






































Microsoft Word - 08mcnally.doc


 History of Meteorology 2 (2005) 105 
 

 

 

 

 

 

 

 

 

 

Reconstruction of Late 18
th

 Century Upper-air Circulation 

Using Forensic Synoptic Analysis 

 

Louis K. McNally, III 

 
L.K. McNally & Assoc., Inc., Portland, Maine, USA 

 

 

Introduction 

 

In this study I attempt to develop an outline of the general atmospheric flow around the 

time of 1785 C.E., in North America.  The results of my reconstruction of daily weather in the 

year 1785 in the northeastern part of North America imply a relatively cold flow over central and 

eastern North America for most of the year (McNally, 2004). The source of the colder air is to 

the north and west, and I begin with the assumption that a north to northwest flow from north 

central Canada was prevalent for most of the year. By comparing proxy data from other locations 

around the world to identify the location of the polar front (or edge of the polar cell), it is 

possible to outline a general flow around the northern hemisphere resulting in one similar to that 

theorized by Lamb (1977), and described as a “short-circuit cross-polar” flow, around the edge 

of a displaced and possibly larger polar cell.  

This flow suggests storm tracks that would provide the precipitation necessary for the 

development of the Laurentide and Scandinavian ice sheets. Storms would travel up the east 

coast of the United States and recurve northward into eastern Canada. An additional track would 

supply the moisture for northern Europe. This circulation, also independently described by Flohn 

(1969), is similar to that which may have been prevalent at the onset and during the dissipation 

of the most recent glaciation (Figure 1). 

Wexler (1956) has suggested that a hypothetical reduction of 20% in insolation, whether 

from reduced solar output or from volcanic effects, would result in a circulation characterized by 

a broad trough over eastern North America. Lamb (1977) notes that this is similar, although 

weaker, to the circulation around 1800 C.E. Both are characterized by a strong and persistent 

trough in eastern Canada, extending down the east coast of the U.S. This would result in a 

prevalent cold flow to the northeastern United States from Canada. These patterns may be 

representative of a transitional period between major warm and cold regimes in the northern 

hemisphere. 

Studies covering the 1780s are sparse, and concentrate on the possible cooling from 

volcanic effects in 1783-1785. However, valuable data can be extracted to arrive at the outline of 

the prevalent flow by following the cooling signal, or lack of it, in these proxy studies. 

Additionally, numerous analyses exist focusing around the year 1816, following the eruption of 

Tambora, some of which do extend back to 1785. The dust veil index (DVI) for the 1816 time 



 Reconstruction of Upper-air Circulation 106 
 

 

period is very high, as is that in 1785 (3000 and 1000 respectively vs. 1680-1974 C.E. average of 

475, Lamb, 1972). Some additional inferences may be drawn from and comparisons made to the 

1816 data. Both 1785 and 1816 followed moderate (M+) ENSO events, and the differences 

between the resultant effects also suggest the presence of a generally cold cross-polar flow in 

1785. 

Most of the cooling effect from Tambora in 1815 occurred in the year immediately 

following the eruption. Mid-latitude effects from the eruptions in 1783 appear to have lingered 

for two years (Kington, 1992). Generally, although the decade of the 1780s was already colder 

than normal in many areas, the cold cross-polar flow appears to be amplified by the effects of the 

1783-1785 high-latitude volcanic events in both Iceland and Japan. This amplification is also 

supported by the findings of Quinn and Neal, outlining the possible combination of effects from 

the El Chichon eruption and the El Niño of 1982-1983 (Quinn and Neal, 1983b). It would 

appear, then, that additional forced cooling from reduced insolation when occurring in a 

generally cooler period might trigger the cross-polar flow. 

 

 
Fig. 1. Short-Circuit Flow, from Lamb, for the onset of the Würm/Wisconsinan glaciation.  Thin line denotes 

average 5300 geopotential meter surface to 500 mb thickness for January. Short arrows denote prevailing 

surface wind. Broad arrows indicate storm tracks. 

 

 

 



 History of Meteorology 2 (2005) 107 
 

 

It would also appear that this cross-polar flow “short-circuits” the normal general 

atmospheric circulation, particularly at polar latitudes above 60° north. This flow develops brief, 

yet strong polar outbreaks at the southern end of the cross-polar route in Canada and the eastern 

United States. This further supports the finding of split-flow in the jet stream that may, in fact, 

sometimes have three branches. An expanded or displaced polar cell, evidenced by a polar front 

further south than normal in North America, may take on very different circulation 

characteristics from the usual, or normal, polar cell of today. These differences are most 

pronounced in North America, the North Atlantic Ocean, western Europe, and central Asia. The 

preferential cooling in these areas may point to a displacement of the polar cell to those areas, as 

opposed to a wholesale expansion. I begin the reconstruction, then, with the assumed northwest 

cold flow in Canada, west of Hudson Bay. 



 Reconstruction of Upper-air Circulation 108 
 

 

 
 

Fig. 2. Data Points for Reconstruction of Prevalent Flow in 1785, C.E. See Table 1 for data types used. 

 



 History of Meteorology 2 (2005) 109 
 

 

 
Table 1. Data points, sources, and descriptions 

 
Data 

Point 

Source Description 

1 McNally Documentary and observational data forensic synoptics  

2 Baron Documentary and observational data content analysis 

3 Crowe Documentary evidence triangulation and extrapolation 

4 Fayle Radial and height increment tree growth analysis 

5 Lough Tree ring chronologies, regression analysis 

6 Luckman and Colenutt Latewood marker tree ring chronologies 

7 Fritts and Shao Standardized tree ring width measurement, spatial arrays 

8 Alt, et al Isotope and melt percent values, synoptic analogues 

9 Lamb Monthly mean MSL from observations 

10 Ogilvie Documentary evidence 

11 N.H. Gazette and Pfister Documentary evidence, newspaper, diaries 

12 Schove Tree ring analysis 

13 Moberg Historical observations 

14 Wood Historical observations 

15 Bednarz and Trepinska Width and maximum density tree ring studies  

16 Serre Bachet, et al Tree ring width 

17 Serre Bachet, et al Dendroclimatological and documentary reconstruction 

18 Roy Documentary evidence, Delhi wheat prices,  

19 Wang and Zhang Documentary evidence, Qing Yu Lu 

20 Tsukamura Synoptic analogues 

21 Mikami and Tsukamura Documentary evidence, freeze dates, diaries 

22 Huang Jiyaou Documentary evidence, dryness/wetness grades 

23 Borisenkov Documentary evidence, religious chronicles 

24 Meeker and Mayewski Ice core evidence, ion series 

 

North America 

 

In addition to my results, content analysis by Baron and Gordon (1985) shows that the 

early 1780s in eastern Massachusetts, in general, exhibit shorter growing seasons, more winter 

days with fair-sky conditions, more summer days with thunderstorms, and more winter snowfall 

days than the remainder of the decade. Both shorter growing seasons and more winter days with 

fair-sky conditions indicate a prevalence of clearer Arctic or polar air masses in both summer 

and winter. The variability and vigor of the atmospheric circulation at other times of the year can 

be deduced from the more frequent summer thunderstorms, indicating an active summer Alberta 

storm track. More winter snowfall days would indicate more frequent and slower-moving winter 

storms. Generally, when not under the direct influence of polar air masses, the northeast could be 

construed to be near the polar front, its southern edge determined by the storm tracks in eastern 

North America. These results concur with mine. 

Crowe (1992), in his reconstruction of temperatures for Toronto, Ontario, Canada, finds 

significant effects from the volcanic activity in the summer of 1816, but very little variation 

during the 1783-1785 time period. Mean July temperature in 1816 (16.1°C) is more than 3° 

below the 50-year running mean, while 1783-1785 are all within 1° of the 50-year running mean. 

The similarities in the July mean temperatures of 1783-1785 would indicate a persistent pattern, 

consistent with the interior of a cold air mass as opposed to the dramatic effects in 1816. Again, 

assuming that the polar front is generally south of the area in 1785, the effects of the weaker 



 Reconstruction of Upper-air Circulation 110 
 

 

volcanic activity would not be expected to appear in data from so far inland, inside a continental 

air mass. Further north and west, and deeper into the continent, at Churchill, Manitoba, some 

cumulative effect on tree growth after Tambora is noted with white spruce (Picea glauca) and 

tamarack (Larix laricina) studies (Fayle, et al., 1992), but no major effect is found, further 

supporting the presence of a continental air mass, which is assumed to be present in 1785. 

Tree ring chronologies for the western part of North America developed by Lough (1992) 

show no large deviation from a long-term (1602-1960) mean. As the chronologies are 

reconstructed from sites at middle latitudes in the western third of the continent, this result is not 

surprising. In fact, the continentality expected from these interior sites in the southwestern part of 

Canada and the western USA, within and east of the Rocky Mountains, would prevent or at least 

dampen the detection of any signal from the east or north. These areas are under the effects of 

semi-permanent continental pressure patterns and away from the assumed cold northwest flow 

east of the Rocky Mountains, which marks the westward extent of the polar cell. They would not 

necessarily exhibit the sensitivity that would be expected in locations more proximate to the 

cross-polar flow, such as the area further north in central Canada or in northeastern North 

America. 

Luckman and Colenutt, (1992) conducted additional studies in the high-elevation 

Canadian west, above 50°N. Their results support the assumption that this region is out of the 

cross-polar flow, and influenced more by a flow from the Pacific Ocean. Although there is some 

reduction in radial tree growth, which could be attributed to a general cooling, wide variation is 

found in the mid-1780s. This would be consistent with an onshore flow from the Pacific; any 

stabilizing continental effects would be expected further east and south. At these higher latitudes, 

the effects of this stronger zonal flow would be exhibited here in the Canadian Rockies as 

compared to the region near the United States-Canada border. As they are not, thus ruling out a 

zonal flow extending into the center of the continent, a strong cold northwest flow in central 

Canada may be additionally supported, while still implying a general flow from the Pacific. 

In the western United States, Fritts and Shao (1992) reconstructed temperature and 

precipitation from a variety of spatial arrays of sites. These areas included the Columbia Basin, 

the California valleys, intermountain basins, southwest deserts, the northern high plains and the 

southern high plains. Throughout the entire area, temperature and precipitation data for the 

period from 1750 to 1800 show very little variation from the norm, with the exception of the 

temperature reconstruction for the high plains data set. There is a colder than normal period 

which stretches from 1770-1790 and it is found only in the high plains data set. In the absence of 

any other large deviations over the time frame, one may assume that persistent cold affected only 

the high plains, which is consistent with a steady northwest flow from northern Canada. The 

southern high plains are apparently unaffected, which supports the cold northwest flow from 

central Canada curving eastward towards the Great Lakes and the northeastern United States, and 

south of Toronto. This western edge of the circulation in 1785 is also consistent with Lamb's 

suggested flow. 

 

Arctic and Iceland 

 

In a cross-polar flow, Arctic locations in northern Canada would be directly in the path of 

continental air masses migrating across the North Pole from Asia. In any flow, evidence of 

volcanic activity would be found in high Arctic ice. Indeed, a marked signal is present in ice 

cores from the Agassiz and Devon sites in the high Canadian Arctic (Alt, et al., 1992). However, 



 History of Meteorology 2 (2005) 111 
 

 

similarities in the effects from both the 1815 and 1783-1785 eruptions show that the period of 

both was one of general cooling in the Arctic. Brief episodes of further cooling followed both 

eruptions as well. Circulation patterns following each, however, are different. 

Alt et al, (Alt, 1985, 1987, Alt, et al, 1985), using previously developed analogues, 

further conclude that the general synoptic pattern around the time of the Tambora eruption is 

similar to that in 1972, as characterized by the presence of a strong 500 millibar (50kPa) vortex 

centered near northern Greenland (Figure 3). No analog is presented for 1785. The western edge 

of this circulation is in a similar location to that of the 1785 cross-polar flow. The eastern side, 

however, shows a return flow poleward from the central North Atlantic Ocean. In contrast, my 

results indicate that the circulation in the 1780s appears to continue eastward as a strong zonal 

flow across the Atlantic. In the case of 1785, the center of the vortex (Icelandic Low) would 

most likely be depressed further southward and expanded eastward. This may explain Alt's 

findings, where the circulation pattern after the Laki eruption was different from that of 

Tambora. This expansion and eastward shearing of the low pressure vortex will also be identified 

in the European data discussed below. 



 Reconstruction of Upper-air Circulation 112 
 

 

 
 

Fig. 3. Summer Circulation of 1972, analogous to 1816, from Alt. 

 

Ogilvie (1992) studied sea ice records for the area around Iceland and shows that on a 

decadal scale the 1780s contained the greatest amount of sea ice on record. The series extends 

back to 1501. There is a positive correlation between sea ice incidence and temperature 

(Bergthorsson, 1969, Ogilvie, 1981, 1984a), and the cold regime thus inferred could be caused 

by an increase in the flow around the semi-permanent high over Greenland and an extended low 

pressure area in the Atlantic Ocean. It is possible that a Rex block may have developed, similar 

to the one in Figure 4, but displaced northward. This circulation pattern agrees with the flow 



 History of Meteorology 2 (2005) 113 
 

 

around the high pressure area over Europe as well, and can help explain the mechanism by which 

northern Europe is both colder and drier than normal. 

 

 
 

Fig. 4. 500 mb (50 Kpa) Circulation in mid-March, 2002.  Note Rex block (high north of low) in eastern Atlantic, 

forcing split flow upstream. 

 

Atlantic Ocean and Europe 

 

Lamb (1992) refers to a pattern similar to that in 1785 in reconstruction of pressure 

patterns for the summer of 1816 over the Atlantic Ocean and western Europe. Broad zonal flow 

north of 40°N to 50°N in the Atlantic Ocean is driven by broad low pressure from 50°N to 60N. 

This also agrees with Alt, et al and Ogilvie above. The increased pressure gradient between the 

(Icelandic) low pressure to the north, which is displaced southward and sheared eastward from 

the normal position near Greenland and the subtropical (Bermuda) high to the south drives a 



 Reconstruction of Upper-air Circulation 114 
 

 

strong flow from central and eastern North America across the Atlantic over a flattened ridge. 

This would also force the north side of the subtropical (Bermuda) High in the Atlantic to be 

sheared eastward, increasing the advection of more moist maritime air into southern Europe 

(Figure 5). The implication for stronger surface winds in the ocean under the core of the flow is 

borne out by observations from ships of being “blown off the coast” and unable to make ports in 

North America before supplies ran out (Bermuda Gazette, 1785). 

 

 
 

Fig. 5. Reconstructed flow for July, 1816, after Lamb.  Dashed lines inferred by Lamb. 

 



 History of Meteorology 2 (2005) 115 
 

 

High pressure in northern Europe is larger in this flow, which , if present in 1785, may 

explain the very hot, dry summer in England (Fowle's Gazette, 1785), as well as conditions that 

are drier and warmer than normal in Switzerland (Pfister, 1978, 1992), and drought in Finland 

(Schove, 1954). The northern edge of the flow continues across northern Europe, and then curves 

poleward, aided by a strong high pressure ridge in central Asia. Arctic air masses originating in 

central and eastern Asia then follow the flow eastward and then northward, eventually crossing 

over the Arctic, and completing the short-circuit cross-polar flow. A split in the flow from its 

southern edge into west-central Asia is implied, with an eastward displacement of the subtropical 

high over the Asian subcontinent supplying the moisture for storm tracks to the east. 

Northern Europe, under the influence of cooler and moist air transported rapidly across 

the North Atlantic Ocean could be expected to be cooler, on average, with similar variability as 

that which is noted in the northeastern part of North America.  The average temperatures as 

reconstructed for three stations in central England, Uppsala, Sweden, and De Bilt, the 

Netherlands, do indicate that a cooling period began by 1780, reaching a minimum in 1785 

(Moberg, 1996). Wood (1992) notes that average January temperatures for six European cities 

(Stockholm, Copenhagen, Edinburgh, Berlin, Geneva and Vienna) are below the 31-year normal  

in 1780 and 1781, above normal for 1782 and 1783, well below normal (greater than 1 ) for 

1784, and near normal for 1785 and 1786. The dry summer in England and cooler July readings 

in western and northern Europe can be explained by the presence of a persistent high pressure 

ridge over Britain and the Baltic, the eastern side of which allows cold advection into Europe 

proper. 

Further south and east in Europe, the general cooling noted above for the decade is seen 

in tree ring studies from the Alps and Tatra mountains (Bednarz and Trepinska, 1992). The 

general cooling is evident through the decade of the 1780s, and is most likely the result of 

summer outbreaks of the cold air on the eastern side of the high pressure ridge over Britain and 

the Baltic before its return northward in eastern Europe. The sensitivity to increased variation as 

well as a more pronounced cooling trend is evident more in the Tatra mountains than in the Alps, 

implying that the Tatras in Poland are closer to the edge of the flow, and thus closer to the edge 

of the polar front. Switzerland and central Europe remain under the influence of the expanded 

continental high and implied upper ridge. 

Dendroclimatic reconstructions from tree ring widths in southwestern Europe and 

northwestern Africa for the 1780s (Serre-Bachet et al, 1992) help to define the southern stream 

of the split flow, south of the Rex block, from the eastern Atlantic Ocean. Reconstructions of the 

temperature for eastern France and Rome indicate cooling, which would support a cold flow 

from off the eastern side of the northern European ridge. Precipitation records do not extend 

from the present back to 1785. Serre-Bachet, et al also analyzed reconstructions of temperature at 

four grid points on the Jones (Jones et al., 1985) network. For 1785, the English Channel grid 

point (50°N, 0°W) shows little variation from normal, although colder temperatures are indicated 

in the ensuing decade. The Mediterranean grid point (40°N, 10°E) shows a 7-year cooling trend 

is already underway, reaching a minimum in 1785. The grid point in southern Poland (50°N, 

20°E), shows rapid cooling over a 6-year period, while the grid point in eastern coastal Spain 

(40°N, 0°W) is near the maximum of a 12-year rise which peaks in 1783. 

It is evident from the tree ring studies in the Tatra Mountains above that the cooler flow 

in Poland can be supported. The English Channel area, away from both the northern and 

southern branches of the split flow, is less affected by anomalous conditions, which is consistent 

with a ridge overhead. Precipitation data from Morocco, which is the lowest (driest) in the record 



 Reconstruction of Upper-air Circulation 116 
 

 

from 1500-1975, and the warmer temperatures in Spain could support either the eastward 

displacement of the subtropical high or, more likely, of an African ridge extending northward 

and westward. In either case, a strong flow across the Atlantic, curving both north and south of 

Britain, can be inferred. 

 

India and Asia 

 

In India, regular weather observations do not commence until 1792, but wheat prices in 

Delhi (Roy, 1972) do show a marked increase in 1782 and 1783. The immediate cause may be 

tied to ENSO activity and a failed monsoon. This might be affiliated with a northern hemispheric 

circulation anomaly, in that a reduced flow from the west, part of which has turned poleward, 

might allow a stronger high and ridge to develop over the Indian subcontinent. A split in the flow 

would be implied. However, an eastward migration of the monsoon, causing the spike in wheat 

prices would be consistent with both an El Niño event and the eastward displacement of other 

pressure systems in eastern Africa. There is no corresponding anomaly in wheat prices in 1810-

1820 (Pant, et al, 1992).  

A split flow can be inferred from ion-concentration studies of GISP-2 ice cores (Meeker 

and Mayewski, 2002). They constructed sea level pressure anomalies for winter (December, 

January and February) and spring (March, April and May) during the Little Ice Age (LIA) 

(1400's to 1800's C.E. in this study) from a proxy series extending back 1400 years BP. Results 

do show reduced pressure across the north Atlantic and into south central Asia during the LIA. 

This would support both a southern split from the cross-polar flow and an eastward migration of 

high pressure systems over the Indian sub-continent. 

In a very broad sense, Borisenkov (1992), using documentary evidence, finds that the 30-

year seasonal averages in mid-Russia show little difference from normal for the autumn, winter, 

and spring, but higher than normal readings for the 1780-1800 period. This nearly 1°C anomaly 

is only slightly lower than that reconstructed for the 1800-1830 period, which is the highest in 

the entire 1501-1900 record. Warmer conditions in the summer would be consistent with the 

prevalence of continental air masses that may be advected northward in the cross-polar flow, 

modifying (cooling) over the pole, and then arriving in North America as cold outbreaks in 

summer. 

 

China, Japan and the Pacific Ocean 

 

Generally wetter weather in eastern China can be noted at Nanjing and Suzhou in 1785 

from the Qing Yu Lu or the Clear and Rain records (Wang and Zhang, 1992). References to the 

year 1816 indicate severe flooding in eastern China as well, as a result of an eastward retreat of 

the subtropical high, (Huang Jiayou, 1992). The Yangtze and Yellow river basins would be 

affected in both cases, and a parallel might be drawn to the 1780s. However, in 1816, the polar 

front in Japan shows no change from normal, and, in fact, the summer was warmer and longer 

than normal, implying a more prevalent meridional flow (Tsukamura, 1992). In 1785, increased 

zonal flow and eastward migration of regular pressure patterns, including the displaced 

monsoonal flow over India, is already assumed. The increased rain in 1785 may have come from 

tropical cyclones hitting land further south than normal. 

Polar air masses from central Asia would be steered towards the pole in a cross-polar 

flow, prior to arriving at the east Asian coast. Continuing in the cross-polar flow, these 



 History of Meteorology 2 (2005) 117 
 

 

continental air masses would be directed southward from the Arctic, and become the source for 

the notable cold outbreaks in eastern North America. Any expansion of cold air masses in Asia 

in the summer, however, might affect Japan, which is on the eastern edge of this flow. This 

would be expected to be found in cooler summer conditions, without any attendant increase in 

storminess. In fact, during the 1780s, and from 1783-1785 in particular, Japan did suffer 

extraordinarily cool and wet summers, leading to famine conditions from poor rice harvests 

(Mikami and Tsukamura, 1992). Lake Suwa, Japan, froze 22 days earlier than the decadal 

average, which is earlier than any other year in the series from 1772 to 1794 (Wood, 1992). 

The fact that there is evidence for zonal flow in eastern Asia and Japan in 1785 would 

also point to a more sensitive response somewhere in the western part of North America. A 

strong zonal flow over the northern Pacific, unlike the meridional flow established for the decade 

of 1810-1820 above, could be then be responsible for the signal from high Rocky mountain trees 

in the southwestern Canadian study cited above (Lough, 1992, Luckman and Colenutt, 1992). 

This does not preclude a quasi-meridional flow over a flattened ridge in the Pacific, similar to 

that in the Atlantic, but does imply an onshore flow from the west in the Canadian Rockies. 

Thus, the circulation pattern of the entire northern hemisphere can be outlined, showing 

the “short-circuit” cross-polar flow, excluding those areas which would not be directly affected, 

such as southeastern Asia westward to the Indian Ocean. The efficiency with which this flow can 

drive polar outbreaks from the continental source regions of east and central Asia, the Arctic, and 

northern Canada, eventually arriving in northeastern North America, then appears viable. 

 

Tropics and Southern Hemisphere 

 

Evidence of a more pronounced zonal flow across the Pacific Ocean at lower tropical 

latitudes might be found in the increased total particulate deposition and decreased 
18

O values 

in the Quelccaya ice cap in Peru. This is observed for the first half of the decade, with maxima at 

1783-1785 (Thompson and Mosley-Thompson, 1986). In the southern hemisphere, data are 

sparse, but tree studies of sub-tropical to sub-Antarctic species in South America (Villalba and 

Boninsega, 1992) do show decreased temperature, perhaps in response to the Laki-Asama 

events, after 1783 at 37-39° south latitude for a one year period, very little, if any response at 41° 

south latitude, and a pronounced response three to seven years later at 54° south latitude. It 

would appear that more data and research are required to identify the propagation of a signal 

through the prevailing westerlies in the stronger global circulation pattern in the southern 

hemisphere. 

 

ENSO effects 

 

Quinn's (1983b) research does indicate an amplification of anomalous patterns when 

ENSO and volcanic events occur simultaneously. With the results below, combined with the 

disconnect from the main flow noted above, it is much more reasonable to assume that the cross-

polar flow would exist with little direct connection to ENSO activity. The years preceding 1785 

did show increased ENSO activity, with which with many weather events worldwide can be 

associated. 

Deficient floods in the Nile River are associated with ENSO activity 80.3% of the time 

from the mid-1500's to today (Quinn, 1993a), and in strong events, the correlation is higher. 

Record drought in Chile, a characteristic of Little Ice Age climate patterns (Quinn and Neal, 



 Reconstruction of Upper-air Circulation 118 
 

 

1992), is in evidence through the beginning of the 1780s. The drought conditions are broken only 

by the very strong ENSO event of 1782-1784. Another established teleconnection with El Niño 

and the northeast is a warmer winter (United States Dept. of Commerce, 2002). As this was 

certainly not the case in 1785, this can give more weight to the strength and persistence of the 

cross-polar flow. 

Additional teleconnections associated with El Niño activity include drought in eastern 

and northern Australia, an east monsoon drought over Indonesia, and drought in northeast Brazil. 

Quinn has found that southern California also experiences higher rainfall during El Niño years 

about 87% of the time (Quinn, 1993a). These effects, although expected, could not be proved or 

disproved with the data used in this paper. 

There is a general suppression of tropical storm activity in both the Atlantic and Pacific 

Ocean during El Niño years (Dong, 1988, Gray and Schaeffer, 1991, O'Brien, et al, 1996). This 

reduction in the number of storms does not preclude hurricane development. In fact, one might 

expect to find fewer, stronger and faster moving storms due to the more vigorous steering 

currents at upper levels of the atmosphere, and the larger temperature gradient between the 

colder-than-normal ocean waters to the north and the warm tropics. Dixon points out that there 

are fewer hurricanes in an El Niño year, yet the frequency of landfall of those that do occur in an 

El Niño year is slightly higher (Dixon, 1996).  

In summary, the climatic conditions associated with the very strong (VS) El Niño of 

1782-1784 and the moderate (M+) event of 1785-1786, can help to develop the picture of 

northern hemispheric circulation in 1785. The tropics and areas affected directly by the El Niño 

appear to behave independently of the high latitude westerlies and cross-polar flow, exhibiting 

the hallmarks of a cooler climatic regime resembling the Little Ice Age. This reconstruction, with 

the apparent disconnection between the cross-polar flow and tropical southeast Asia, raises an 

interesting question for future research: Did the eastern displacement of the pressure and wind 

systems from the Indian Ocean to eastern Asia allow the polar cell to drift toward North America 

and the Atlantic Ocean, or was the preferential movement of the polar cell to those areas the 

forcing factor? Figure 6 represents the reconstructed complete short-circuit cross-polar flow for 

1785, C.E. 



 History of Meteorology 2 (2005) 119 
 

 

 
 

Fig. 6. Reconstructed short-circuit cross-polar flow for 1785, C.E.  

 

 



 Reconstruction of Upper-air Circulation 120 
 

 

References 

 

Alt, B.T., 1987. Developing synoptic analogues for extreme mass balance conditions on Queen 

Elizabeth Island ice caps. J. Climate Appl. Meteorol. 26 (12), 1605-1623. 

Alt, B.T., R. M. Koerner, D. A. Fisher and J. C. Bourgeois, 1985. Arctic climate during the 

Franklin era as deduced from ice cores, The Franklin Era in Canadian Arctic History, Pat 

Sutherland, ed. Natural Museum of Man, Mercury Series, Archaeological Survey of 

Canada Paper 131, 69-92. 

Alt, Bea Taylor, Davis A. Fisher and Roy M. Koerner, 1992, Climatic Conditions for the Period 

Surrounding the Tambora Signal in Ice Cores from the Canadian High Arctic Islands, in 

Harington, ed., Year Without a Summer, 309-327. 

Baron, William R., and Geoffrey A. Gordon, 1985, A Reconstruction of New England Climate 

Using Historical Materials, Syllogeus 55, Climate Change in Canada 5, Critical Periods 

in the Quaternary Climatic History of Northern North America, C.R. Harington, ed., 

National Museum of Natural Sciences, National Museums of Canada, Ottawa. 

Bednarz, Zdzislaw and Janina Trepinska, 1992, Climatic Conditions of 1815 and 1816 from 

Tree-Ring Analysis in the Tatra Mountains, in Harington, ed., Year Without a Summer, 

418-421. 

Bermuda Gazette and General Advertiser, 1785-1786. Archives of the Bermuda Historical 

Society, Hamilton, Bermuda. 

Bergthorrsen, P., 1969. An estimate of drift ice and temperature in 1000 years, Jokull 19, 94-101 

Borisenkov, E. P., 1992. Documentary evidence from the U.S.S.R., in Bradley and Jones, eds., 

Climate Since A.D. 1500, 171-183. 

Bradley, R.S. and P.D. Jones, eds., 1992. Climate Since A.D. 1500. Routledge, London. 

Crowe, R. B., 1992. Expansion of Toronto Temperature Time-Series from 1840 to 1778 Using 

Various United States and Other Data, in Harington, ed., Year Without a Summer, 145-

161. 

Dong, K., 1988. El Nino and tropical cyclone frequency in the Australian region and the 

northwest Pacific, Australian Meteorl. Mag. 36, 219-226. 

Fayle, David C., Catherine V. Bentley and Peter A. Scott, 1992. How did Treeline White Spruce 

at Churchill, Manitoba Respond to Conditions around 1816? in Harington, ed., Year 

Without a Summer,  281-290. 

Flohn, H., 1969. Ein geophysikalisches Eiszeit-Modell, Eiszeitalter und Gegenwart 20, 204-231, 

Ohringen/Wurtt. 

Fowle’s New Hampshire Gazette and General Advertiser, 1785. Athenaeum Library, 

Portsmouth, New Hampshire. 

Fritts, H.C., and X.M. Shao, 1992. Mapping climate using tree-rings from western North 

America, in Bradley and Jones, eds., Climate Since A.D. 1500, 269-295. 

Gray, William M. and John D. Schaeffer, 1991. El Nino and QBO influences on tropical cyclone 

activity, in Glantz,, Katz, and Nicholls, eds., Teleconnections Linking Worldwide Climate 

Anomalies, Cambridge University Press, 535 p. 

Harington, C.R., ed. 1992. The Year Without a Summer, World Climate in 1816. Canadian 

Museum of Nature, Ottawa, Canada.  

Huang Jiayou, 1992. Was there a Colder Summer in China in 1816?, in Harington, ed., Year 

Without a Summer,  448-451. 



 History of Meteorology 2 (2005) 121 
 

 

Jones, P.D., S.C.B. Raper, B.D. Santer, B.S.G. Cherry, C. Goodess, R.S. Bradley, H.F. Diaz, 

P.M. Kelly and T.M.L. Wigley, 1985. A grid point surface air temperature data set for the 

Northern Hemisphere, 1851-1984. DoE Technical Report TR022. U.S. Dept. of Energy, 

Carbon Dioxide Research Division, Washington, DC. 

Kington, John, 1992. Weather Patterns over Europe in 1816, in Harington, ed., Year Without a 

Summer, 358-371. 

Lamb, H. H., 1972. Climate: Present, Past and Future, Vol. 1, Fundamentals and Climate Now, 

Methuen & Co., London, 613 p. 

Lamb, H. H., 1977. Climate: Present, Past and Future, Vol. 2, Climatic History and the Future, 

Methuen & Co., Ltd., London, 835 p. 

Lamb, Hubert H., 1992. First Essay at Reconstructing the General Atmospheric Circulation in 

1816 and the Early Nineteenth Century, in Harington, ed., Year Without a Summer, 355-

357. 

Lough, J. M., 1992. Climate in 1816 and 1811-1820 as Reconstructed from Western North 

American Tree-Ring Chronologies, in Harington, ed., Year Without a Summer, 97-114. 

Luckman, B. H.,  and M. E. Colenutt, 1992. Early Nineteenth-Century Tree-Ring Series from 

Treeline Sites in the Middle Canadian Rockies, in Harington, ed., Year Without a 

Summer, 266-280. 

McNally, Louis K III, 1994. Application of Forensic Synoptic Analysis Techniques for 

Extraction and Reconstruction of Meteorological Information from a Non-Homogeneous 

Qualitative Data Set, MS Thesis, University of Maine, Orono, 90 p. 

McNally, Louis K, III, 2004. The Weather of 1785: An Interdisciplinary Approach to 

Meteorological Reconstruction Using Forensic Synoptic Analysis, Ph.D. Dissertation, 

University of Maine, Orono, Maine. 

Meeker, Loren D., and Paul A. Mayewski, 2002. A 1400-year high-resolution record of 

atmospheric circulation over the North Atlantic and Asia, The Holocene 12, 257-266. 

Mikami, T., and Yasufumi Tsukamura, 1992. The Climate of Japan in 1816 as Compared with 

and Extremely Cool Summer Climate in 1783, in Harington, ed., Year Without a Summer, 

462-476. 

Moberg, Anders, 1996. Temperature Variations in Sweden Since the 18th Century, 

Avhandling/Dissertation No. 5, Naturgeografiska Institutionen, Stockholms Universitet. 

O’Brien, J., T. S. Richards and A. C. Davis, 1996. The Effect of El Nino on U.S. Landfalling 

Hurricanes, Bull. Amer. Meteorol. Soc. 77 (4), 773-774. 

Ogilvie, A.E.J., 1981. Climate and society in Iceland from the medieval period to the late 18
th
 

century. Ph.D. dissertation, University of East Anglia, Norwich, U.K. 

Ogilvie, A.E.J., 1984a. The past climate and sea-ice record from Iceland, Part 1: data to A.D. 

1780. Climatic Change 6, 131-152. 

Pant, G.B., B. Parthasarathy, and N.A. Sontakke, 1992. Climate over India during the First 

Quarter of the Nineteenth Century, in Harington, ed., Year Without a Summer, 429-435. 

Pfister, C., 1978. Fluctuations in the duration of snow cover in Switzerland since the late17
th

  

century. Danish Meteorological Institute, Climatological Papers 4, 1-6. 

Pfister, C., 1992. Monthly Temperature and Precipitation in Central Europe 1525-1979: 

Quantifying Documentary Evidence on Weather and Its Effects, in Bradley and Jones, 

eds., Climate Since A.D. 1500, 118-142. 

Quinn William H., 1993a. The Large Scale ENSO Event, the El Nino and Other Important 

Regional Features, Bull. Inst. Fr. Etudes andines 22 (1), 13-34. 



 Reconstruction of Upper-air Circulation 122 
 

 

Quinn, W. H. and V. T. Neal, 1983. Recent Climatic Change and the 1982-1983 El Nino. 

Proceedings of the Eighth Climate Diagnostics Workshop, Canadian Climate Center, 

Toronto, Ontario, Canada. 

Quinn, W. H. and V. T. Neal, 1992. The Historical record of El Nino events, in Bradley and 

Jones, eds., Climate Since A.D. 1500, 623-648. 

Quinn, William H. and Victor T. Neal, 1983b. Recent Climatic Change and the 1982-1983 El 

Nino, Proceedings of the Eighth Climate Diagnostics Workshop, October 17-21, 1983, 

Canadian Climate Centre, Toronto, Ontario, Canada. 

Roy, S., 1972. A rare document on Delhi wheat prices 1763-1835. Indian Econ. and Soc. Hist. 

Rev. 9 (1), 91-99. 

Schove, D. J., 1954. Summer temperatures and tree rings in North Scandinavia, A.D. 1461-1950. 

Geografiska Ann. 37, 40-80. 

Serre-Bachet, F., J. Guiot and L. Tessier, 1992. Dendroclimatic evidence from southwestern 

Europe and northwestern Africa, in Bradley and Jones, eds., Climate Since A.D. 1500, 

349-365. 

Thompson, Lonnie G. and Ellen Mosley-Thompson, 1986. Evidence for Changes in Climate and 

Environment in 1816 as Recorded in the Ice Cores from the Quelccaya Ice Cap, Peru, the 

Dunde Ice Cap, China and Siple Station, Antarctica, in Harington, ed., Year Without a 

Summer, 479-492. 

Tsukamura, Yasufumi, 1992. The Reconstructed Position of the Polar Frontal Zone around Japan 

in the Summer of 1816, in Harington, ed., Year Without a Summer, 453-461. 

U. S. Department of Commerce, 2002. USDOC/NOAA/PMEP/TAO/El Nino, 

http://www.pmel.noaa.gov/tao/elnino/gif/summer_winter1-nns.gif. 

Villalba, Ricardo and Jose A. Boninsega, 1992. Changes in South American Tree-Ring 

Chronologies following Major Volcanic Eruptions between 1750 and 1970, in Harington, 

ed., Year Without a Summer, 493-509. 

Wang, P.K., and D. Zhang, 1992. Reconstruction of 18
th

 century summer precipitation of 

Nanjing, Suzhou, and Hangzhou, China, based on the Clear and Rain Records, in Bradley 

and Jones, eds., Climate Since A.D. 1500, 184-209. 

Wexler, H., 1956. Variations in insolation, atmospheric circulation and climate, Tellus 8, 480-94. 

Wood, Charles A., 1992. Climatic Effects of the 1783 Laki Eruption, in Harington, ed., Year 

Without a Summer, 58-77. 

 

 


