New Zealand to Easter IslandEaster Island, New Zealand
From Europe around the globe for ?1093! Okinawa, South Korea, New Zealand, Tahiti, Easter Island and Chile!
We recommend this astonishing route if you are a keen traveller! On your round the globe tour you will be visiting Okinawa, Seoul, Auckland, Tahiti, Easter Island and Santiago de Chile! A great chance to see six intriguing places in a wonderful voyage! Your first travel destination is Japan.
You' ll have 5 working day to discover the magnificent Okinawa-Archipel. It is a range of sub-tropical islets with one of the most attractive sea fauna and one of the most spectacular corals in the game. South Korea's capitol with more than 10 million inhabitants can be described as one of the largest towns in the atlantic.
The next plane takes you to an even more beautiful and remote location - Polynesia. Soak up the relaxing moments in Tahiti, enjoy a moist wind over the blue waters and explore the beautiful reefs, ideal for snorkeling, scuba divers and swim. Would you like to be even more impressive of your trip?
Their next stop is Easter Island! Dive and snorkel in one of the most remote places on earth, explore the world's most enigmatic sculptures and enjoy a whole weekend in this amazing area! Travelling dates: Keep in mind that the trip data is fully customisable so that you can eventually determine the trip data yourself.
Route and example data: Click HERE >> and go to the Jeju Air website or simply click on the link below and change your booking details. The three itineraries must be reserved separately: Reserving samples: Luggage during the trip:
Climatic window for Polish trips to New Zealand and Easter Island
The South Pacific migratory trails used in the Eastern Polynesia colonisation (800-1500 AD) were accepted as usually against the wind on the basis of an understand of contemporary climatic models. Instead, our novel paleo-wind reconstructed fields with bidecadic dissolution show that migratory trails from eastern Polynesia lie in lee during the known periods of New Zealand's first settlement and Easter Island.
These findings are important to show that a nautical capability was not indispensable in the Polyynesian colonisation voyage and that long-term variations in sail condition due to the extent of the tropical regions were important in determining the history of colonisation. The discussion on the early stages of man's migrations in the vast area of Eastern Polynesia has concentrated on maritime technologies, both in terms of navigational and canoeing skills, while less consideration has been given to the timing variations in sail condition, particularly due to the effects of climatic changes.
There is a pattern of the Polyynesian voyage that the pre-historic colonisation channels that travel eastwards and through Eastern Polynesia (CEP: Society, Tuamotu, Marquesas, Gambier, Southern Cook, Austral Islands) and to Easter Island currently dominate the Pacific Ocean. A similar argument was put forward for the CEP's trip to New Zealand against the predominant western wind countries.
Another opinion is that the migrations demanded dependable off-wind sailways. Investigating the ocean environment and possible itineraries during the Medieval Climatic Anomaly (MCA), 800-1300 AD, when the first colonisation of CEP and New Zealand took place. Palaeoclimate datasimulation is used to rebuild Pacific Ocean Floor pressures and windfield samples with bidecadic resolutions during the MCA.
It is argued here that varying windfield designs associated with the MCA have created circumstances in which travel to and from the most remote Eastern Polynesia, New Zealand and Easter Island was easily possible through off-wind winds. By 1140-1260 AD, the deepening and widening of the Pacific sub-tropical anti-cyclone opened an abnormal climatic frame for off-wind sail lines from the southern Austral Isles, the southern Cook Isles and the Tonga/Fiji Isles to New Zealand.
Archaeological, palaeoecological and language testimonies to colonisation in Eastern Polynesia indicate that CEP ( "Central Polynesia") (Fig. 1) was colonised around 1025-1120 AD by Samoa and the peripheral islands of Hawaii, Easter Island and New Zealand around 1190-1290 AD, New Zealand being accessed from the southern Cook or Austral Islands and Easter Island through the Gambier Islands (1, 2).
The argument is that long-distance transports later decreased and were practically discontinued until 1500 AD (4, 5). It is one opinion that the colonisation journey brought with it a certain ability to go upwind, if one assumes that the travel season was dominant for contemporary climatic models (4, 6?-8). This together with the belief that short-lived trade winds in connection with westerly winds or El Niño incidents may have supported the voyage is mirrored in the real and artificial operations of eastern Polynesia's experiential yachting canoe (, 4, 6??-9).
A comparison of early historic observation and language information indicates, however, that before exposure to the late sailing technique in western Polynesia, after about 1500 AD, eastern Polynesian twin-cannoes did not have a firm masts and were essentially limited to off-wind sections (10?-12). The Pacific Ocean with the largest part of the Polynesian colonization delta across the Middle East Pacific (CEP), the extratropic South West Pacific and the sub-tropical North Pacific, which was colonised to A.
Samoa and Tonga Islands were colonised in an early migratory sequence about 3,000 y B.P. (Inset) The area of off-wind marine orientation for Polyynesian yachting boats here is directly leeward plus up to ±30° to each side thereof. When available, the benefit of off-wind sections is that long-range (wind on both sides) or leeward sails avoid the mechanics of the updraft and the need to tack up to fourfold as far as possible to achieve the same goal (6, 9).
Shorts to the east were possible in west weather seasons, and long haul sections outside the breeze could have appeared if west breezes caused large sub-tropical tradewind in very strong El Niño weather but these were probably brief and rather erratic periods (5, 7, 13?-15). Whereas the propellers of Polish migrations are not known, any deliberate or systemic sail participating in the colonisation voyage, whether with ships with only one headwind or with greater capabilities, would have profited from multi-decadal displacements to favourable equitable prevailing winches.
This occurs when extraterrestrial and sub-polar currents travelling as high pressures travel from their central climatic positions in the sub-tropics (the highest point of hypertrophic pressures is at the sub-tropic ridge, STR) to the poles and weaken the predominant westerly zone wind in favour of meridionic wind (16).
Advanced re-enactments of Polyynesian travels (4, 6) show that apron winds are possible in the extra-tropical south-west Pacific when north-east trade winds substitute westerly winds in season in winter as well as early in the year. Marine Sea Surface Anomaly ( "SLPa") over the sub-tropical to extra-tropical southern hemisphere was recently restored for several decades during the medieval climate anomaly (MCA) (17).
MCA includes two climatic modes, 800-1100 A.D. and 1100-1300 A.D., each of which has a different abnormal weather in early and late summer in comparison to the contemporary one. Biennial to centenary climatic changes in the Pacific during the MCA can be described in relation to the latitude of the tropical regions and the mean state of the Pacific in relation to the duration or abundance of the two phases of El Niño-Southern Oscillation (ENSO).
Changes in the mean climatic condition are due to a variation in the time and incidence of the continuing summer-winter seasons and the windstorm rate. Shifting to more El Niño-like (La Niña-like) climates, for example, leads to a higher incidence of western (southeastern) wind in the southern hemisphere tropical regions and, on a decadiccale, depicts the appearance of El Niño (La Niña) perennial occurrences as observed in the last few years 1980-2000 (1950-1970).
Similarly, polar extension (equatorial contraction) of the tropical regions leads to more (less) common, quasi-stationary anti-cyclones in the sub-alpine and more (less) north-south (west-east) storms in the southwestern Pacific. Between 800-900 A.D. and 1000-1100 A.D., the century-old mean climatic patterns resemble a displacement to the El Niño patterns of the Central Pacific (Modoki), with southwestern windfields over New Zealand and abnormal western windfields (trade fair blips ) over the Central Pacific, in combination with a polar tropical landscape (17).
Between 900 and 1000 AD, the Pacific was marked by the El Niño model and the equatorial tropical regions, where the west side winds are in the west to northwest. Between 1000-1300 AD the Pacific trade winds were generally reinforced by the polar expanse of the tropical regions and the associated prolonged sub-tropical anticyclone, first via East and South Australia and the Tasman Sea before 1100 AD, then later via New Zealand and to the east.
Between 1140-1260 AD, the Pacific was marked by a displacement of the mean climatic conditions towards the Modoki La Niña model, together with an increase and polar extension of the cyclone subtropics (17). D. 1140 Samples affect CEP by cold ocean surfaces temperature (SSTs) and droughts in the tropics of the Pacific Ocean and the branching of trade winds and ocean current near the Austral and Tuamotu Islands (18).
The polar expanse of the tropical regions during the MCA therefore opened up an abnormal climatic frame for off-wind sail lines into the south-west Pacific extratropes, especially to New Zealand. After 1300 A.D., with the exception of a short time after the mid 1400s, our earlier climatic reconstruction ( "17") shows that no further ecadic climatic periods for dependable off-wind sail between the subtropical and the southwestern extratropes arose.
Comeback to the mid-latitude zones in the west and sub-tropical eastern trade has largely eradicated the climatic window for off-wind cruising on the relevant trade lanes, as a more stubborn El Niño-like weather and equatorial sub-tropics dominate the South Pacific ?A Bidecadic resolution of these climatic window was studied in this paper using the palaeoclimate procydata assimilative method (17) to additionally reconstitute ocean floor pressures and long time mean climatic conditions for 1300-2010 A.D. with regard to SAD.
Reconstructed for each current 20-year-old windows between 800 and 1600 AD that includes the southern winters (June, July, August) and springs (September, October, November) periods (materials and methods and SI appendix). Dataassimilation maintains modelled interviewable relations; therefore, both ocean floor pressures and windfield reconstruction are dynamic coherent.
In order to further explore possible routs for canoeing, we created an enveloping curve of sea breezes at each raster point of the restoration to help us determine the off-wind sail vectors (directly in the windward direction plus a ±30° spread to allow a long range). To investigate the significant sea breeze abnormalities for every 20 year period, we planned possible sail lines from the Pacific Ocean, which would have led to landings in New Zealand or Easter Island.
Bidecadic climatic re-construction makes several suppositions (see materials and methods and SI appendix) that lead to the following insecurities, e.g.: proximity datability, insulation of the proximity climatic signals from non-climatic interference, restricted geographic distributions of available proximity climatic information, constraints of the climatic models and the algorithms of association.
To minimise the uncertainty of dates, we have found a bidecadic timing within the estimates of the uncertainty of dates for the contained perxy-dates. Climatic squelch uncertainty is minimised by analysing the proxysignal intensity at each timing increment and only taking into account those which indicate a unique one. Reconstructing is partially limited by the geographical densities of the available proximity information for each of the windows, as the greater densities allow for more redundancy in localization.
Geographical distributions of the prospective X-ray images are shown in the SI Annex, Figure -1. Increased densities of paleoclimatic information reduce the uncertainty in the bidecadic climatic signals resulting from the intrinsic chronic ambiguity of the information or the ages of each paleoclimatic prop. Therefore, in this trial, the greater densities of palaeoclimate after A.
First MCA climatic window for off-wind sailings from CEP to Easter Island appeared with continuing west winds during 800-820 A.D., 830-910 A.D., 1010-1030 A.D., 1040-1060 A.D. and 1080-1100 A.D. (SI annex, Fig. S5 A and B), the last data coinciding with the first CEP colonisation (2).
Windfields from 830-850 AD and 860-890 AD show powerful south-east-east sail lines from Samoa to Easter Island, across the Gambier Islands or Austral Islands. The Easter Island sailways, which originated around 1250 AD from the Gambiers and especially from the Australians in 1250-1280 AD, were also made accessible by west winds (SI Annex, Fig. S5C).
The two groups of islands are likely starting points for Easter Island passageways, and both have colonisation estimations (Rapa Island in South Australia 1100-1200 A.D.; Gambiers 1108-1275) simultaneously with those of Easter Island (A. D. 1200-1253) (1, 2, 24). Nevertheless, a journey around 800-900 AD cannot be ruled out.
Interesting is that the windfield re-construction according to A. In this way, an earlier assumption is supported that the Polynesian travellers to the East could have been sailing in the volume S, which is experiencing a seasonsal transitions between south-east and west trade in the present weather (9). Also the Australian Easter Island tour was open:
The last travel window resulted from a tropical and subtropical contracture and an equatorial displacement of the westerly winds, which took place in conjunction with a displacement of the mean climatic condition towards continuing El Niño-districts. Reconstituted bidecadic mean surface pressures (black lines), surface pressures abnormalities (color, hectopascal) and the associated windfield abnormality vector (grey) for the seasons (A) A.D. 1080-1100, (B) A.D. 1140-1160 and (C) A.D. 1170-1190.
In addition, the 30° boundary of the off-wind sail markers (solid black) for Polynesia canoeing is shown for each winding vectra (see Fig. 1). Length of the aomaly and the associated off-wind sail vessels represent the Pacific's absolute velocity differential and are proportionate to the re-constructed atmosphere pressures.
Possible downwash paths for each climatic frame are indicated by the large grey arrow. It was possible to travel back from Easter Island to CEP when a climatic displacement repaired the eastern and north-eastern winds in the sub-tropical Pacific at 1090-1120 A.D. and 1200-1250 A.D. (SI Annex, Fig. S5 B and C).
An anti-cyclonal windfield in connection with the eastern migrations of the sub-tropical anti-cyclone eastwards of New Zealand and a reinforced STR further eastwards provided dependable trade winds and opened off-wind sail lines for the voyage back from Easter Island to CEP. Windfields in 1110-1130 A.D., 1200-1220 A.D. and 1230-1250 A.D. (SI appendices, Fig. S5 B and C) show reinforced trade winds between Easter Island and the Austral Islands, which create an off-wind path to the Australis.
The east-west sail would have been easy to sail on using latitude-starpathing. Another or extra assumption for the Easter Island colonization is the journey of Indians from South America. We have reconstructed off-wind sail lines to Easter Island of Central and Northern Chile in 910-930 A.D., 930-950 A.D. (SI Annex, Fig. S5A) and 1140-1170 A.D. (Fig. 2B), and 1220-1260 A.D. (Fig. 3 A and B).
This follows the equatorial leg of the subtropical anticyclone of the Eastern Pacific Ocean, in the Humboldt Current, which flies about 30°S to the north-west and then to the west in the direction of Easter Island. Possible ways back to Chile were open in 1260-1290 A.D. and closed with an increased southern current around 1300 A.D. (Fig. 3 C and SI annex, Fig. S5 C and D).
Reconstituted bidecadic mean surface pressures (black lines), surface pressures abnormalities (color, hectopascal) and associated windfield abnormality vector (grey), together with a 30 off-wind sail vector for the seasons (A) A.D. 1220-1240, (B) A.D. 1240-1260 and (C) A.D. 1300-1320. Possible downwash paths for each climatic frame are indicated by the large grey arrow.
The New Zealand sail route was already possible from 910-970 AD (SI Annex, Fig. S5A) and from 1140-1260 AD (Fig. L and 3B ), when the strengthening of the sub-tropical anti-cyclone opened up an abnormal climatic field for off-wind sailroute in the South-West Pacific extratropes. The climatic patterns led to three possible trips to the New Zealand region: from the Southern Australs, the Southern Cooks and Tonga/Fiji.
The Coriolis effect in the southern hemisphere, which redirects the sea currents to the south west of the sub-tropical sea winds, finally supported New Zealand's landing on the Southern Australs and Southern Cook itineraries. Between 940-970 and 1170-1230 A.D., in Tonga and Fiji, possibly even in Vanuatu or New Caledonia, prospective sail lines emerged from the sub-tropical Pacific to New Zealand.
2C ), off-wind sail paths from Tonga take a southern route before they encounter northeast breezes at 35°S, which are focusing potential landings along the eastern seaboard of South Island and southwards to the sub-Antarctic Auckland Islands. It worsened after 1240 A.D. and vanished after 1270 A.D. when the southern winter spring west breezes were re-established over the Auckland Islands.
New Zealand's oldest climatic frame for the journey of the South Australians is 910-930 A.D. to 940-960 A.D., which precedes the archeological estimations of the colonisation of the Austral Islands. The South Australian-New Zealand downwash was also open in 1140-1170 (A. D. 1140-1160 in Figure 2B) and 1200-1240 (A. D. 1220-1240 in Figure 3A); these seasons include recent archeological estimations of the colonisation period (2, 3, 13).
Canoeing westbound in the 27°S to 30°S latitude range may have become entangled in the powerful east-northeast locations of the fortified sub-tropical anti-cyclone over the southwestern Pacific. Between the southern Cook Islands and New Zealand there was a downwash at 940-960 AD (SI Annex, Fig. S5A) and from 1140-1160 AD, 1210-1230 AD, especially 1240-1260 AD (Fig. L2 B and 3B), with sustained eastern trade winds over the southern cooks approaching the date line in a northeast direction.
The majority of sail lines from possible South Cook or Australian origin concentrate on land fall opportunities on the north-east coastline of New Zealand's North Island, in accordance with Maori canoeing traditio. Returning from New Zealand to CEP called for southwest winds to the Northern Cooks and west winds in the latitude S to the Northern Australs.
The journey back to the Southern Cooks is limited to 1280-1300 AD, while the journey back to the Southern Australs has a longer windows from 1250-1270 AD and after 1300 AD (Fig. 3C) until the early 1600s. In 1270-1290 AD, these times were dotted with a possible opening for the Australians to travel back to New Zealand.
An equatorial displacement of the STR and the persistent seasons of the travelling anti-cyclones led to short-term possibilities for a three-way trip on the Bidecad map. Reconstruction of our windfield indicates that there were no other climatic window on the bicadal level for dependable off-wind sail between the sub-tropics and the south-west Pacific extratropes to A.
D. 1,600 (19?????-23), and other westerly winds zones and sub-tropical eastern businesses were recovered (SI annex, Fig. S5D). While this does not exclude that some years may offer favourable seasons for offshore winds between the sub-tropical Pacific and New Zealand, it emphasises that these would not be standard or dependable for decades.
Our windfield re-constructions are therefore in line with the archeological proof s that the long-distance journey to New Zealand was either stopped or was scarce after 1300 AD. To summarize, there were multi-decadal climatic window for off-wind travel during the MCA, (Fig. 4) from: (i) CEP to Easter Island between 800 and 900 AD, 1000 AD and 1100 AD, and during the time after Christ's birth.
1250-1280; (ii) the east edges of East Polynesia (including Easter Island) in a northwestern sense, especially 1140-1160 A.D. and 1180-1250 A.D.; and (iii) CEP to New Zealand in 910-960 A.D. and 1140-1260 A.D. All of the window from 1100-1300 A.D. correspond to the actual archeological and related clues. Supplementary window frames between Tonga/Fiji and New Zealand during 940-970 A.D. and 1170-1230 A.D. offer the opportunity to investigate interrelationships that are otherwise generally neglected.
We also show that the voyage back was possible at regular intervals by off-wind cruising in every sense, and we are documenting a broader spectrum of off-wind cruising options than currently foreseen, which include between the Australians and Easter Island, between Fiji or Tonga and New Zealand, and between Easter Island and extra-tropical Chile.
Possible Polish itineraries to the CEP and the extra-tropical southwestern Pacifc post-A. A. D., 1010-1030 A. D., 1040-1060 A. D., 1080-1100 A. D. and 1250-1280 A. D. The green trails were open. More southerly itinerary from the Australian Isles was open from 1290 AD to 1440 AD, 1500-1540 AD, 1550-1570 AD and 1590-1610 AD.
The way back on the dark and dark trails was open in the years 960-990 AD and mainly from 1260 AD to 1550 AD. During the time of the Eastern Polynesia colonisation, our newly restored sailings would have made it possible to negotiate all known colonisation trails and others, sometimes archeologically, with unwind-up mayoes.
It is our point that the climate proofs suggest that an updraught capacity was not necessary for the investigation and settlement of the isolated eastern Polyynesian isles during the archaeological time. Furthermore, the lack of efficient wind-up capacity could have been important for the reduction of long-distance cruises in Eastern Polynesia after the sail window discussions here were over.
The results indicate that recent modelling of marine engineering in Polynesia during the MCA (27, 28) and the simulation and experiment journey need to be considered. It is our conclusion that climatic changes offered a great chance for Polynesia's offshore migrations during the Late Holocene. The mean surface water table pressures (SLP) and windfields for the South Pacific Ocean basins were restored with a 20 year dissolution from 800-1600 A.D. using an establish and previously described (17) palaeoclimate datassimilation method.
We are building on other current approaches (23, 29) by using the combination of a multi-variate IP networks to choose climatic condition analogues from an established group. In this section we describe the standardisation of process and modelling information, the analogue choice and the computation of recovered SLP and windfields.
SI Annex contains: the SI protocol (SI Annex, Fig. S1 and Table S1), the results of a dummy protocol (SI Annex, Fig. S3), the protocol used to record each interval in Fig. 2 and 3 and SI Annex, Fig. S4, and the bi-kadic interval reconstructed in the results and discussion of A.D. 800-1600 (SI Annex, Fig. S5 A-D).
Cheatcodes. Multi-variate and multi-variate database contains single sets of proxies, reconstructed and released local multi-proxy reconstructed (SI attachment, Fig. S1 and Table S1). Prior to assessment, all annual resolution was converted to decadic value, decadic resolution was used without resampling, and discreet temperature information was set to decadic value.
In order to make it easier to compare different types of progressive oxide datasets, all decadic datasets were standardized with respect to the long-term mean 1300-2000 AD. Mean temperature signals for each of the probes were computed for each individual 20 year interval. Recognizing the high signal-to-noise associated with proximity sensing signals, each 20-y cycle was rebuilt using only those signals that indicate a unique environmental condition sign (SI Annex, Figure S4), as the 20-y mean value that exceeds ±0.
For each 20 year term, the standardized value for each remaining procxy was summarized into a single array (P) (SI Annex, Fig. S4). A 10,000-year-old Holocene controller used in this report is the Commonwealth Scientific and Industrial Research Organization (CSIRO) Mk3L climatic system modelling 1.2.
is a fully-linked, high-resolution spectral solar thermal simulation tool developed specifically for millennium scales (30). Atmospherical components of Merc3L are a mathematically effective versions of the atmospherical components of the Merc3 models used in the World Research Programme Code of Intercomparison Models Project Phase 3 (WCRP-CMIP3) and the Intergovernmental Panel on the Fourth Assessment Report on the Fourth Assessment of Change (30, 31).
It is a fairly good simulator of a contemporary environment, which includes a real engineered environment, albeit with some of the distortions described in ref. Mk3L also provides a real-world amplitudes and space properties simulations of Southern Annular Mode and Pacific South American Mode 1 and 2 (17). We used a 10,000-year simulated Holocene weather forecast with consistent parameters:
In addition, an A.D. 800-2000 Mk3L Tranverse Simulator with re-constructed sun (33), volcano (34) and CO2 (35) is used as a destination for pseudooproxy analysis (SI Annex, Figure S3). In order to be assimilated, both models and probabilities must be in a standardised form. Therefore, the models are presented by a series of normalised yearly mean time series (S) based on the same variable and location as the P propositions.
Palaeoclimate data assimilation. The term refers to a standard width (i) of standardized procxy levels that concisely describe the climatic condition for a given multi-decadal span of it. S1-i is an arrays of timelines (latitude = i and longitude = n) deriving from the modelled climates at the corresponding geographical places and climatic parameters as members of P2. Each S1-i field equals each P1-i field, and each S1-n field equals one member of the cast group ( "one year").
Averaging the 50 BMA is used to determine the climatic condition for the desired timeframe. Each modelled tag can be broken down by composite of the 50 BMA-emble mean; however, only modelled tags with a mechanical relation to the data set of the protocol should be safely interpret. Abnormalities are computed in relation to the full 10,000 year cycle and thus constitute variations from the modelled Holocene climatic conditions and are not directly indicative of variations from the CEP.
Reconstruction of spatio-temporal areas for the period of times mentioned under Results and Discussion are shown in the SI Annex, Figure S5 A-D. The methodical analysis using pseudo oxide shows that the method should precisely recreate the SLP areas for most of the South Pacific and Southern Ocean (see SI Annex, Fig. S4).
Reconstructing " a realistic climatic modelling using a synthesized climatic map; since the climatic condition to be reconstructed is already known, the ability of the reconstructive method can be demonstrated (according to refs. 36 and 37). A 800-2000 Mk3L simulated atmosphere with re-constructed solars (33), volcanoes (34) and CO2 (35) is known.
Pseudo proxies were deduced from the simulated models using the same set of parameters and places as the pseudo-proxies shown in the SI Annex, Table1. Then the 1200-year old modelling simulations were rebuilt with a 20 year dissolution using the multi-variate asymmetry method to choose climatic condition analogues from the casual 10,000 year old simulations (see methods) which are only basing on the climatic signal from the dummy propxy-link.
S3 shows halftone point relationships between modelled and rebuilt SLP over 1,200 years at a 20 year dissolution; it gives an idea of the capability of the reconstructive method in the present IPN ("SI Appendix", Fig. S1 and Table S1). Genuine SLP and windfield reconstructions should be understood with the utmost certainty in those areas characterized by significant positives correlation.
Reconstructing the image achieves a higher relevance for times with higher geographical densities of the prospective datas. Research on climatic change is a part of the Australia Eastern Seaboard Climatic Change Initiative. I. D. G. researched; I. D. G. and S. A. B. researched; I. D. G., S. A. B. and A. J. A. analysed dates; I. D. G., S. A. B. and A. J. A. penned the papers; and I. D. G. and S. A. B. sketched numbers.