Mid-Late Holocene Vegetational History and Land-use Dynamics in County Monaghan, Northeastern Ireland–the Palynological Record of Lough Muckno Carlos Chique1,*, Karen Molloy1, and Aaron P. Potito1 Abstract - We conducted high-resolution palynological analysis on a sediment core obtained from Lough Muckno, County Monaghan, Ireland. The results presented represent the first paleoecological account of Mid-Late Holocene vegetational change and land-use dynamics in the study region. Human activity and agriculture is first recorded during the Early Neolithic (ca. 3870–3500 B.C.). After a period of undiscernible human activity of ~900 years, farming resumes during the Early Bronze Age (ca. 2600 B.C.). Henceforth, human presence on the landscape is constant with fluctuating levels of intensity. During the Bronze Age, anthropogenic activity is most pronounced during ca. 2000–1750 B.C. and ca. 1500–1300 B.C. followed by a phase of reduced intensity in the Late Bronze Age (ca. 1000–650 B.C.). Farming activity increases during the Iron Age and is disrupted with the onset of a period of rapid woodland regeneration from ca. 200 B.C. to A.D. 200. During the prehistorical period agriculture has a strong focus on pastoral grazing with a limited arable component. An upsurge in agricultural activity is recorded in the historical period from ca. A.D. 400 in which a mixed agricultural economy placing more emphasis on cereal-crop cultivation is adopted. Arable farming attains its maximum levels ca. A.D. 990–1140. Evidence of farming disruptions in the pollen record may reflect of a period of local “conflict” during the Viking Age/Medieval period (ca. A.D. 800–1190). We explore the characterizing features of the pollen assemblage of this large lake system and its use in reconstructing past cultural landscape change. 1Palaeoenvironmental Research Unit, School of Geography and Archaeology, National University of Ireland–Galway, Galway, Ireland. *Corresponding author - c.chique1@nuigalway.ie. Introduction Pollen-analytical investigations have played a major role in the development of vegetation history and landscape reconstruction in Ireland. A significant number of pollen diagrams are now available from the island (Mitchell et al. 2013), with several focusing on Mid-Late Holocene environments (viz. Walker et al. 2012), which has allowed for attempts at identifying changes in vegetation structure and land-use dynamics during different cultural periods (e.g., Hall 2000, O’Connell and Molloy 2001, Plunkett 2009). However, a limitation of the Irish palynological record lies in its uneven geographical distribution precluding a uniform representation of landscape change across the island. The Irish Pollen Site Database (IPOL 2013) categorizes County Monaghan as one of the least studied areas in Ireland from a palynological perspective (cf. Mitchell et al. 2013). To date, a total number of 3 pollen datasets with only 1 radiocarbon (14C) date are available from the county (Coope et al. 1979, Mitchell 1951, Vaughan et al. 2004). These investigations mostly relate to pre-/Early Holocene environments (pre-6250 B.C.), with no profiles available spanning the Late Holocene. While the number of pollen diagrams in Ireland as a whole is high, there are still few high-resolution investigations focusing on different cultural periods (e.g., Ghilardi and O’Connell 2013; Molloy and O’Connell 1991, 2004). This study aims to address a clear gap in the state of knowledge and expand the spatial extent of the Irish palynological record by implementing high-resolution pollen analysis on a sediment core obtained from Lough Muckno, County Monaghan. The objective is to provide a meticulous account of Mid–Late Holocene landscape change with no precedent in an area still unexplored in palynological terms. We attempted to critically evaluate human impacts on vegetation structure and describe the development of the local agricultural component through pollen-inferred land-use dynamics. The regional setting, inclusive of Monaghan and adjoining counties, has traditionally drawn limited archaeological interest, leading to an underrepresentation of archaeological records relative to other parts of Ireland (Ó Drisceoil et al. 2014:13). Accordingly, the current investigation also aims to provide a palynological background of relevance for past and future regional archaeological research. We took an innovative approach to Irish palynological reconstructions with the application of high pollen counts in the sedimentary record of a system with a large surface area. Such an endeavor is expected to produce an account of vegetational change at a considerable spatial scale (local and regional) due to the projected pollen source area of the lake, a function of its large deposition basin (Sugita 2007; Hellman et al. 2008). Pollen reconstructions from large lakes 2017 Journal of the North Atlantic No. 32:1–24 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 2 rarely feature in Irish palynology, which is invariably concerned with change at the local level. The location of the study site within the inter-drumlin belt provides insights into vegetation dynamics in a landscape with distinctive topographical and edaphological properties (Aalen et al. 1997), which can be extrapolated to palynological reconstructions in similar settings. Taking into account the novel nature of the results presented, we emphasize drawing comparisons with pertinent palynological data (Fig. 1; Table 1). Pollen diagrams from 4 sites located in close proximity to L. Muckno that comprise the regional extent of palynological data are considered in detail (Table 4). We expected the collation of regional pollen data to (i) allow for the comparison and potential integration of the high-resolution profile of L. Muckno with neighboring (more local) pollen records, and (ii) explore land-use dynamics at a finer spatial scale in this part of Ireland while drawing contrast with patterns inferred elsewhere in the island. Local and Regional Setting Lough Muckno (Figs. 1, 2) is a large inter-drumlin lake (3.57 km2; 86 m a.s.l.) composed of 3 main basins joined by narrow channels, located next to the modern town of Castleblayney. The lake has an average depth of 5.4 m with an estimated shoreline length of 26.3 km, and is fed by 6 main fluvial inputs draining an extensive catchment area (109.08 km2). The geology of the area is composed of Silurian quartzite overlain by poorly drained glacial till soils. The post-glacial landscape provides an undulating topographical setting manifested in frequent oval-shaped features (drumlins) often merging with lake margins (Fig. 2b). Figure 2a provides an account of archeological features within the area under consideration. Historical records (i.e., post-A.D. 400) predominate with ringforts (raths), enclosed farmsteads characteristic of Early Historic open landscapes (ca. A.D. 600– 850; Kerr 2009), being the most prevalent (n =1 13). The remains of the Early Christian (ca. A.D. 700) Muckno Monastery are located on the northern shore of the lake. A structure (fortification) northwest of the coring location was suggested to be part of an expansion of local monastic activity following ca. A.D. 1100 (Ó Dufaigh 2011). Distinctive prehistoric features include a Bronze Age lake settlement (Cullyhanna) northeast of the lake consisting of an oak Figure 1. Map of Ireland illustrating the location of Lough Muckno and the study area (□) relative to pollen sites referred to in the text. The approximate extent of the inter-drumlin belt is shown with a grey background. The location of relevant archaeological features outside the range of Figure 2a are shown with a “▲” symbol and include Iron Age sites Aghareagh West and The Dorsey. Detailed information for each pollen site is given in Table 1. Sites considered part of the “regional” record are denoted with an “×” symbol: (1a) Redbog II; (1b) Essexford Lough; (2) Whiterath Bog; (3) Loughnashade. Other pollen sites are shown with a “●” symbol and include the following: (4) Garry Bog; (5) Sluggan Bog; (6) Caheraphuca Lough; (7a) Caherkine Lough; (7b) Mooghaun Lough; (8a) Barrees; (8b) Loch Beag; (9) Mount Gabriel; (10) Glen West Bog; (11) Abbeyknockmoy Bog; (12) An Loch Mór; (13) Ballinphuil Bog; (14) Church Lough; (15) Lough Sheeauns; (16) Clonfert Bog; (17) Clonenagh Bog; (18) Moyreen Bog; (19) Corlea Bog; (20) Belderrig; (21) Céide Fields; (22) Owenduff Bog; (23) Emlagh Bog; (24) Mongan Bog; (25) Kilbegly; (26) Lough Cooney; (27) Lough Dargan; (28) Loughmeenaghan; (29) Templevanny Lough; (30) Derryville Bog; (31) Ballynagilly Bog; (32) Lough Catherine; (33) Claraghmore Bog; (34a) Killimady Lough; (34b) Weir’s Lough; (35) Scragh Bog. Journal of the North Atlantic 3 2017 No. 32 C. Chique, K. Molloy, and A. Potito palisade and hut structure (Hodges 1958), with onsite radiocarbon returns of 3475 ± 75 BP (ca. 1800 B.C.) and 3305 ± 50 BP (ca. 1580 B.C.) (Smith et al. 1973). A cluster of 5 fulachta fia (burnt-mounds) were uncovered on a former island southwest of the coring location (Fig. 2b). Fulacht fia are predominantly Bronze Age features, with 14C evidence suggesting 2 main periods of proliferation in Ireland during ca. 1700–1300 B.C. and ca. 1000–800 B.C. (Ó Néill 2003). Methods Fieldwork We used a Usinger piston corer (see Mingram et al. 2007) to recover 2 parallel sediment cores (LM-I, II), both ~9.5 m long from the deepest part of the western basin of the lake at a water depth of 15 m (Fig. 2b). The data presented pertains to depths 284–744 cm of core LM-II. Table 1. Details of pollen sites shown in Figure 1. The deposit type indicates the nature of the profile subject to palynological analysis. A indicates raised bog and B is used for blanket bog. Map Deposit Coordinates Dating Site label type (lat/long) County methods Reference Redbog II 1a BogA 53°58'28", 6°37'22" Louth 14C, tephra Weir (1995) Essexford 1b Lake 53°58'6", 6°38'44" Louth Cross-matching Weir (1995) Whiterath Bog 2 Lake 53°55'37", 6°25'24" Louth Cross-matching Weir (1995) Loughnashade 3 Lake 54°21'00", 6°41'26" Armagh 14C, cross-matching Weir (1993) Garry Bog 4 BogA 55°13'27", 6°28'13" Antrim Tephra Plunkett (2009) Sluggan Bog 5 BogA 54°44'12", 6°14'47" Antrim Tephra Plunkett (2009) Caheraphuca 6 Lake 52°55'50", 8°55'1" Clare 14C Molloy and O’Connell (2011, 2012) Caherkine 7a Lake 52°47'13", 8°51'21" Clare 14C O’Connell et al. (2001) Mooghaun 7b Lake 52°47'4", 8°52'45" Clare 14C Molloy (2005) Barrees 8a BogB 51°42'30", 9°55'10" Cork 14C Overland and O’Connell (2008) Beag 8b Lake 51°42'4", 9°55'32" Cork 14C Overland and O’Connell (2008) Mount Gabriel 9 BogB 51°33'18", 9°33'48" Cork 14C Mighall et al. (2008) Glen West Bog 10 BogA 54°24'50", 8°2'22" Fermanagh Tephra Plunkett (2009) Abbeyknockmoy Bog 11 BogA 53°26'35", 8°45'22" Galway 14C, tephra Lomas-Clarke and Barber (2004) An Loch Mór 12 Lake 53°3'28", 9°30'36" Galway 14C, tephra; varves Molloy and O’Connell (2004, 2007) Ballinphuill Bog 13 BogA 53°16'16", 8°30'5" Galway 14C Molloy and O’Connell (2016) Church 14 Lake 53°37'25", 10°12'45" Galway 14C O’Connell and Ní Ghráinne (1994) Sheeauns 15 Lake 53°33'21", 10° 4'39" Galway 14C Molloy and O’Connell (1987, 1991) Clonfert Bog 16 BogA 53°15'8", 8°3'12" Galway Tephra Hall (2005) Clonenagh Bog 17 BogA 52°59'57", 7°25'56" Laois Tephra Hall (2005) Moyreen Bog 18 BogB 52°31'58", 9°10'54" Limerick Tephra Plunkett (2009) Corlea Bog 19 BogA 53°37'18", 7°52'2" Longford 14C, dendrochronology Caseldine and Hatton (1996) Belderrig 20 BogB 54°18'8", 9°34'21" Mayo 14C Verrill and Tipping (2010) Céide Fields 21 BogB 54°18'15", 9°27'32" Mayo 14C Molloy and O’Connell (1995) Oweduff Bog 22 BogB 53°59'50", 9°39'43" Mayo Tephra Plunkett (2009) Emlagh Bog 23 BogB 53°47'7", 6°45'12" Meath 14C, tephra Newman et al. (2007) Mongan Bog 24 BogA 53°19'59", 7°55'52" Offaly Tephra Hall (2005) Kilbegly 25 Mire 53°17'42", 8°12'10" Roscommon 14C Overland and O’Connell (2011) Cooney 26 Lake 54°12'0", 8°32'14" Sligo 14C O’Connell et al. (2014) Dargan 27 Lake 54°12'10", 8°25'13" Sligo 14C Ghilardi and O’Connell (2013) Loughmeenaghan 28 Lake 54° 5'35", 8°23'31" Sligo 14C Stolze et al. (2012) Templevanny 29 Lake 54°2'5", 8°24'18" Sligo 14C Stolze et al. (2013) Derryville Bog 30 BogA 52°46'1", 7°40'53" Tipperary Tephra Hall (2005) Ballynagilly Bog 31 BogB 54°40'46", 6°51'28" Tyrone 14C Pilcher and Smith (1979) Catherine 32 Lake 54°41'54", 7°26'15" Tyrone 14C Hirons and Edwards (1986) Claraghmore Bog 33 BogA 54°37'53", 7°26'48" Tyrone 14C wiggle-matching, Plunkett (2009) tephra Killimady 34a Lake 54°29'12", 6°47'54" Tyrone 14C Hirons (1984), Hirons and Edwards (1986) Weir's 34b Lake 54°29'12", 6°47'54" Tyrone 14C Hirons (1984), Hirons and Edwards (1986) Scragh Bog 35 Lake 53°35'36", 7°20'59" Westmeath Cross-matching O’Connell (1980, 1991) 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 4 Figure 2. (A) Lough Muckno at the catchment scale including simplified hydrology and archaeological setting. The legend indicates the time-period classification assigned to each archaeological record in the area. The annotation is used to differentiate among archaeological records including specific sites mentioned within the text. Unclassified Neolithic tombs are shown unannotated with a “▲” symbol. Individual ringforts (including univallate, bivallate, and cashel-type) are also illustrated unannotated using the “●” symbol. The large multivallate Lisleitrim Fort is shown separately. No further discrimination is made among the remainder of records with an “Uncertain (prehistoric)” classification, which are shown unannotated with a “*” symbol. (B) Western basin of L. Muckno showing coring location (depth in m), topography (a.s.l. in m), and local archaeological records. Journal of the North Atlantic 5 2017 No. 32 C. Chique, K. Molloy, and A. Potito Stratigraphy, chronology, and loss-on-ignition Visual description and use of photography were our primary means of documenting the stratigraphy of the sediment core. To attach a chronology to the profile, 8 sediment samples were selected for AMS 14C dating (between 160-708 cm) based on significant changes in the pollen record. Sediment samples consisted of 2-cm–thick slices sieved through a 125-μm–mesh sieve. We examined the material retained using a Leica MZ125 stereomicroscope, and submitted terrestrial macrofossils (fruits, plant remains and charcoal) for AMS 14C dating. The upper 2-m section of the sediment core was dated using 210Pb methods, which are presented elsewhere (Chique et al., in review). Sediment samples (2 cm3) were taken at the same depths as pollen samples (i.e., every 4 cm from 284 cm to 744 cm) to calculate loss-on-ignition (LOI %). Samples were dried for 24 hours at 105 °C and ashed for 4 hours at 550 °C (Heiri et al. 2001). Pollen analysis The 116 samples used for pollen analysis in this study consisted of 1 cm3 of sediment taken every 4 cm from 284 cm to 744 cm in depth. Sample preparation followed standard methods implemented at the Palaeoenvironmental Research Unit (PRU). Lycopodium clavatum spore tablets (2 tablets, containing a total of 41,696 spores) supplied by the Department of Geology, University of Lund (batch no. 1031) were added to the samples prior to laboratory processing to facilitate calculation of pollen concentration. The samples were deflocculated with KOH, sieved through a 100-μm–mesh sieve, treated with 60% HF solution followed by acetolysis, sieved through a 5-μm–mesh in an ultra-sonicator in order to remove remaining particles/debris, mounted in glycerol, and examined for pollen using a Leica DM 4000-B microscope fitted with x10 oculars with a x40 objective for routine counting and phase-contrast x63 and x100 (oil immersion) for detailed analysis of pollen grains. Pollen and spore identification and nomenclature follow Moore et al. (1991) and Beug (2004). We regularly consulted the pollen reference collection at the PRU. Cereal-type pollen were identified following the guidelines provided by Beug (2004), but with a minimum-size criterion of 40 μm instead of 37 μm for inclusion of grains into the cereal-type category, and differentiation of Secale-type (rye) pollen. Based on the length of the longest axis, we grouped cereal-type pollen into 40–44 μm, 45–49 μm, and >50 μm categories (excluding Secale-type pollen). Identification of Isoetaceae microspores is based on size–frequency curves by Birks (1973). Routine counting observations and measurement (longest diameter parallel to furrow excluding perine) of 200 microspores divided equally among 10 samples suggest dominance of Isoetes echinospora (average spore size = 27.4 μm). Non-pollen palynomorphs (NPP) counted include algal remains, Pinus stomata and (micro-) charcoal (>37 μm). A minimum of 1000 total terrestrial pollen (TTP) and spores were counted in each sample. Pollen percentage and concentration calculations as well as diagrams were generated using CountPol ver 3.3 (obtained from I. Feeser, University of Kiel, Kiel, Germany). Percentage values are based on TTP, with the pollen sum (PS) excluding pollen and spores from aquatic taxa and NPPs. Main pollen categories comprise arboreal pollen (AP) and non-arboreal pollen (NAP). We divided the latter into distinct ecological groups including tall shrubs, grassland/pastoral (NAPp), arable/disturbed, ferns, and bog/heath. Results Stratigraphy The section of the sediment core under consideration (284–744 cm) is uniformly composed of postglacial fine dark brown lake mud intercalated with lighter-colored banding at random intervals. 14C dating and age–depth modelling Available 14C dates are given in Table 2. The uppermost dates LMII-1 and LMII-2 were not incorporated into the age–depth model on the grounds of 14C reversal. However, the age–depth model is constrained at the top (190 cm) by the lowermost date (-88 BP ± 130) of the 210Pb chronology in the upper 2 m of LM-II, which enables an age-depth chronology to be established in parts of the profile where 14C dates reversed. The age–depth model (Fig. 3) was established using Bacon v. 2.2 Bayesian age–depth modelling software (Blaauw and Christen 2011) utilizing IntCal13 (Reimer et al. 2013) embedded on R as an interface. The output of the dating model provides an age determination for every centimeter depth of the sediment core. Prior values assigned to the Markov-chain Monte Carlo iterations are as follows: accumulation shape = 1.5, accumulation mean = 10 years/cm, memory strength = 4, memory mean = 0.7. Modification of the pre-set prior valueaccumulation mean was based on the software estimate recommendation. The 14C date at 708 cm (ca. 3920 BC) represents the lowermost age determination from LM-II included in the age–depth model. Considering the lack of palynological indicators of human activity in that section of the core, we derived the chronology for samples below this depth (i.e., 712–744 cm) by assuming a constant sedimentation rate. In addition to Figure 3, supplemental data is 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 6 provided to highlight the age range associated with age–depth modelling corresponding to the historical period and calendar year annotated events (see Supplementary Table 1, available online at http:// www.eaglehill.us/JONAonline/suppl-files/J061316- Chique-s1, and, for BioOne subscribers, at http:// dx.doi.org/10.1656/J061316.s1). Pollen and additional data Establishment of pollen assemblage zones (PAZs) is based on changes in percentage/concentration pollen data while zone age determination is based on the output of Bacon age–depth modelling. The percentage contribution towards TTP of various pollen ecological groups is presented in a composite diagram in Figure 4a. Individual percentage pollen curves are presented in Figures 4b and 5a, and selected concentration curves are given in Figure 5b. LOI (%) and dry-mass sediment accumulation rate are shown in Figures 5c and 5d, respectively. Calculation of sediment accumulation rate is based on the working age–depth model inclusive of (accepted) 14C dates and the bottom-most 210Pb date (Fig. 3). The percentage contribution (averaged) of terrestrial pollen ecological groups in each PAZ is presented in a pie chart format in Figure 6. Cereal-type and Secale-type pollen statistics are shown in Figure 7. Table 3 includes a summary of the most relevant palynological features of each PAZ while providing information on contemporary cultural/historical periods. The division of the prehistoric period (ca. 4400 B.C.–A.D. 400) is based on Waddell (2010) Table 2. 14C dates from core LM-II. Sample depths are given according to upper depth of section taken (maximum 2 cm thick). Age range is quoted in calibrated years B.C./A.D. (negative values indica te B.C.). Sample Depth Age range 14C lab. code ID (cm) 14C (BP) (1σ, 68.3%) (2σ, 95.4%) Material description Comments UBA-26860 LMII-1 160 756 ± 31 1247–1281 1221–1285 Charcoal, wood fragments, moss Reversal, rejected stems, Betula fruit UBA-26861 LMII-2 301 1494 ± 36 542–611 532–489 Charcoal, wood fragments, moss stems Reversal, rejected UBA-30293 LMII-7 320 1108 ± 34 940–980 869–1017 Leaf and wood fragments, unidentified Accepted bud scale, Alnus fruit UBA-29462 LMII-5a 372 1472 ± 30 565–621 545–643 Leaf and wood fragments, Betula fruit Accepted UBA-29463 LMII-6 420 2348 ± 46 -489 to -379 -545 to -357 Wood fragments, Betula fruit and bract Accepted UBA-26862 LMII-3 492 3119 ± 32 -1431 to -1385 -1451 to -1287 Wood fragments, moss stems, charred Accepted plant remains UBA-26863 LMII-4 628 4237 ± 37 -2904 to -2867 -916 to -2851 Wood fragments, Betula remains Accepted UBA-26864 LMII-5 708 5174 ± 40 -3998 to -3957 -4051 to -3936 Charcoal, wood fragments, moss Accepted stems, Betula bud scale Figure 3. Age–depth model based on Bayesian modelling of 14C dates in conjunction with 210Pb data from core LM-II. 14C results (uncalibrated) are also shown including the rejected date at 301 cm denoted by *. Black bands indicate probability distribution for calibrated 14C dates. The red curve follows the “best” model based on the weighted mean age for each depth, with parallel grey lines indicating the 95% confidence interval. Journal of the North Atlantic 7 2017 No. 32 C. Chique, K. Molloy, and A. Potito Figure 4. Percentage pollen diagram plotted to depth with matching age-scale in calibrated years B.C./A.D. (rounded to the nearest 10). Ecological groups are differentiated using color coding. Zones/subzones are indicated on the right hand side. (A) Composite percentage pollen diagram. (B) Selected pollen percentage curves and pollen sum (x1000 grains). Scales used are indicated at the base of the individual curves, with normal curves presented at x1 and hatched curves exaggerated (x2, x5, x10). Dots indicate very low values. The following abbreviations are used for rare occurrences. Tall Shrubs: P = Prunus; S = Sambucus; C = Crataegus. NAPp: V = Valeriana officinalis. Arable/Disturbed: H = Cannabis/Humulus-type. NPPs: PS = Pinus stomata. 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 8 Figure 5. [Caption on following page.] Journal of the North Atlantic 9 2017 No. 32 C. Chique, K. Molloy, and A. Potito Figure 5 (previous page). (A) Percentage pollen curves for taxa with low but relevant occurrence (conventions follow those in Fig. 4) (B) Selected pollen concentration curves including total pollen, AP, and NAP. (C) Calculated loss-on-ignition (%). (D) Dry-mass sediment-accumulation rate. Note that values from depths 712–744 cm are calculated based on the assumption of a constant sedimentation rate. 14C dates (uncalibrated BP) are given according to sample depth towards the right hand side. The rejected date (LMII-2) is shown with a hollow symbol and accepted dates with black fill symbol. Core sections (each 2 m long) are given in the left hand side of the diagram. Cultural/historical periods are shown on the right hand side and are color coded according to the overall degree of inferred human activity: blue = undiscernible or limited, light green = low–moderate, orange = moderate–high, and red = high. Table 3. Summary of palynological features of profile LM-II. The Chronology (14C) column is based on the results of age-depth modelling from core LM-II. The Period/Age Range column provides individual cultural/historical periods and their corresponding timeframe. All dates are given in calibrated years B.C./A.D. A indicates the period is based on Waddell (2010), Bfollows O’Brien (2012) and CO’Sullivan (1998). Spectra PAZs (cm) Chronology (14C) Period/age range Main features and subzones 9 332–284 A.D. 800 –1190 Viking Age (A.D. 800–1100)C – Initial recovery of AP followed by reductions; further Medieval (A.D. 1100–1350)C expansion of NAP; Cereal-type curve disrupted. Subzone 9c: Cereal-type curve disrupted. Subzone 9b: Corylus, Quercus, Betula, and Fraxinus decline; Poaceae and P. lanceolata expand; Cereal-type curve expands. Subzone 9a: Increase in Corylus and Fraxinus; Slight contraction of P. lanceolata; Cereal-type curve disrupted. 8 384–336 A.D. 220–800 Late Iron AgeA –Early Historic Reduction in AP; NAP expansion. (A.D. 400–800)C Subzone 8b: Corylus, Quercus, and Alnus decline; Poaceae, P. lanceolata, and Pteridium expand; Increase in Pediastrum, I. echinospora, and micro-charcoal; Cereal-type curve expands. Subzone 8a: Corylus, Betula, and Fraxinus decline; Poaceae and P. lanceolata expand. 7 408–388 220 B.C.–A.D. 220 Late Iron AgeA AP recovers; low NAP. 6 436–412 650–220 B.C. Early Iron Age (600 B.C.– Decline in Quercus, Ulmus, and Corylus; Poaceae, P. lanceolata, A.D. 400)A and Pteridium expand; Increase in Pediastrum, I. echinospora, and micro-charcoal. 5 500–440 1490–650 B.C. Middle Bronze Age (1600–1100 Initial decline in AP and NAP expansion with gradual reversal of B.C.)A –Late Bronze Age (1000– trend. 650 B.C.)A Subzone 5c: AP recovers; P. lanceolata declines further. Subzone 5b: Reductions in Fraxinus and Alnus; NAP contract and stabilize; Increase in Pteridium. Subzone 5a: Quercus, Corylus, and Fraxinus decline; Poaceae, P. lanceolata,and Pteridium expand; Increase in Pediastrum and micro-charcoal. 4 600–504 2590–1490 B.C. Early Bronze Age (2500–1600 AP fluctuates along with NAP expansion in three consecutive B.C.)A; inclusive of Chalcolithic phases. (2500–2000 B.C.)B Subzone 4d: AP tends to stabilize; NAP decrease from preceding subzone and remains constant. Subzone 4c: Decrease in Quercus, Ulmus, Fraxinus; Poaceae and P. lanceolata increase. Subzone 4b: AP fluctuates; Poaceae and P. lanceolata expand Subzone 4a: Decline in Quercus, Ulmus, Corylus, Fraxinus and Alnus; Poaceae and P. lanceolata expand; Cereal-type pollen forms curve. 3 672–604 3490–2590 B.C. Middle - Late Neolithic (3100– High AP; NAP very low. 2500 B.C.)A Subzone 3c: Third Elm Decline. Subzone 3b: Second Elm Decline; Fraxinus begins to expand. Subzone 3a: High Corylus, Quercus, Alnus, and Ulmus recovers. 2 708–676 3970–3490 B.C. Early Neolithic (4000–3600 B.C.)A AP declines; NAP increases for the first time. –Middle Neolithic (3600–3100 Subzone 2c: Recovery in AP; NAP declines. B.C.)A Subzone 2b: Decline in Quercus and Betula with Alnus expanding at the end of subzone; Poaceae and P. lanceoloata expand . Subzone 2a: Elm Decline. 1 744–712 4400–3970 B.C. Mesolithic (8000–4000 B.C.)A High AP including Corylus, Ulmus, Quercus; NAP very low. 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 10 General considerations Based on theoretical guidelines (Sugita 2007), the pollen record of a large system such as L. Muckno can potentially reflect the averaged vegetation composition of an area of ~104–105 km2. These estimates relate to pollen from tree taxa (AP) characterized by enhanced dispersal in contrast to pollen from low-growing plants (NAP), which derive from a much more localized area (Hellman et al. 2009). NAPp and especially cereal-type pollen are expected to be reflective of the vegetation structure on mineral soils near the lake basin (Broström et al. 2005, Sjögren et al. 2015). In these terms, the sedimentary record of L. Muckno is expected to incorporate regional to local pollen (Hellman et al. 2008). Stream-bound pollen from the catchment area of the lake may be an important contributor towards TTP (Sugita 2007). In order to facilitate comparisons with other sites, we recalibrated 14C dates from other pollen profiles (Table 1) using the IntCal13 curve and Calib ver. 7.1. All dates are referred to in calibrated years B.C./A.D. Some of the regional pollen diagrams (Tables 1, 4) lack independent dating, with the chronology of Essexford Lough and Whiterath Bog based on crosscorrelation of pollen events with the 14C chronology established at Redbog II (Weir 1995). At Loughnashade, a combination of local 14C evidence and crosscorrelation of events (based on data by Hirons 1984) were used to derive an age–depth chronology (Weir 1993). Pollen assemblage zones (PAZs) The main palynological features summarised in Table 3 are described in more detail as follows. PAZ-1 (744–712 cm; ca. 4400–3970 B.C.). AP accounts for 97% of TTP. Corylus is the main contributor (41%–24%) and O’Brien (2012), and the historic period (ca. A.D. 400–1190) follows O’Sullivan (1998). We also classified the data using informal periods in order to better illustrate chronological trends in palynological data, with the Iron Age (600 B.C.–A.D. 400) subdivided into Early Iron Age (ca. 650–220 B.C.) and Late Iron Age (ca. 220 B.C.–220 A.D.). Table 4 integrates main palynological features and landuse dynamics inferred from regional pollen profiles (Weir 1993, 1995). Figure 6. Pie charts showing the percentage contribution of terrestrial pollen ecological groups in Zones/Subzones 1–9. Journal of the North Atlantic 11 2017 No. 32 C. Chique, K. Molloy, and A. Potito Figure 7. Cereal-type pollen statistics. Mean size (bar column), maximum/ minimum size (dashed line), and standard deviation (solid line) for cereal-type grains is given for each zone and, in the case of Zones 8–9, for each subzone. Bar data labels xx/yy denote the amount of cereal- and Secale-type recorded in each zone, with xx indicating cereal-type and yy Secale-type. Zones 1 and 3 have no cereal-type records. The dotted vertical line indicates the minimum-size criterion (40 μm) used for inclusion of pollen grains into the cereal-type category. Table 4. Summary of land-use dynamics and palynological features inferred from regional profiles (Weir 1993, 1995). Arrow symbols (↓↑) are used to indicate an increase (↑) or decrease (↓) in human activity at each site. The first column provides corresponding PAZs established at Lough Muckno along with cultural/historical period and age-d epth chronology (14C) to facilitate comparisons. PAZ, period, and chronology (14C) Redbog II Essexford Lough Loughnashade Whiterath Bog 1. Pre-Neolithic landscape Elm Decline: 3980 B.C. N/A N/A N/A (pre-3970 B.C.) 2. Early-Middle Neolithic Onset of activity at 3800 B.C. N/A N/A N/A (3970–3490 B.C.) 3. Middle-Late Neolithic Subdued activity, Elm Subdued activity, N/A Subdued activity (3490–2590 B.C.) Decline: 3230 and 2800 B.C Elm Decline: 2900 B.C. 4. Early Bronze Age ↑ pastoral farming from ↑ 2300 B.C. ↑ 1900 B.C. ↑ 2300 B.C. (2590–1490 B.C.) 2300 B.C. Arable component from 1600 B.C. 5. Middle–Late Bronze ↑ 1600–1400 B.C. and ↑ 1200 B.C. and ↑ 1400–1300 B.C. ↑ 1300 B.C. Age (1490–650 B.C.) 1000–800 B.C. 1000 B.C. particularly arable farming ↓ 1000–600 B.C. 6. Early Iron Age Subdued activity ↑ 600–200 B.C. ↑ 600–200 B.C. ↑ 400–200 B.C (650–220 B.C.) 7. Late Iron Age Subdued activity and ↓ 200 B.C.–A.D. 25 ↓ 200 B.C.–A.D. 200 ↓ 200 B.C–A.D. 25 (220 B.C.–A.D. 220) ↑ A.D. 25–540 ↑ A.D. 25–540 ↑ A.D. 25–200 8. Late Iron Age –Early ↑ A.D. 25–540 ↑ A.D.25–540 and 700–800 ↑ A.D. 200–500 and N/A Historic period ↓ and woodland separated by ↓ and 600–900 separated by (A.D. 220–800) regeneration A.D. 540–850 woodland regeneration, ↓ and woodland Secale introduction A.D. regeneration, Secale 500 introduction A.D. 400 9. Viking Age and ↑ 850–1000 Tendency towards ↑ with Tendency towards ↑ N/A Medieval (A.D. 800–1190) ↓ in arable farming with ↓ in arable A.D. 900–1000 farming A.D. 900–1050 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 12 and P. lanceolata increase to 6.7% and 1.6%, respectively, in conjunction with additional NAPp. Subzone 4c (552–528 cm; ca. 2060–1750 B.C.): AP representation falls to 83%. This is accounted for by a decrease in values of Quercus (19%), Ulmus (1%) and Fraxinus (1%). Corylus (40%) and Betula (7%) show higher representation. NAPp increases again, primarily Poaceae (7%) and P. lanceolata (3%). Subzone 4d (524–504 cm; ca. 1750–1490 B.C.): AP curves remain relatively stable. Poaceae and P. lanceolata values contract to 4% and 0.7%, respectively. The cereal-type curve is briefly interrupted. PAZ-5 (500–440 cm; ca. 1490–650 B.C.) Subzone 5a (500–488 cm; ca. 1490–1300 B.C.): AP falls from 90% to 69% with reductions in Quercus (24% to 16%), Fraxinus (6% to 2%), and Corylus (40% to 26%). Conversely, an increase in Betula (6% to 12%) and Alnus (13% to 17%) values is recorded. The expansion of NAP (31% of TTP) is primarily accounted for by an increase in Poaceae (3.7% to 15%) and P. lanceolata (1% to 4%). The cereal-type pollen curve averages 0.2% for most of PAZ-5. Pediastrum, I. echinospora, and microcharcoal values increase in PAZ-5, in particular in Subzone 5a. Subzone 5b (484–468 cm; ca. 1300–1030 B.C.): AP oscillates from 86% to 79% with reductions in Fraxinus (2%), Betula (6%), and Alnus (20%), although Corylus values increase (36%). Taxus forms a curve continuing uninterrupted until PAZ-9b. Poaceae and P. lanceolata curves contract to ~7.8% and ~1.3%, respectively. Subzone 5c (464–440 cm; ca. 1030–650 B.C.): Most AP curves remain stable. Values of Poaceae (~6.5 %), P. lanceolata (~0.6%), and most NAPp indicators tend to decrease. A disruption in the cerealtype curve is recorded. PAZ-6 (436 cm–412 cm; ca. 650–220 B.C.) AP falls to 71% with a decrease in Quercus (17% to 12%) Fraxinus (~7% to 1%), Corylus (36% to 25%), and Alnus (21% to 15%). Betula values increase from 6% to 10%. Higher NAPp values are recorded with Poaceae increasing from 8% to 14% and P. lanceolata peaking at 4%. Cereal-type pollen is continuously recorded. Levels of Pediastrum, Botryococcus, I. echinospora, and micro-charcoal increase. PAZ-7 (408–388 cm; ca. 220 B.C.–A.D. 220) AP representation increases from 71% to 91%, accounted for by Corylus (25% to 39%), Ulmus (1% to followed by Quercus (27%–21%), Alnus (19%–15), Ulmus (17%–11%), Betula (4%–8%), and Pinus (4%–1%). NAP values are low. Aquatic indicators and micro-charcoal are at low levels. PAZ-2 (708–676 cm; ca. 3970–3490 B.C.). Subzone 2a (708–704 cm; ca. 3970–3870 B.C.): AP dominates (96% of TTP) with Ulmus values falling from 13% to 5%. Micro-charcoal is at low levels through Zone 2. Subzone 2b (700–684 cm; ca. 3870–3600 B.C.): AP representation falls to 79% with Ulmus declining to 1%. Quercus and Betula values decrease from 29% to 13% and 8% to 4%, respectively. Alnus values increase to 40%. NAP attains 21% with an increase in Poaceae (11%) and Plantago lanceolata (4%). Two cereal-type pollen grains are recorded at 700 cm. Pediastrum, Botryococcus, and Isoetes echinospora show increased representation. Subzone 2c (680–676 cm; ca. 3600–3490 B.C.): AP representation recovers to 97%; see in particular Corylus (41%), Alnus (25%), Quercus (21%), and Ulmus (5%). NAPp falls to 2% with Poaceae and P. lanceolata curves contracting to 1.2% and 0.6%, respectively. PAZ-3 (672–604 cm; ca. 3490–2590 B.C.) Subzone 3a (672–656 cm: ca. 3490–3220 B.C.): AP dominates (98%–95%), and NAP values remain low throughout Zone 3. Alnus values decline from 29% to 16%. A continuous Fraxinus curve (~0.4%) is established. Subzone 3b (652–636 cm; ca. 3220–2960 B.C.): Ulmus values decline from 11% to 5% before recovering to 12%. Fraxinus representation increases to ~1.1%. The first records of Taxus baccata are observed at 644 cm. Subzone 3c (632–604 cm; ca. 2960–2590 B.C.): Ulmus values decline from 12% to 1% before recovering to 6%. Fraxinus continues to increase, attaining 5%. PAZ-4 (600–504 cm; ca. 2590–1490 B.C.) Subzone 4a (600–592 cm; ca. 2590–2460 B.C.): AP declines from 96% to 89% with reduced representation of Quercus (38% to 31%), Ulmus (6% to 2%), Fraxinus (5% to 1%), and Alnus (21% to 13%). Poaceae and P. lanceolata values peak to 5.3% and 2.7%, respectively. Cereal-type pollen is recorded in Zone 4 at low levels (~0.1%). Small increases in micro-charcoal follow closely the expansion of NAP in all subzones. Subzone 4b (588–556 cm; ca. 2460–2060 B.C.): Slight fluctuations in most AP curves are recorded. Alnus peaks at 26% before falling to ~19%. Poaceae Journal of the North Atlantic 13 2017 No. 32 C. Chique, K. Molloy, and A. Potito activity, this section mainly follows cultural/historical periods (Table 3). These are in turn contained within 3 broad headings: (i) pre-Neolithic landscape, (ii) prehistoric vegetation and land-use dynamics, and (iii) historic vegetation and land-use dynamics. Pre-Neolithic landscape Pre-elm decline woodland environment (PAZ-1) (pre-3970 B.C.). The landscape is dominated by mixed deciduous woodland; oak is the main tall-canopy tree with lesser amounts of elm, pine, birch, and alder likely to be prevalent on wetter soils. Hazel represents the primary understory component. Ferns are common within these woodlands (Polypodium vulgare and Dryopteris-filix-mas-type curve; Figs.. 4b, 5a). The very low NAPp values (~0.9%) suggest open habitats were not present at this time, and apart from the initiation of an Ilex curve at ca. 4080 B.C. (Fig. 4b), there is no evidence of pre-Neolithic woodland instability (viz. O’Connell and Molloy 2001). While the presence of micro-charcoal can potentially be linked to pre-Neolithic human activity (Mighall et al. 2008, Warren et al. 2014), in view of the lack of anthropogenic pollen indicators in this section of the profile, a stronger argument can be made for charcoal originating from natural sources. Elm-decline (PAZ-2a) (ca. 3970–3870 B.C.). The Mid-Holocene Elm Decline, a ubiquitous feature of Irish pollen diagrams dated to the Early Neolithic, is recorded in PAZ-2a with a decrease in Ulmus values from 13% to 5% at ca. 3920 B.C. The Elm Decline as recorded here does not appear to be linked with human activity and precedes a Neolithic Landnam by ~50 years. A similar pattern is reported in different parts of Ireland (cf. Ghilardi and O’Connell 2013, Molloy and O’Connell 1987, Stolze et al. 2013). Considering the lack of palynological indicators of human activity at the onset of the event, it is assumed that anthropogenic factors have limited relevance in the sudden decline of elm. The hypothesis of an epidemic disease caused by a fungal pathogen differentially affecting elm (Clark and Edwards 2004) seems valid in this context. However, it is also possible that factors not discernible from the pollen record were involved in the process. The spread of disease by earlier and inconspicuous human disturbance of the woodland environment (Warren et al. 2014), perhaps undiscernible considering the expected low level of activity and localized nature of NAP, is a distinct possibility. The onset of the decline corresponds with records from other locations, with a post-4000 B.C. start registered within the region (Table 4) and at several other sites (O’Connell and Molloy 2001, Parker et al. 2002, Whitehouse et al. 2014). 3%), Betula (8% to 15%), Fraxinus (4% to 7%), and Taxus (0.6% to 1.4%). Quercus values fall from 17% to 10% and then stabilize at ~15%, a trend supported by concentration data. Alnus values oscillate between 13% and 20%. NAP is low with Poaceae and P. lanceolata curves contracting from 13% to 5% and 3% to 0.1%, respectively. The cereal-type curve (~0.1%) is disrupted. A decrease in the representation of aquatic indicators and micro-charcoal is recorded. PAZ-8 (384–336 cm; ca. A.D. 220–800) Subzone 8a (384–376 cm; ca. A.D. 220–440): AP falls from 90% to 74% with declining Corylus (37% to 28%), Betula (12% to 5%), and Fraxinus (5% to 1%) values. NAPp shows higher representation with Poaceae and P. lanceolata values increasing from 5.7% to 12.1% and 1% to 2%, respectively. The cereal-type curve peaks briefly (0.5%) and then contracts (~0.1%). Subzone 8b (372–336 cm; ca. A.D. 440–800): AP declines to 61%. Corylus (21%), Quercus (9%) and Alnus (15%) are primarily affected. Poaceae and P. lanceolata peak to 19% and 3%, respectively. Cereal- type values increase (1.3%), with Secale pollen recorded sporadically. Pediastrum, Botryococcus, I. echinospora, Litorella uniflora, and micro-charcoal show increased representation. PAZ-9 (332–284; ca. A.D. 800–1190) Subzone 9a (332–316 cm; ca. A.D. 800–960): AP attains 79% with Corylus values increasing from 25% to 32% and Fraxinus from 1% to 4%. Poaceae values remain at ~10%, P. lanceolata decreases (2.3% to 1.1%), and additional NAPp is recorded. Cereal-type values fluctuate from 0.6% to 0.1%, and Secale pollen eventually becomes frequent. Microcharcoal levels are high in Zone 9. Subzone 9b (312–292 cm; ca. A.D. 960–1140): AP declines to 59% with Corylus, Quercus, Betula, and Fraxinus reaching lows of 23%, 8%, 5%, and 0.7%, respectively. Poaceae and P. lanceolata values increase to 20% and 3.2%, respectively. NAPp attains 28% of TTP. Cereal-type records peak at 1.1% concurrent with Secale values of 0.6%. Aquatic indicators (Pediastrum, Botryococcus, I. echinospora, and L. uniflora) have a high representation. Subzone 9c (288–284 cm; ca. A.D. 1140–1190): AP and NAP values remain stable, with the exception of the cereal-type curve contracting to 0.4%. Discussion In order to facilitate a detailed chronological account of vegetation change paralleled with human 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 14 tern is reported in other inter-drumlin sites (Hirons and Edwards 1986) and at a Neolithic field system located in a steep river valley in North Mayo (Verril and Tipping 2010). Eventual woodland recovery along with a decline in NAP in PAZ-2c reflects the abandonment of land from ca. 3600–3490 B.C. The pollen profile in closest proximity, Redbog II (Table 4), also indicates commencement of Neolithic activity following ca. 3800 B.C., with a comparable chronology reported in other locations (Caseldine and Hatton 1996, O’Connell 1991, O’Connell and Molloy 2001, Stolze et al. 2013). The inferred period of concentrated human activity at L. Muckno (~270 years) may be considered brief (cf. ~700 years of main Landnam phase; Ghilardi and O’Connell 2013), and in terms of intensity, either modest (cf. Molloy and O’Connell 1987, 1995) or substantial (cf. Mighall et al. 2008, Molloy and O’Connell 2012, Weir 1995), a factor that highlights geographical variability in land-use patterns. The palynological evidence from L. Muckno and elsewhere (Caseldine and Hatton 1996; Ghilardi and O’Connell 2013; Molloy and O’Connell 1987, 2016; O’Connell 1991; O’Connell and Molloy 2001; Pilcher and Smith 1979; Stolze et al. 2012, 2013) suggests agricultural activity is predominantly pastoral, with discreet evidence of arable farming during the earliest stages of the Early Neolithic. This interpretation is also supported by archaeobotanical evidence (McClatchie et al. 2014). In terms of archaeological records, the prevalence of court tombs in the area (n = 9), considered to be predominantly Early Neolithic features (ca. 3700–3570 B.C.; Schulting et al. 2012), corresponds well to the period of concentrated Neolithic activity inferred from pollen data. Middle–Late Neolithic (PAZ-3) (ca. 3490 - 2590 B.C.). High AP representation (~97% of TPP) in PAZ-3 indicates the presence of a fully closed woodland environment primarily dominated by hazel, oak, and elm with conditions similar to the pre-Elm Decline environment (PAZ-1) (Figs. 4a, 6). Open habitats are no longer present locally, and human presence in the landscape is not discernible from palynological evidence. These observations are comparable to patterns at both a regional and wider scale (Table 4; Caseldine and Hatton 1996; Molloy and O’Connell 1991, 2016; O’Connell and Molloy 2001; Pilcher and Smith 1979; Stolze et al. 2013). This period of reduced activity has been tentatively attributed to a decline in population (O’Connell and Molloy 2001) or climatic deterioration (Whitehouse et al. 2014) at the Irish scale. The consecutive reductions in Ulmus values in PAZ-3 appear to be additional Elm Decline events. Vegetation and land-use dynamics in the prehistorical period Early–Middle Neolithic (PAZs 2b-2c) (ca. 3970–3490 B.C.). The first definite evidence of human activity is recorded in PAZ-2 spanning a period of ~380 years, with activity concentrated during ca. 3870–3600 B.C. (PAZ-2b), followed by a phase of declining activity ca. 3600–3490 B.C. (PAZ-2c). Clearance of tall-canopy oak and birch is construed from reduction in Quercus and Betula values. Further reductions in Ulmus values may reflect a combination of disease and woodland clearance affecting elm. The stability of Pinus values (Fig. 4b) indicates that pine is not affected by clearings, with pine pollen possibly deriving from the wider region. The brief peak in Corylus values in PAZ-2b is most likely due to increased pollen production associated with reductions in canopy-forming trees oak and elm (Feeser and Dörfler 2014). The increase in NAP values, in particular Poaceae, provide the first indication of expansion of open areas, inclusive of a species-rich pastoral component (see P. lanceolata, Rumex-type, Ranunculus-type, and Liguliflorae; Fig. 4b), suggesting the presence of areas dedicated to pastoral grazing. Two cereal-type records indicate small-scale arable farming during the earliest stages of the Landnam phase (ca. 3810 B.C.). Overall, woodland dominates the landscape, reaching its minimum representation at ca. 3760 B.C. (Fig. 4a), indicating the presence of small local openings. Values of Isoetes, a palynological indicator of catchment erosion (Vuorela 1980), in conjunction with LOI % values (Fig. 5c), reflect increased catchment disturbance and sediment in-wash into the lake. An increase in nutrient inputs may be envisaged from the higher representation of Botryococcus and Pediastrum. The latter has no parallel until the Late Bronze Age (PAZ-5) (see Pediastrum curve; Fig. 4b) and may indicate activity relatively close to the coring location. As reported elsewhere (Ghilardi and O’Connell 2013, Molloy and O’Connell 1987), fire does not appear to have played a role in Neolithic woodland clearance. The substantial expansion of alder following the main phase of woodland clearance at ca. 3600 B.C. likely derives from a combination of the local interdrumlin conditions and anthropogenic disturbance. Reduced woodland cover would have facilitated increased overland flow and gravitational soil erosion into inter-drumlin hollows and steep lake margins. In turn, this is expected to have led to higher soil hydraulical saturation as well as soil stabilization, ultimately resulting in the expansion of suitable alder habitat (Bennett and Birks 1990). A similar patJournal of the North Atlantic 15 2017 No. 32 C. Chique, K. Molloy, and A. Potito Including the original collapse at ca. 3920 B.C. (PAZ-2a), 3 distinct Elm Decline events are inferred from the pollen record of L. Muckno, with the second event recorded at ca. 3220 B.C. (PAZ-3b) and the final event at ca. 2960 B.C. (PAZ-3c). Hereafter Ulmus values do not recover to pre-Elm Decline levels for the remainder of the pollen profile. Multiple Elm Decline events have been reported in other locations (Hirons and Edwards 1986, O’Connell 1980) including a comparable chronology at Redbog II (Table 4). Weir (1995) attributed Elm Decline events at that site to human influence, but in the case of L. Muckno the lack of anthropogenic indicators hints at disease as the primary cause for all 3 Elm Decline events. The gradual expansion of Fraxinus marks a change in woodland structure. Light-demanding ash, previously present in the landscape as a frequent but underrepresented taxon, expands in open areas within woodlands (O’Connell and Molloy 2001). A pine stomata recorded at 644 cm suggests local pine presence (see Pinus curve; Fig. 4b), but its representation in the landscape gradually declines from ca. 2850 B.C. onwards. The decline in Alnus values in Zone 3 is likely a result of the inherent dynamism of alder woodlands (Bennett and Birks 1990), in this case due to the absence of catchment disturbance facilitating the creation of suitable habitat for its expansion. Early Bronze Age (PAZ-4) (ca. 3490 - 2590 B.C.). Human activity is constant but at fluctuating levels of intensity during the Early Bronze Age. NAPp representation suggest 3 episodes of increased agricultural activity concentrated during ca. 2590–2460 B.C. (PAZ-4a), ca. 2330–2150 B.C. (PAZ-4b), and ca. 1970–1750 B.C. (PAZ-4c). Renewed catchment disturbance is recorded from 2590 B.C. with small-scale clearance envisaged from modest reductions in AP components and expansion of Poaceae and P. lanceolata curves. Increased representation of Cyperaceae and Salix (Fig. 4b) may provide an indication of openings along lake fringes. The cereal-type curve suggests smallscale cereal cultivation spanning most of the period ca. 2590–1490 B.C following a period of ca. 1220 years (3810–2590 B.C.) with no pollen evidence of arable farming in the vicinity of L. Muckno. Stability in most AP curves suggest limited woodland clearance in PAZ-4b, with NAPp indicating a slight expansion in the extent of grassland/pastoral areas along temporal spread of alder into wet areas. A more intense phase of woodland clearance follows in PAZ-4c with reductions in oak, elm, and ash. AP representation falls to a minimum of 83% by ca. 1840 B.C. (Fig. 4a). Hazel and birch benefit from modifications in woodland canopy structure and slowly expand. The presence of a vegetation Mantel layer primarily composed of rowan/whitebeam (see Sorbus curve, and Prunus, Crataegus, and Sambucus records; Fig. 4b) reflects openings in the woodland environment and (possibly) grazing pressure on woodland fringes (Ghilardi and O’Connell 2013). Increased representation of additional components of the woodland edge community, holly and ivy (Ilex and Hedera curves; Fig. 4b), as well as a better expression of ferns (P. vulgare and Filicales curve; Figs. 4b, 5a), emphasize woodland structure modifications. Pteridium values peak during each of the episodes of increased human activity (particularly in PAZ-4c; Fig. 4b), reflecting bracken colonization of newly opened areas. The establishment of a Filipendula curve represents the presence of wet grassland patches. Higher incidence of arable weeds in Zone 4 (Brassicaceae, P. major/media, and Chenopodiaceae; Figs. 4b, 5a) may reflect disturbance within open habitats possibly related to arable farming (Behre, 1981). In PAZ-4d, apart from a brief decrease in Fraxinus values which may imply further ash clearance, AP curves remain relatively stable, with lower P. lanceolata values suggesting contraction in the extent of pastoral areas, and a disruption of the cereal curve perhaps reflective of a brief lull in arable farming during ca. 1710–1660 B.C. The pollen record indicates that the economy in the Early Bronze Age is still largely grassland-based with a minor but now relatively permanent arable component. Overall, the correspondence of small peaks in microscopic charcoal with each episode of pronounced human activity hints at an anthropogenic source, related to either settlement and/or woodland clearance. Evidence of oak clearance may derive from the settlement at Cullyhanna (Fig. 2a), with 14C evidence (Smith et al. 1973) providing potential chronological correspondence with the phase of increased human activity at ca. 1800 B.C. Regional palynological evidence (Table 4) and additional Irish investigations (Ghilardi and O’Connell 2013; Mighall et al. 2008; Molloy 2005; Molloy and O’Connell 1991, 2004; O’Connell et al. 2014) also provide evidence of land-use intensification, predominantly pastoral, but often with some evidence of cereal-crop cultivation during the early 2nd millennium B.C. Middle–Late Bronze Age (PAZ-5) (ca. 1490–650 B.C.). A substantial expansion in farming activity is recorded during the Middle Bronze Age. This trend is strongest during ca. 1490–1300 B.C. (PAZ-5a), with intensity decreasing but still at a considerable 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 16 level ca. 1300–1030 B.C. (PAZ-5b). A period of lower activity occurs during the Late Bronze Age ca. 1030–650 B.C. (PAZ-5c). During PAZ-5a, tall-canopy oak and elm trees are subject to a degree of removal; however, scrub vegetation is the main target of clearances with strong reductions in hazel and to a lesser extent ash, facilitating increased birch flowering (see peak in Betula; Fig. 4b). Higher values of Poaceae and additional NAPp (P. lanceolata, Rumex-type, Ranunculus- type, Filipendula) represent an expansion of open grassland/pastoral habitats. The impact on woodland cover at ca. 1400 B.C. (AP = 69%) is larger than that recorded in the earlier part of the Bronze Age at ca. 1840 B.C. (PAZ-4c) (Fig. 4a). Frequent records of Osmunda (Fig. 5a) are suggestive of open wet areas, and 2 phases of bracken expansion into newly cleared areas are discernible (see Pteridium curve; Fig. 4b): the first corresponds to the initial phase of woodland clearance at ca. 1400 B.C. (PAZ- 5a), while the second occurs in conjunction with a reduction in Fraxinus percentage and concentration values at ca. 1080 B.C. (PAZ-5b). The latter could be interpreted as indicative of exploitation of ash for timber as there is no evidence for expansion of anthropogenic indicators or for major changes in woodland composition. The gradual expansion of Alnus towards PAZ-5b once again reflects the dynamism of alder woodlands expanding onto wet soils. Hazel scrub expands, possibly replacing oak and/or ash in areas previously subject to clearance. The establishment of a slender Taxus curve (Fig. 4b) suggests that yew benefits from changes in woodland structure, but remains a minor component of the woodlands. AP values in PAZ-5c indicate stability in woodland structure, with hazel scrub dominating the landscape and a degree of elm regeneration. Decreased NAPp values between ca. 1030–650 B.C. indicate further contraction of pastoral/grassland habitats from. Continuity of cereal-type records in PAZ-5 (see also arable weeds; Figs. 4b, 5a) imply prevalence of small-scale arable farming during the Middle–Late Bronze Age, with a possible disruption in cereal cultivation at ca. 700 B.C. In PAZ-5a, catchment disturbance, increased soil erosion, and lake productivity are reflected in micro-charcoal, I. echinospora, and Pediastrum representation. Higher minerogenic inputs are also evident at ca. 1400 B.C. (Fig. 5c). If the local fulachta fia (Fig. 2) are associated with the construction phase of 1700–1300 B.C. (Ó Néill 2003), then an argument can be made for localized activity that may be reflected in higher Pediastrum levels. Regional evidence points to increased agricultural activity during the Middle Bronze Age, including frequent signs of cereal cultivation (particularly post-1400 B.C.) (Table 4). Palynological data from different parts of Ireland (e.g., Plunkett 2009), including a number of profiles in close proximity in the west (Molloy 2005; Molloy and O’Connell 2011, 2012; O’Connell et al. 2001), serve to highlight the spatio-temporal variability in land-use dynamics characterizing this period. However, a widespread upsurge in agricultural activity corresponding to the onset of the Late Bronze Age (ca. 1000 B.C.) (Ghilardi and O’Connell 2013; Molloy and O’Connell 2004, 2016; O’Connell et al. 2014; Overland and O’Connell 2008; Plunkett 2009), also evident in most regional profiles (Table 4), is in clear contrast with the low levels of human activity recorded during ca. 1030–650 B.C. at L. Muckno and perhaps Loughnashade (ca. 1000–600 B.C). Early Iron Age (PAZ-6) (ca. 650–220 B.C.). Increased human activity is indicated as woodland clearance involving ash, oak, and to a lesser degree elm. Hazel scrub, by this point a dominant component in the context of woody vegetation, is subject to pronounced clearance at ca. 300 B.C. The expansion of birch corresponds to reductions in the tall-canopy tree layer and diminished competition from hazel scrub. High NAPp values indicate considerable expansion of grasslands and especially pastoral habitats. Concomitant bracken expansion (see Pteridium curve; Fig. 4b) provides further evidence of open areas. The inferred intensity of pastoral activity and impact on woodland vegetation is comparable to land-use dynamics in the Middle Bronze Age (PAZ- 5a) (Fig. 4a). Frequent records of cereal-type pollen (see also Brassicaceae curve; Fig. 4b) suggest continuity in cereal cultivation. Catchment disturbance is again reflected in LOI % values (Fig. 5c), aquatic indicators, and micro-charcoal. Two linear earthwork features located southeast of the lake (Fig. 2) may indicate Iron Age activity in the wider area of the study site. Along with a number of regional archaeological features (The Dorsey and Aghareagh West; Fig. 1), it was suggested that these comprised a series of Iron Age fortifications in the north midlands (Davies 1955, Kane 1909, Lynn 2008, Walsh 1987), although recent interpretations suggest alternative scenarios (see Ó Drisceoil et al. 2014). With the exception of Redbog II, regional pollen evidence indicates a period of intensifying farming activity (Table 4) often reported in different parts of Ireland (Ghilardi and O’Connell 2013; Molloy 2005; Molloy and O’Connell 1991, 2004, 2012, 2016; Overland and O’Connell 2008; Plunkett 2009; Pilcher and Smith 1979). Journal of the North Atlantic 17 2017 No. 32 C. Chique, K. Molloy, and A. Potito Late Iron Age (PAZ-7) (ca. 220 B.C.–A.D. 220). Zone 7 represents the so-called Late Iron Age Lull (LIAL), a period of subdued agricultural activity during which widespread woodland regeneration may have occurred across Ireland. Locally, this phase dates from ca. 220 B.C. to A.D. 220, with rapid woodland regeneration involving hazel, birch, and ash, but also elm and yew at a slower rate. A fall in Quercus values suggests decreased oak representation in the landscape during ca. 150–70 B.C. However, given the generalized contraction in NAP during this period, local oak clearance is not envisaged. A pollen “screening” effect provided by alder woodlands, presumed to be predominant around the lake, may provide an explanation for the underrepresentation of oak pollen, a feature also reported from other inter-drumlin settings (Hirons 1984). The decline in Alnus representation at ca. 150 B.C. is also unlikely to reflect alder clearance, and it is difficult to explain a reduction in Alnus pollen influx given the liekly prevalence of alder on wetter soils around the lake. Competition from an expanding ash population, a contemporary component of wet woodlands present around the lake today (Foss and Crushell 2012) may provide a viable explanation. Expansion of woody taxa along wetter areas, involving either alder and/or ash, may also be a factor in the absence of Cyperaceae pollen (Fig 4b). Rapid modifications in the organization or “regularity” of the local to regional vegetation mosaic (Hellman et al. 2009) seem to acount for the palynological features noted above. Woodland representation during the Late Iron Age is at its maximum extent since the Early Bronze Age (PAZ-4d) (Fig. 4a). Regeneration of the woodland environment is also reflected in the higher representation of holly (Ilex curve; Fig. 4b) benefiting from the expansion of woodland fringes. The contraction of NAPp points to a considerable reduction of open habitats and loss of diversity (e.g., Urtica, Fabaceae). The very low records of P. lanceolata (0.1%) only have parallel in the Late Neolithic (PAZ-3) ~2600 years earlier. Open areas dedicated to pastoral grazing are negligible by the early 1st century A.D. Cereal-type records indicate that arable farming in the area ceased briefly at ca. 70 B.C., resuming shortly after at a small scale. The effects of declining catchment activity are well represented in the decreased sediment accumulation rate (Fig. 5d) (see also decline in micro-charcoal, Pediastrum, Botryococcus, and I. echinospora; Fig. 4b). Weir (1995) (see also Baillie 1993) argued for a potential link between climatic deterioration and population reduction occurring during 200 B.C.– A.D. 200 in Ireland. More recently, Baillie and Brown (2013) consider the possibility of an environmental downturn at 44–42 B.C. leading to a collapse in population in the ensuing centuries. From a palynological perspective, a degree of chronological variability concerning “features” comparable to the LIAL is evident. In the west, they are often registered during the early 1st millennium A.D. (Molloy and O’Connell 2004, 2016; Lomas-Clarke and Barber 2004) and slightly later in the south (Overland and O’Connell 2008). Dates tend to oscillate around the 1st and 2nd centuries A.D. in other locations considered (Ghilardi and O’Connell 2013; Molloy 2005, Newman et al. 2007). This is in contrast with the evidence available from L. Muckno and nearby profiles (Table 4) where the LIAL is registered centuries earlier than in the west and which suggest a synchronous woodland regeneration phase in northeastern Ireland. The variability observed throughout Ireland is highlighted by modelling of 14C evidence from several pollen profiles placing the onset of the LIAL at various times within the window of 200 B.C.–A.D. 200 questioning the validity of a climatedriven lull and supporting the role of cultural factors (Coyle-McClung 2013). Vegetation and land-use dynamics in the historical period Late Iron Age–Early Historic period (PAZ-8) (ca. A.D. 220–800). A period of intense human activity characterized by the unprecedented expansion of open habitats and emphasis on cereal-crop cultivation is reflected in PAZ-8. In Subzone 8a, clearances involving hazel scrub, ash, and birch and the concomitant expansion of grassland/pastoral habitats (Fig. 4b) indicate renewed focus on pastoral activity. Arable farming also experiences a sudden and brief expansion at ca. A.D. 300. Continued clearance involves oak at ca. A.D. 760 (PAZ- 8b) along with alder removal, which may point to agricultural expansion onto marginal land. Elm and yew representation in the landscape is minimal from this point onwards. Deforestation reaches its climax at ca. A.D. 700 with AP falling to its lowest levels yet recorded (61%; Fig. 4a). The expansion of open/pastoral habitats, inclusive of the re-establishment of nutrient-rich areas (see Urtica values) and wet grassland patches (see Filipendula and new records of Valeriana officinalis; Fig. 4b) suggests intense pastoral grazing. Sedges expand once more along lake edges, and bracken colonization of open areas is widespread (Fig. 4b). From ca. A.D. 500, a substantial increase in cereal-type records indicates cereal cultivation first becomes an important component of the farming economy (see also an increase in arable weeds; Figs. 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 18 4b, 5a). Secale-type pollen is recorded sporadically during ca. A.D. 530–760, and by ca. A.D. 630 the arable component is at its maximum extent although an economy dominated by grazing prevails. High levels of catchment disturbance (see micro-charcoal curve and fluctuations in LOI %; Figs. 4b, 5c) have a considerable but now continuous effect on aquatic indicators. The high representation of Isoetes indicates unprecedented catchment erosion and sediment inwash into the lake. The sediment accumulation rate increases from this point onwards (~7 yr/cm; Fig. 5d), emphasizing the combined effects of higher landto- water material export and (possibly) increased lake productivity (Bennett and Buck 2016). The prevalence of shoreweed in shallow areas (L. uniflora curve; Fig. 4b) may provide an indication of modifications in physico-chemical parameters in the littoral of the lake (Preston et al. 2002:537). The increase in Pinus values in PAZ-8a merits consideration, with the pine curve expanding in two instances in the profile, at the onset of PAZ-5 and PAZ-8 (Pinus curve; Fig. 4). After a widespread decline at ca. 2000 B.C., pine populations in Ireland were thought to have survived in isolated pockets before becoming extinct at ca. A.D. 400 (Roche et al. 2009). However, some evidence suggests the survival of relict pine stands through the historical period (McGeever and Mitchell 2016; Overland and O’Connell 2008, 2011). In L. Muckno, increased influx of Pinus pollen may indeed reflect the presence of pine within the regional source area of the lake that is only expressed during periods of decreased influx of other AP. However, considering the degree of anthropogenic “forcing” on catchment soils during PAZ-5 and particularly PAZ-8, a logical argument for this feature would also be reworking of old pollen into the lake (Edwards and Whittington 2001). While there is some evidence for reworked material higher up in the profile (see 14C reversal at 301 cm; Figs. 3, 5), there is nothing to suggest this is a factor in this part of the sediment record. Woody vegetation clearance and expansion of agricultural activity through PAZ-8 is gradual, with human activity in Subzone 8a (ca. A.D. 220–440) representing a brief transitional period from the LIAL into the historical period (PAZ-8b), also reported in regional profiles (Table 4), and which has been suggested to reflect Romano-British influence in Ireland (McCormick 2014). Given the active role that many ecclesiastical settlements played in the agrarian economy (Ó Corráin 2005), a potential link between this phase of increased cereal cultivation and the monastery at L. Muckno (Fig. 2) is possible. Ó Dufaigh (2011) suggests that the monastery was already established by ca. A.D. 700 (note peak in cereal-type pollen at ca. A.D. 630). The pollen record from Church Lough (O’Connell and Ní Ghráinne 1994) reflects a comparable scenario, with an upsurge in cereal cultivation seemingly corresponding to the foundation of a monastic site next to the lake in the late 7th century. On the other hand, Hall (2005) found no palynological evidence to support increased arable farming along with the establishment of a number of monastic settlements in different parts of Ireland. The local prevalence of ringforts (Fig. 2) reflects landscape openness and a settlement pattern based on individual farmsteads characteristic of the time (Edwards 2005:238, McCormick et al. 2011) and supports palynological evidence for increased farming from ca. A.D. 500. The modern townland Tullanacrunat (Tulaight na Cruithneachta), translating into the “the hill of bread-wheat” (Carville and Duffy 2011:15), situated in the southern shore of the coring location may provide placename evidence for cereal cultivation. The word cruithnecht or “breadwheat” is included in a list of cereal-crop varieties in a Medieval Irish law text as a staple cereal grain of the time (Kelly 1997:219–221). The introduction of Secale at ca. A.D. 530 corresponds well with available Irish macrofossil evidence (McClatchie et al. 2015), suggesting its adoption among cereal-crop varieties during the early stages of the historical period. However, it does not appear to have become of major importance in this area. The results from L. Muckno integrate well with the post- A.D. 700 regional expansion of agricultural activity and focus on arable farming (Table 4). However, a phase of woodland regeneration recorded regionally during the 6th century and tentatively attributed to climatic deterioration (Weir 1993, 1995) is not evident in the pollen record of L. Muckno. Examples from further afield (Ghilardi and O’Connell 2013; Lomas-Clarke and Barber 2004; Molloy and O’Connell 2004, 2016; Newman et al. 2007; Overland and O’Connell 2008) tend to suggest an increase in agricultural activity with the introduction of a sizable arable component between the 5th and 8th century A.D. (see Overland and O’Connell 2011 for a review). These steps towards a mixed agricultural regime seem to have occurred slightly earlier in L. Muckno and possibly in its regional context. Kerr et al. (2009) suggest that the 8th and 9th centuries A.D. in Ireland were marked by a climatic downturn, which might have led to a shift towards cereal-crop production as winter conditions became too severe to support the traditional cattle-based economy. This pattern may be (partially) reflected in the pollen record of L. Muckno, with increased cereal production during the 8th century A.D., but it Journal of the North Atlantic 19 2017 No. 32 C. Chique, K. Molloy, and A. Potito is clear that an expansion in cereal cultivation begins earlier in the 7th century A.D. Viking Age and Medieval (PAZ-9) (ca. A.D. 800– 1190). Based on cereal-type records, 3 distinct phases in arable farming are apparent in PAZ-9, with an initial disruption during ca. A.D. 830–960 (PAZ-9a), followed by an increase ca. A.D. 990–1140 (PAZ- 9b), and a further interruption ca. A.D. 1160–1190 (PAZ-9c). A degree of contraction in the extent of pastoral habitats along with hazel and ash expansion is recorded ca. A.D. 830–960 in PAZ-9a. Two distinct disruptions in the cereal-type curve at ca. A.D. 830 and 960 along with (overall) lower cereal records may provide evidence of local disturbance in cerealcrop cultivation (see also contraction of Brassicaceae curve in PAZ-9a; Fig. 4b). In PAZ-9b, woody vegetation targeted for clearance includes hazel, oak, birch, ash, and alder. Grassland/pastoral habitats expand, pointing to renewed emphasis on pastoral grazing. Increased cereal-type records (inclusive of Secale) suggest recovery in arable farming following the A.D. 960 disruption, attaining its maximum expression during ca. A.D. 990–1140 (see also arable weeds; Figs. 4b, 5a). Cannabis/Hummulus-type records (Fig. 4b) may also provide discreet evidence for hemp cultivation. The representation of aquatic indicators and micro-charcoal indicate unparalleled levels of catchment disturbance and human influence on the lake environment following ca. A.D. 1100. Overall, the sedimentary record reflects the predominance of a mixed agrarian economy until a fall in cereal-type records suggests arable farming declines once more at ca. A.D. 1160. It is possible that these distinct fluctuations in cereal-type pollen relate to a period of local conflict registered in documentary sources. Three events are recorded in a range of monastic chronicles collectively known as the Irish “annals” (Hennesy, 1866, 1887; O’Donovan 1848). The sacking of Muckno by Vikings is described during A.D. 832 and 933. One additional record suggests Muckno was plundered “completely” during A.D. 1110 (see also Carville and Duffy 2011:30–38). The correspondence of these events with the 3 phases of disruption in cereal cultivation recorded at L. Muckno is noteworthy. However, the uncertainty associated with age–depth modelling in this section of the core and each relevant depth must be highlighted: 332 cm (A.D. 830 ± 250 years), 316 cm (A.D. 960 ± 250 years), 288 cm (A.D. 1160 ± 320 years) (Fig. 3; see also Supplementary Table 1, available online at http:// www.eaglehill.us/JONAonline/suppl-files/J061316- Chique-s1, and, for BioOne subscribers, at http:// dx.doi.org/10.1656/J061316.s1). Chronological standardization of the Irish annals (McCarthy 1998, 2008) provides reliability in the accuracy of these annotated events which are tentatively interpreted as reflective of a period of “conflict” ca. A.D. 800–1190. It is possible that the evidence presented here represents a (palynological) measurable impact on farming activity. Viking raids are considered to have had a disruptive effect on local economies and demographics (Edwards 2005:295, Hughes 2005:638, Mitchell and Ryan 1997:297), and in this scenario, a disruption in cereal cultivation and pastoral grazing is conceivable. Alternative explanations include crop failure and livestock losses resulting from both disease and climatic deterioration reported in Medieval Irish sources (see events at A.D. 700 and A.D. 964; Ó Corráin 2005:576). In terms of climate, the hypothesis put forward by Kerr et al. (2009) may find some support in the inferred contraction of pastoral grazing at ca. A.D. 800, while the upsurge in cereal production following ca. A.D. 990 occurs within a background of fair climatic conditions in the form of the Medieval Warm Period (ca. A.D. 950–1250) (Mann et al. 2009). If climate amelioration indeed characterizes the later stages of the Medieval, then perhaps an even stronger argument can be made for conflict resulting in the final disruption in cereal cultivation at ca. 1160 A.D. From ca. A.D. 800 onwards, the Irish palynological record indicates continuous expansion towards a mixed farming economy in which pastoral grazing remains a fundamental component. The regional evidence (Table 4) points to increased agricultural activity during the Viking Age/Medieval, but it is also clear that cereal cultivation is affected at different times, a factor attributed by Weir (1995) to climatic deterioration and/or conflict in the form of Viking raids. The results presented here may therefore support the hypothesis of conflict affecting farming activity not only locally but also regionally. Overall, excluding distinct disruptions in arable farming, the pollen evidence from L. Muckno finds good agreement with general patterns observed in Ireland. During the period ca. A.D. 990–1140, cereal-crop cultivation expands, achieving levels hitherto unseen during the Holocene. Conclusion High-resolution palynological analysis in combination with a chronology based on 14C and 210Pb 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 20 from a sediment core from L. Muckno have enabled the first account of Mid–Late Holocene landscape change in County Monaghan. Following a brief Neolithic Landnam, a period of undiscernible human activity in a predominantly wooded environment is apparent until the Early Bronze Age. From this point onwards, human activity in the landscape is continuous, but at variable levels of intensity, with hazel scrub and herbaceous taxa gradually becoming dominant components of the vegetation structure. A more homogeneous trend towards increased activity and a mixed agrarian economy is recorded from the onset of the historical period. Farming activity, removal of woodland cover and expansion of open habitats climax during the Viking Age/Medieval period. The possibility of a period of “conflict” resulting in consecutive fluctuations in farming activity serves to emphasize the importance of collation of documentary and paleoecological records, particularly if a high sampling resolution and reasonable chronological constraint are available. The high expression of arboreal pollen is a characterizing feature of the pollen profile as a whole and is a factor attributable to the large surface area of the lake. The enhanced dispersal capacity of AP is reflected in its elevated representation throughout prehistorical and historical periods even when woodland cover in the landscape is at its minimum (Figs. 4a, 6). Nevertheless, the results presented here demonstrate that the pollen record of this large lake system can provide insights into localized events. AP overrepresentation may partially account for the low amount of cereal-type records through the profile, forming a slender curve as often reported in Irish pollen diagrams, and indicative of constant arable farming in mineral soils close to the lake from the Early Bronze Age onwards. The “continuity” in cereal-type records, and hence the inferred arable component, validates the benefit of implementing high pollen counts in Irish palynological reconstructions, particularly in systems with a tendency towards high AP representation. The local interdrumlin setting exerts a strong influence on the pollen assemblage of the lake exemplified by the high variability in Alnus pollen influx. Tentative chronological associations with local and regional archaeological records are established. The integration of the results from L. Muckno with patterns recognized in neighboring profiles results in a more comprehensive regional palynological dataset now encompassing areas of counties Armagh, Louth, and Monaghan. A comparable collapse of elm populations is recorded at L. Muckno and Redbog II, whereas contrasting levels of activity at L. Muckno, and perhaps Loughnashade, with a range of sites during the Mid-Late Bronze Age is evident. An argument can be made for an earlier onset of the LIAL, as well as the subsequent upsurge in human activity and woodland clearance towards the Early Historic period in parts of Monaghan, but also in Louth and Armagh, in contrast to several other locations in Ireland. Acknowledgments We gratefully acknowledge the support and help received throughout this investigation from those listed below. Pat O’Rafferty, Michelle McKeown, Seamus McGinley, and Stephen Galvin helped during fieldwork. Daisy Spencer assisted in the inspection of pollen data. Michael O’Connell gave general advice on this investigation. Heritage Officer Shirley Clerkin (Monaghan County Council) kindly provided information on different aspects of the study area and Grace Moloney (Clogher Historical Society) provided regional historical information. Niall Roycroft (National Roads Authority) was kind enough to provide feedback on local archaeological records. Dan McCarthy (Trinity College Dublin) gave assistance in the interpretation and chronological standardization of the Irish annals. Carlos Chique was financially supported by a Postgraduate Fellowship from NUI, Galway (Hardiman Research Scholarship). We also express gratitude to The Irish Quaternary Research Association (IQUA) and the 14CHRONO Centre at Queen’s University Belfast for providing a number of radiocarbon dates through the Bill Watts 14 Chrono Award. Literature Cited Aalen, F.H.A., K. Whelan, and M. Stout. 1997. Atlas of the Irish Rural Landscape. Cork University Press, Cork, Ireland. 352 pp. Baillie, M. 1993. Dark ages and dendrochronology. Emania 11:5–12. Baillie, M., and D. Brown. 2013. A Chronological framework for the period from 208 B.C. to A.D. 600. Emania 21:59–71. Behre, K.E. 1981. Anthropogenic indicators in pollen diagrams. Pollen et Spores 32:225–245. Bennett, K.D., and C.E. Buck. 2016. Interpretation of lake sediment accumulation rates. The Holocene 26:1092–1102. Bennett, K.D., and H.J.B. Birks. 1990. Postglacial history of alder (Alnus glutinosa (L.) Gaertn.) in the British Isles. Journal of Quaternary Science 5:123–133. Beug, H.J. 2004. Leitfaden der Pollenbestimmung für Mitteleuropa und Angrenzende Gebiete. Pfeil, Munich, Germany. 542 pp. Birks, H.J.B. 1973. Past and Present Vegetation of the Isle of Skye–A Palaeoecological Study. Cambridge University Press, Cambridge, UK. 415 pp. Blaauw, M., and J.A. Christen. 2011. Flexible paleoclimate age–depth models using an autoregressive gamma process. Bayesian Analysis 6:457–474. Journal of the North Atlantic 21 2017 No. 32 C. Chique, K. Molloy, and A. Potito Broström, A., S. Sugita, M.J. Gaillard, and P. Pilesjö. 2005. Estimating the spatial scale of pollen dispersal in the cultural landscape of southern Sweden. The Holocene 15:252–262. Carville, G., and J. Duffy. 2011. On the Way: The Journey of the People of Muckno with a Survey of the Placenames of Muckno. Muckno Parish Pastoral Team, Castleblayney, Ireland. 264 pp. Caseldine, C., and J. Hatton. 1996. Early land clearance and wooden trackway construction in the third and fourth millennia B.C. at Corlea, Co. Longford. Biology and Environment (Proceedings of the Royal Irish Academy) 95B:11–19. Chique, C., A. Potito, K. Molloy, and J. Cornett. In review. Tracking recent human impacts on a nutrient sensitive Irish lake–Integrating landscape to water linkages. Hydrobiologia. Clark, S.H.E., and K.J. Edwards. 2004. Elm bark beetle in Holocene peat deposits and the northwest European elm decline. Journal of Quaternary Science 19:525–528. Coope, G.R., J.H. Dickson, J.A. McCutcheon, and G.F. Mitchell. 1979. The Late glacial and Early Postglacial Deposit at Drumurcher, Co. Monaghan. Proceedings of the Royal Irish Academy Section B: Biological, Geological, and Chemical Science 79:63–85. Coyle-McClung, L.C. 2013. The Late Iron Age Lull—not so Late Iron Age after all! Emania 21:73–83. Edwards, K.J., and G. Whittington. 2001. Lake sediments, erosion, and landscape change during the Holocene in Britain and Ireland. Catena 42:143–173. Edwards, N. 2005. The Archaeology of Early Medieval Ireland, ca. 400–1169: Settlement and economy. Pp. 235–296, In D. Ó Cróinín (Ed.). A New History of Ireland–I Prehistoric and Early Ireland. Oxford University Press, Oxford, UK. 1219 pp. Feeser, I., and W. Dörfler. 2014. The glade effect: Vegetation openness and structure and their influences on arboreal pollen production and the reconstruction of anthropogenic forest opening. Anthropocene 8:92–100. Foss, P.J., and P. Crushell. 2012. Wetland survey County Monaghan II. Monaghan County Council/The Heritage Council, The Glen, Monaghan, Ireland. 89 pp. Ghilardi, B., and M. O’Connell. 2013. Fine-resolution pollen-analytical study of Holocene woodland dynamics and land use in north Sligo, Ireland. Boreas 42:623–649. Hall, V.A. 2000. The documentary and pollen analytical records of the vegetational history of the Irish landscape A.D. 200-1650. Peritia 14:342–71. Hall, V.A. 2005. The vegetation history of monastic and secular sites in the midlands of Ireland over the last two millennia. Vegetation History and Archaeobotany 15:1–12. Heiri, O., A.F. Lotter, and G. Lemcke. 2001. Loss on ignition as a method for estimating organic and carbonate content in sediments: Reproducibility and comparability of results. Journal of Paleolimnology 25:101–110. Hellman, S., M.-J. Gaillard, A. Broström, and S. Sugita. 2008. The REVEALS model, a new tool to estimate past regional plant abundance from pollen data in large lakes: validation in southern Sweden. Journal of Quaternary Science 23:21–42. Hellman, S., M.J. Bunting, and M.-J. Gaillard. 2009. Relevant source area of pollen in patchy cultural landscapes and signals of anthropogenic landscape disturbance in the pollen record: A simulation approach. Review of Palaeobotany and Palynology 153:245–258. Hennessy, W.M. (Ed.). 1866. Chronicum Scotorum. Longmans, Green, Reader, and Dyer, London, UK. Hennessy, W.M. (Ed.). 1887. Annals of Ulster. Alex Thom and Co, Dublin, Ireland. Hirons, K.R. 1984. Palaeoenvironmental investigations in east Co. Tyrone, Northern Ireland. Unpublished PhD Thesis. Queen’s University, Belfast, Ireland. 419 pp. Hirons, K.R., and K.J. Edwards. 1986. Events at and around the first and second Ulmus declines: Palaeoecological investigations in Co. Tyrone, Northern Ireland. New Phytologist 104:131–153. Hodges, H.W.M. 1958. A hunting camp at Cullyhanna Lough near Newton Hamilton, County Armagh. Ulster Journal of Archaeology 21:7–13. Hughes, K. 2005. The Irish Church, 800 –c.1050. Pp. 635 –655, In D. Ó Cróinín (Ed.). A New History of Ireland–I Prehistoric and Early Ireland. Oxford University Press, Oxford, UK. 1219 pp. Irish Pollen Site Database (IPOL). 2013. Site Table. Available on line at: http://www.ipol.ie/index.html. Accessed 10 May 2016. Kane, W.F.V. 1909. The Black Pig’s Dyke: The ancient boundary fortification of Uladh. Proceedings of the Royal Irish Academy 27C:301–328. Kelly, F. 1997. Early Irish Farming: A Study Based Mainly on the Law-Texts of the 7th and 8th Centuries A.D.. School of Celtic Studies, Dublin Institute for Advanced Studies, Dublin, Ireland. 751 pp. Kerr, T. 2009. The height of fashion: Raised raths in the landscape of north-west Ulster. The Journal of Irish Archaeology 18:63–75. Kerr, T., G.T. Swindles, and G. Plunkett. 2009. Making hay while the sun shines? Socio-economic change, cereal production, and climatic deterioration in Early Medieval Ireland. Journal of Archaeological Science 36:2868–2874. Lomas-Clarke, S.H., and K.E. Barber. 2004. Palaeoecology of human impact during the historic period: Palynology and geochemistry of a peat deposit at Abbeyknockmoy, Co. Galway, Ireland. The Holocene 14:721–731. Lynn, C. 2008. Navan Fort—Legendary Capital of Prehistoric Ulster. Archaeology Ireland, Heritage Guide 40:1–6. Mann, M.E., Zhang, Z. Rutherford, S. Bradley, R.S. Hughes, M.K. Shindell, D. Ammann, C. Faluvegi, G. and F. Nei. Global signatures and dynamical origins of the Little Ice Age and Medieval climate anomaly. Science 326:1256–1260. McCarthy, D.P. 1998. The chronology of the Irish annals. Proceedings of the Royal Irish Academy 98C:203–255. McCarthy, D.P. 2008. The Irish Annals: Their Genesis, Evolution, and History. Four Courts Press, Dublin, Ireland. 416 pp. 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 22 McClatchie, M., A. Bogaard, S. Colledge, N.J. Whitehouse, R.J. Schulting, P. Barratt, and T.R. McLaughlin. 2014. Neolithic farming in north-western Europe: Archaeobotanical evidence from Ireland. Journal of Archaeological Science 51:206–215. McClatchie, M., F. McCormick, T.R. Kerr, and A. O’Sullivan. 2015. Early medieval farming and food production: A review of the archaeobotanical evidence from archaeological excavations in Ireland. Vegetation History and Archaeobotany 24:179–186. McCormick, F. 2014. Agriculture, settlement and society in Early Medieval Ireland. Quaternary International 346:119–130. McCormick, F., T. Kerr, M. McClatchie, and A. O’Sullivan. 2011. The archaeology of livestock and cereal production in Early Medieval Ireland, A.D. 400-1100. Early Medieval Archaeology Project (EMAP) Report 5.1. 144 pp. Available online at: http://www.emap.ie/ emap_reports.html. Accessed 25 March 2016. McGeever, A.H., and F.J.G. Mitchell. 2016. Re-defining the natural range of Scots pine (Pinus sylvestris L.): A newly discovered microrefugium in western Ireland. Journal of Biogeography 43:2199–2208. Mighall, T.M., S. Timpany, J.J. Blackford, J.B. Innes, C.E. O’Brien, W. O’Brien, and S. Harrison. 2008. Vegetation change during the Mesolithic and Neolithic on the Mizen Peninsula, Co. Cork, south-west Ireland. Vegetation History and Archaeobotany 17:617–628. Mingram, J., J.F.W. Negendank, A. Brauer, D. Berger, A. Hendrich, M. Köhler, and H. Usinger. 2007. Long cores from small lakes: Recovering up to 100-mlong lake sediment sequences with a high-precision rod-operated piston corer (Usinger-corer). Journal of Paleolimnology 37:517–528. Mitchell, G.F. 1951. Studies in Irish Quaternary deposits: No. 7. Proceedings of the Royal Irish Academy 53B:111-206. Mitchell, F., and M. Ryan. 1997. Reading the Irish Landscape. Town House and Country House, Dublin, Ireland. 392 pp. Mitchell, F.J.G., B.S. Stefanini, and R. Marchant. 2013. A catalogue of Irish pollen diagrams. Biology and Environment (Proceedings of the Royal Irish Academy) 113B:103–133. Molloy, K. 2005. Holocene vegetation and land-use history at Mooghaun, south-east Clare, with particular reference to the Bronze Age. Pp. 255–301, In E. Grogan (Ed.). The North Munster Project. Vol. 1. The Later Prehistoric Landscape of South-East Clare. Discovery Programme Monographs 6. Wordwell, Bray, Ireland. 341 pp. Molloy, K., and M. O’Connell. 1987. The nature of the vegetation changes about 5000 BP with particular reference to the elm decline: Fresh evidence from Connemara, Western Ireland. New Phytologist 106:203–220. Molloy, K., and M. O’Connell. 1991. Palaeoecological investigations towards the reconstruction of woodland and land-use history at Lough Sheeauns, Connemara, western Ireland. Review of Palaeobotany and Palynology 67:75–113. Molloy, K., and M. O’Connell. 1995. Palaeoecological investigation towards the reconstruction of environment and land-use changes during prehistory at Céide Fields, western Ireland. Probleme der Kiistenforschung im siidlichen Nordseegebiet 23:187–225. Molloy, K., and M. O’Connell. 2004. Holocene vegetation and land-use dynamics in the karstic environment of Inis Oírr, Aran Islands, western Ireland: Pollen analytical evidence evaluated in light of the archaeological record. Quaternary International 113:41–64. Molloy, K., and M. O’Connell. 2011. Boom and bust or sustained development? Fossil pollen records and new insights into Bronze Age farming in County Clare. Pp. 57–71, In S. Conrad, E. Danaher, and M. Stanley (Eds.). Past Times, Changing Fortunes–Proceedings of a public seminar on archaeological discoveries on national road schemes, August 2010. National Roads Authority, Dublin, Ireland. 170 pp. Molloy, K., and M. O’Connell. 2012. Prehistoric farming in western Ireland: Pollen analysis at Caheraphuca, Co. Clare. Pp. 109–122, In S. Delaney, D. Bayley, E. Lyne, S. McNamara, J. Nunan, and K. Molloy (Eds.). Borderlands: Archaeological Investigations on the Route of the M18 Gort to Crusheen Road Scheme, NRA Scheme Monographs 9. National Roads Authority, Dublin, Ireland. 184 pp. Molloy, K., and M. O’Connell. 2016. Farming impact in Ireland from the Neolithic to recent times with particular reference to a detailed pollen record from east Galway. Pp. 27-43, In M. O’Connell, F. Kelly, and J.H. McAdam (Eds.). Cattle in Ancient and Modern Ireland: Farming Practices, Environment and Economy. Cambridge Scholars Publishing, Newcastle upon Tyne, UK. 230 pp. Moore, P.D., J.A. Webb, and M.E. Collinson. 1991. Pollen Analysis, 2nd Edition. Blackwell, London, UK. 216 pp. Newman, C., M. O’Connell, M. Dillon, and K. Molloy. 2007. Interpretation of charcoal and pollen data relating to a Late Iron Age ritual site in eastern Ireland: A holistic approach. Vegetation History and Archaeobotany 16:349–365. O’Brien, W. 2012. The Chalcolithic in Ireland: A chronological and cultural framework. Pp. 211–225, In M.J. Allen, J. Gardiner, and A. Sheridan (Eds.). Is there a British Chalcolithic? People, Place, and Polity in the Later 3rd Millennium. Oxbow Books, Oxford, UK. 340 pp. O’Connell, M. 1980. The developmental history of Scragh Bog, Co. Westmeath, and the vegetation history of its hinterland. New Phytologist 85:301–19. O’Connell, M. 1991. The Neolithic environment. Pp. 44– 47, In R. Michael (Ed.). The Illustrated Archaeology of Ireland. Country House, Dublin, Ireland. 224 pp. O'Connell, M., and E. Ní Ghráinne. 1994. Inishbofin: Palaeoecology. Pp. 61–101, In P. Coxon, and M. O'Connell (Eds.). Clare Island and Inishbofin: Field Guide 17. Irish Association for Quaternary Studies, Dublin, Ireland. 102 pp. O’Connell, M., and K. Molloy. 2001. Farming and woodland dynamics in Ireland during the Neolithic. Biology and Environment (Proceedings of the Royal Irish Academy) 101B:99–128. Journal of the North Atlantic 23 2017 No. 32 C. Chique, K. Molloy, and A. Potito O’Connell, M., K, Molloy, and H. McMahon. 2001. Reconstructing prehistoric farming activity and human impact at a fine spatial resolution: Palaeoecological investigations at Mooghaun, Co. Clare, western Ireland. Pp. 161-186, In P. Schauer (Ed.). Beiträge zur Siedlungsarchäologie und zum Landschaftswandel. Regensburger Beiträge zur Prähistorischen Archäologie 7. O’Connell, M., B. Ghilardi, and L. Morrison. 2014. A 7000-year record of environmental change, including early farming impact, based on lake-sediment geochemistry and pollen data from County Sligo, western Ireland. Quaternary Research 81:35–49. O’Donovan, J. (Ed.). 1848. Annals of the Kingdom of Ireland. Hodges and Smith, Dublin, Ireland . O’Sullivan, A. 1998. The Archaeology of Lake Settlement in Ireland. Discovery Programme Monographs 4. Royal Irish Academy, Dublin, Ireland. 236 pp. Overland, A., and M. O’Connell. 2008. Fine-spatial palaecological investigations towards reconstructing late Holocene environmnental change, landscape evolution, and farming activity in Barress, Beara Peninsula, southwestern Ireland. The Journal of the North Atlantic 1:37–73. Overland, A., and M. O’Connell. 2011. New insights into late Holocene farming and woodland dynamics in western Ireland with particular reference to the early medieval horizontal watermill at Kilbegly, Co. Roscommon. Review of Palaeobotany and Palynology 163:205–226. Ó Corráin, D. 2005. Ireland ca. 800: Aspects of society. Pp. 549–607, In D. Ó Cróinín (Ed.). A New History of Ireland: I. Prehistoric and Early Ireland. Oxford University Press, Oxford, UK. 1219 pp. Ó Drisecoil, C., M. Leenane, S. Davis, B. Fitzgibbon, and M. Teehan. 2014. The Black’s Pig Dyke regional project report. Vol. 1. The Heritage Council and Cavan/ Monaghan/Roscommon/Longford County Council, The Glen, Monaghan, Ireland. 184 pp. Ó Dufaigh, S. 2011. Maoldoid of Muckno. Clogher Record 20:473–490. Ó Néill, J. 2003. Lapidibus in igne calefactis coquebatur: The historical burnt mound “tradition”. The Journal of Irish Archaeology 12:79–85. Parker, A.G., A.S. Goudie, D.E. Anderson, M. Robinson, and C. Bonsall. 2002. A review of the mid-Holocene elm decline in the British Isles. Progress in Physical Geography 26:1–45. Pilcher, J.R., and A.G. Smith. 1979. Palaeoecological investigations at Ballynagilly, a Neolithic and Bronze Age settlement in County Tyrone, Northern Ireland. Philosophical transactions of the Royal Society of London. Series B, Biological sciences 286:345–369. Plunkett, G. 2009. Land-use patterns and cultural change in the Middle to Late Bronze Age in Ireland: Inferences from pollen records. Vegetation History and Archaeobotany 18:273–295. Preston, C.D., D.A. Pearman, and T.D. Dines (Eds). 2002. New Atlas of the British and Irish Flora. Oxford University Press, Oxford, UK. 912 pp. Reimer, P., E. Bard, A. Bayliss, J. Beck, P. Blackwell, C. Ramsey, C. Buck, H. Cheng, R. Edwards, M. Friedrich, P. Grootes, T. Guilderson, H. Haflidason, I. Hajdas, C. Hatte, T. Heaton, D. Hoffman, A. Hogg, K. Hughen, K. Kaiser, B. Kromer, S. Manning, M. Niu, R. Reimer, D. Richards, E. Scott, J. Southon, R. Staff, C.S. Turney, and J. van der Plicht. 2013. IntCal13 and Marine13 radiocarbon age calibration curves 0–50,000 years cal BP. Radiocarbon 55:1869–1887. Roche, J.R., F.J.G. Mitchell, and S. Waldren. 2009. Plant community ecology of Pinus sylvestris, an extirpated species reintroduced to Ireland. Biodiversity and Conservation 18:2185–2203. Schulting, R., E. Murphy, C. Jones, and G. Warren. 2012. New dates from the north and a proposed chronology for Irish court tombs. Proceedings of the Royal Irish Academy. Section C: Archaeology, Celtic Studies, History, Linguistics, Literature 112C:1–60. Sjögren, P., W.O. van der Knaap, and J.F.N. van Leeuwen. 2015. Pollen dispersal properties of Poaceae and Cyperaceae: First estimates of their absolute pollen productivities. Review of Palaeobotany and Palynology 216:123–131. Smith, A.G., G.W. Pearson, and J.R. Pilcher. 1973. Belfast radiocarbon dates VI. Radiocarbon 15:599–610. Stolze, S., W. Dörfler, T. Monecke, and O. Nelle. 2012. Evidence for climatic variability and its impact on human development during the Neolithic from Loughmeenaghan, County Sligo, Ireland. Journal of Quaternary Science 27:393–403. Stolze, S., R. Muscheler, W. Dörfler, and O. Nelle. 2013. Solar influence on climate variability and human development during the Neolithic: Evidence from a high-resolution multi-proxy record from Templevanny Lough, County Sligo, Ireland. Quaternary Science Reviews 67:138–159. Sugita, S. 2007. Theory of quantitative reconstruction of vegetation I: Pollen from large sites REVEALS regional vegetation composition. The Holocene 17:229–41. Vaughan, A.P.M., L.A. Dowling, F.J.G. Mitchell, S.E. Lauritzen, A.M. McCabe, and P. Coxon. 2004. Depositional and post-depositional history of warm stage deposits at Knocknacran, Co. Monaghan, Ireland: Implications for preservation of Irish last interglacial deposits. Journal of Quaternary Science 19:577–590. Verrill, L., and R. Tipping. 2010. Use and abandonment of a Neolithic field system at Belderrig, Co. Mayo, Ireland: Evidence for economic marginality. The Holocene 20:1011–1021. Vuorela, I. 1980. Microspores of Isoetes as indicators of human settlement in pollen analysis. Memoranda Societatis pro Fauna et Flora Fennica 56:13–19. Waddell, J. 2010. The Prehistoric Archaeology of Ireland, 3rd Edition. Wordwell, Dublin, Ireland. 455 pp. Walker, M.J.C., M. Berkelhammer, S. Björck, L.C. Cwynar, D.A. Fisher, A.J. Long, J.J. Lowe, R.M. Newnham, S.O. Rasmussen, and H. Weiss. 2012. Formal subdivision of the Holocene Series/Epoch: A discussion paper by a working group of INTIMATE (Integration of ice-core, marine and terrestrial records) and the Subcommission on Quaternary Stratigraphy (International Commission on Stratigraphy). Journal of Quaternary Science 27:649–659. 2017 Journal of the North Atlantic No. 32 C. Chique, K. Molloy, and A. Potito 24 Walsh, A. 1987. Excavating the Black Pig’s Dyke. Emania 3:5–11. Warren, G., S. Davis, M. McClatchie, and R. Sands. 2014. The potential role of humans in structuring the wooded landscapes of Mesolithic Ireland: A review of data and discussion of approaches. Vegetation History and Archaeobotany 23:629–646. Weir D.A. 1993. Dark ages and the pollen record. Emania 11:21–30. Weir D.A. 1995. A palynological study of landscape and agricultural development in County Louth in the second millennium A.D.. Pp. 77–126, In Discovery Programme Reports: 2–Project Results 1993. Royal Irish Academy/Discovery Programme, Dublin, Ireland. 126 pp. Whitehouse, N.J., R.J. Schulting, M. McClatchie, P. Barratt, T.R. McLaughlin, A. Bogaard, S. Colledge, R. Marchant, J. Gaffrey, and M.J. Bunting. 2014. Neolithic agriculture on the European western frontier: The boom and bust of early farming in Ireland. Journal of Archaeological Science 51:181–205.