monographs in Systematic Botany 36.

73.

Ruddiman

(2005) How did humans first alter global climate? Scientific American 292(3): 34–41.

74.

Salazar

(2008) Pre-Columbian mound complexes in the Upano river valley, lowland Ecuador. In: Silverman

Isbell

(eds) The Handbook of South American Archaeology. New York, NY: Springer, pp.263–278.

75.

Sales

(2022) The impacts of Holocene climate and pre-Columbian humans on Neotropical Andean forests. PHD Dissertation, Florida Institute of Technology. Department of Ocean Engineering and Marine Sciences.

76.

Smith

(1994) Explorers of the Amazon. Chicago, IL: University of Chicago Press.

77.

Stockmarr

(1971) Tablets with spores used in absolute pollen analysis. Pollen et Spore 13(4): 615–621.

78.

Thompson

Mosley-Thompson

Davis

, et al. (2013) Annually resolved ice core records of tropical climate variability over the past ~1800 years. Science 340(6135): 945–950.

79.

van Vliet

Adams

Vieira

ICG

, et al. (2013) “Slash and burn” and “shifting” cultivation systems in forest agriculture frontiers from the Brazilian Amazon. Society & Natural Resources 26(12): 1454–1467.

80.

Vuille

Bradley

Werner

, et al. (2003) 20th century climate change in the tropical Andes: Observations and model results. Climatic Change 59(1-2): 75–99.

81.

Walker

Ribeiro

(2011) Bayesian phylogeography of the Arawak expansion in lowland South America. Proceedings of The Royal Society B Biological Sciences 278(17–18): 2562–2567.

82.

Watling

Iriarte

Mayle

, et al. (2017) Impact of pre-Columbian “geoglyph” builders on Amazonian forests. Proceedings of the National Academy of Sciences 114(8): 1868–1873.

83.

Weng

Bush

Athens

(2002) Two histories of climate change and hydrarch succession in Ecuadorian Amazonia. Review of Palynology and Paleobotany 120: 73–90.

84.

Whitlock

Larsen

(2002) Charcoal as a fire proxy. In: Smol

Birks

Last

, et al. (eds) Tracking Environmental Change Using Lake Sediments. Terrestrial, Algal, and Siliceous Indicators. Dordrecht: Springer, pp.75–97.

85.

Whitney

Rushton

Carson

, et al. (2012) An improved methodology for the recovery of Zea mays and other large crop pollen, with implications for environmental archaeology in the Neotropics. The Holocene 22(10): 1087–1096.

86.

Whitney

Smallman

Mitchard

, et al. (2019) Constraining pollen-based estimates of forest cover in the Amazon: A simulation approach. The Holocene 29(2): 262–270.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB

Late-Holocene maize cultivation,fire,and forest change at Lake Ayauch i,Amazonian Ecuador

Abstract

Keywords

Introduction

Study area

Materials and methods

Results

Lake Ayauchi zone description

Zone 1 (680 BC–AD 530; 170–28 cm 2007-core; 83–82 cm 2016-core)

Zone 2 (AD 530–1420; 28–0 cm 2007-core; 82–32 cm 2016-core)

Zone 3 (AD 1420 to AD 2016; 32–0 cm 2016-core)

Discussion

Evidence of occupation and land use

Evidence of forest regrowth

Times of cultural change in pre-Columbian Ecuador

Conclusions

Supplemental Material

sj-pdf-1-hol-10.1177_09596836231151833 – Supplemental material for Late-Holocene maize cultivation, fire, and forest change at Lake Ayauchi, Amazonian Ecuador

Footnotes

Acknowledgements

Authors’ contribution

Funding

Data archiving

ORCID iDs

Supplemental material

References

Supplementary Material

Lake Ayauchⁱ zone description