New York City Weather Archive

Today in New York Weather History: November 16

 

1933

The day after a winter-like high/low of 32/21, today was even colder, with a high/low of 30/17 (twenty degrees colder than average).  Today's low, a record for the date that still stands, is the earliest reading in the teens.

1978

The 0.05" of rain that fell overnight was the first precipitation this month (and since Oct. 27).   

1983

Overnight, New York's all-time annual precipitation record was set, exceeding the 67 inches that fell in 1972.  Still, reservoirs were only filled to 42% of capacity since NYC's heavy rains had regularly missed these upstate watershed areas.  And at the end of the year it was revealed that the rain gauge in Central Park had been malfunctioning since the spring, thus calling into question the year's reported precipitation.

1985

2.39" of rain fell, a record for the date.  Most of it fell after 12 noon and it came down heaviest between 9-11PM when 1.11" was measured.

 

1996

Today was the sixth day in a row with temperatures 10 degrees or more chillier than average.  The average high/low of 41/32 during these days was 12.5 degrees below normal.  Five days had morning lows of freezing or below.  

1997

Temperatures today and the past two days were very similar and never got out of the 30s; highs/lows were: 38/32, 37/32 and 39/31 (today).  These temperatures were thirteen degrees below average.   

2011

Today was the first dreary, rainy day since the area's freak snowstorm two-and-a-half weeks ago.  Rain fell for much of the day and amounted to 0.75".

Save

Save

October 2018: Started Out Like September, Ended Like It Was November

Although October wasn't far from average in terms of its overall temperature, few would think of it as average as the first eleven days of the month had temperatures comparable to the start of September followed by nearly three weeks of temperature more like early November.  Through the 11th temperatures were ten degrees above average (high/low of 75/65), then the rest of the month was about twenty degrees colder (57/45), and four degrees below average.  Another contrast was between the departure from normal of the month's average high, which was a titch below average (-0.3 degrees), and the average low, which was nearly two degrees milder than average (+1.9 degrees).  Despite it being 0.8 degrees above average, this was the chilliest October in seven years.

 

Additional points of interest:

• Similar to August and September, the difference between October's average high and low (i.e., diurnal variation) was one of the smallest on record for October - 11.6 degrees (the average is 15.4 degrees).  In the past 100 years only October 2012 had a smaller variation.

• Just a year after the latest date for a low in the 70s occurred (on 10/9) a new mark was set this year when a low of 71° (five degrees warmer than the average high for the date) was reported on the 10th.

• The month was well above average in terms of days with lows of 60° or milder (eleven).  Only October 1879 (thirteen), 2012 (warmest October on record) and 2007 had more lows of 60+.  And this October's eleven days were consecutive (Oct. 1-11), which set a new October record and included five days in a row with mean temperatures that were ten or more degrees above average.  However, despite these balmy conditions just one day, 10/10, had a high in the 80s (80°).

• During the chilly portion of the month there were a dozen days with highs in the 50s, the most since 2009 (the typical number is eight).  And the number of lows in the 30s was three, the most since 2011 (the chilliest reading was 38°).

• The high of 64° on Halloween (five degrees above average) was the month's mildest reading since 10/15 and the first Halloween to have temperatures in the 60s in five years.

• Finally, much of October's 3.59" of rain (the smallest amount in five years) came from the remnants of catastrophic hurricane Michael on 10/11-12, which brought 1.27", and a nor'easter on 10/27 that produced 1.29".  On 10/20 the year's total precipitation passed 50 inches, and by the end of the month 51.42" had been measured, which is close to two inches more than what falls in a typical year.  This ranks as the the ninth greatest amount of precipitation (since 1869) through the end of October.

Here are recaps of previous Octobers:

2017

2016

Can 'Trace' Amounts of Precipitation Add Up to Measurable Amounts?

When a day's precipitation is reported as 'T', which signifies a "trace" amount, this indicates that less than 0.01" fell (ranging from just a few raindrops or snowflakes up to 0.0044").  From 2000 thru 2017 there have been, on average, two days each month with a trace of precipitation.  In the years I've been writing this blog I've become more and more curious whether trace amounts falling on multiple days can add up and become a measurable amount - something that weather records don't recognize.  (Additionally, on days when a trace falls there are often multiple hours with a trace.  Furthermore, there are days with measurable precipitation that report an hour or more of trace amounts.)

 

ILLUSTRATING MY POINT

Let's say we have a month in which five days had just a trace of precipitation:

Although the amounts are considered traces, precise measurement might report very small amounts of liquid:

Therefore, the five-day total of these trace amounts would indicate that 0.0102" fell during these five day (or 0.01").  The example above would also apply to a day with multiple hours of traces of precipitation. 

 

SOME ADDITIONAL FACTS ABOUT TRACE AMOUNTS

December 1989 had the most days on record, fifteen.  Four of the five years with the most days with a trace of precipitation occurred from 1938 to1943.  The four years with the least were all in the 2000s.  Curiously, the average number per year since 1999 has been half the number in the prior years (46 days vs. 23).  However, it appears 2018 is on its way to having the most in 20 years.

 

IN CONCLUSION

With the today's measurement technology able to measure more granular amounts of liquid, weight, speed, etc., surely a device must exist that can show that trace amounts added together may result in a measurable amount.  Granted, these additional measured amounts won't be large (and which date would be given credit for a measurable amount over the course of multiple days?) but they would make weather records more precise, which is something "Big Data" types could appreciate.  (Perhaps precipitation can be reported to three decimal places?  In a previous post I made a case for reporting temperatures to one or two decimal places). 

 

September 2018 - A Month With Unusually Mild Nights

September 2018 was 2.7 degrees warmer than average.  Like August, its average low was considerably more above average (+4.4 degrees) than its average high (+1.0).  This resulted in a vast difference in their historical rankings, with the average high ranking 60th warmest while the average low was ensconced among the ten warmest, at #4 (overall, the month was the 21st warmest).  The month's other significant story was its rainfall, which amounted to 6.19" - the most in September in seven years.  Furthermore, this was the sixth month this year with more than five inches of precipitation, joining ten other years with this many or more.  And if that wasn't impressive enough, 2018 became just the second year to have these six months concentrated in the first nine months of the year (1998 was the other year). 

 

Additional points of interest:  

• The remnants of two hurricanes brought 1.38" of rain on 9/9-10 (Gordon) and 1.19" on 9/18 (Florence).  However, the month's biggest rainstorm was on 9/25 when 1.95" fell. Nearly half of that day's rain poured down in about an hour early in the afternoon.

• The first full week of the month (week of 9/3) had four days in a row with highs that ranged from 89° to 93°.  The average high/low of 91/76 during these days was twelve degrees above average.  The first day of this hot spell fell on Labor Day, which had a high in the 90s for just the fourth time since 1980. 

• This was the fourth September in a row with an average temperature warmer than 70.0 degrees, the longest such streak on record.  What's more, the month set a September record for the most days with lows of 67° or warmer (seventeen; the average number is six).  Despite this surfeit of warm nights, the number of days with highs of 80° or warmer was a bit below average (nine).

• The month's average diurnal variation (the gap between the average high and low) was the smallest for September in more than 100 years (since 1908).  In 2018 it was 11 degrees, well below the average of 16 degrees (based on the past 100 years).

• This was the record-tying seventh year in a row in which September experienced highs in the 90s, equaling in length the streak of 1940-1946.

• Like September, August's average low also ranked fourth (its average high ranked 32nd).

• Typically September is 2.5 degrees cooler than June, but this year it was one degree warmer.  This is the fifth time it's happened since 1980 (most recently in 2015).

 

 

Here are recaps of the four previous Septembers:

2017

2016

2015

2014

 

August 2018 Weather Recap: Unusually Wet, With Unusually Warm Nights

July's rainy and warm conditions carried over into August, which was even warmer and wetter, ending up as the ninth warmest August on record and the twelfth wettest.  It was 2.9 degrees above average and warmer than July (something that's happened in 22% of the years since records began in 1869).  Despite August being warmer than July (by 0.5 degrees), July's average high was slightly higher (84.9 vs 84.3); it was August's average low that made it warmer overall (72.0 vs. 70.4).  In terms of rainfall, August's 8.59" was about an inch more than what was measured in July (7.45"), with much of it falling during heavy downpours during thunderstorms.  Here are additional observations worth noting:

 

• There were seven days in the 90s, which occurred during two heat waves - one was four days in length (8/5-8) and one lasted three days (8/28-30).  The four-day heat wave was one of the least impressive among those of this length. The hottest reading of the month was 94° on 8/28.  The coolest reading of the month, which was reported twice, was 65°, which tied August 2005 and 1906 for the mildest minimum temperature in the month of August.  The typical coolest low temperature in August is 59° (in the years since 1980).

• This was the thirty-second warmest August in terms of average high, but it had the fourth warmest average low.  (The low was 4.2 degrees above average, the high was 1.7 degrees above.)  This produced the ninth warmest August overall (tied with Aug. 1955, which was even rainier than this August, with nearly eleven inches measured).  Three of the past four Augusts have now placed among the ten warmest.

 

 

• Twenty-three days had lows in the 70s, tying August 1980 and 2005 for the most in the month of August (the typical number of such days is 13).  The month's first 18 days all had lows in the 70s, only the second time it's occurred, joining August 1988.  It appeared the City was going to set an August record for most lows of 70 or warmer, but it was not to be as the mercury fell to 69° the evening of 8/31.  On 8/29 the low of 81° established a new record for the latest date for a low in the 80s, smashing the previous record by two weeks (which occurred in 1985 and 1988).  This summer was just the eighth in which both July and August had average lows above 70° (two of the other summers were in 2015 and 2016).

• The month's diurnal variation (i.e., the difference between the high and low) was just 12.3 degrees, which was the smallest in August since 2000 (also 12.3 degrees).  The only variations smaller than this occurred more than 100 years ago, when eight Augusts in the years before 1911 had variances between 10.4 and 11.7 degrees (at the other end of the spectrum, the greatest diurnal variation in August is 20.4 degrees, which occurred in both 1964 and 1944.)

• This year has had the twelfth wettest August, the fourth wettest July-August, and the twelfth wettest summer.  Almost all of July and August's rain fell in a soggy six-week period (July 12 - Aug. 22).  Ironically, while August is the year's wettest month (so far), it also had the longest streak with no measurable rainfall - eight days, from the 23rd to the 30th.

 

 

Here are previous August recaps:

2017

2016

2015

 

 

 

Lackluster Heat Waves

This post was inspired by this week's rather tepid four-day heat wave (Aug. 5-8), which had an average high of 90.8 degrees (with highs of 91, 92, 90 and 90).  Of the 66 four-day heat waves that have occurred since 1870 only one, in July 1896, had a lower average high, 90.3 degrees.  (2018's low-grade heat wave matched one in August 2009.)  On average, four-day heat waves have had an average high close to 94°.  (The hottest four-day heat wave on record took place in the summer of 2010, when the high temperature from July 4-7 was 99.5 degrees.)

 

 

 

 

However, the story changes when low temperatures are included in the analysis.  For example, this August's heat wave had an average low of 75.0 degrees, which was 1.3 degrees warmer than the average four-day streak and warmer than two-thirds of the four-day heat waves examined.  When combined with the average high, the mean temperature ranked as 27th coolest - quite a difference in ranking compared to its average high alone.  (This follows the weather storyline of this century whereby nighttime temperatures in New York are warming more than daytime temperatures.)

 

 

Finally, while the typical four-day heat wave had a 20-degree difference between its high and low, this August's was 15.8 degrees apart, which was the third smallest diurnal variation of the heat waves studied (a heat wave in July 1995 had the smallest, 14.7 degrees, while the second smallest was in July 1870).  

Stuck in the 70s

Despite what the headline suggests, this post isn't about people obsessed with 1970s pop culture.  Rather, it's a discussion about days that have had both their high and low temperatures in the 70s.  These type of days have occurred infrequently - about once or twice each year.  33 years (since 1869) have never registered such a day, 53 have had one instance; the most days in one year was nine, which happened in 1905, 1890 and 1869.  In 2017 and 2018 there were five occurrences, joining just ten other years that have had five or more (see chart below).  Temperatures stuck in the 70s usually occur when tropical air is in place, the day is hazy/overcast and rain is in the air. 

 

FURTHER DETAILS

• The warmest low in the 70s to be accompanied by a high temperature below 80° is 74°, which has happened twice - first in August 1901 and then seventy-two years later in August 1973.  Both dates had a high of 79°.  (For reference, the average high when lows are between 70°-74° is around 86°).
• The most consecutive days with temperatures in the 70s all day was five, in 1905 (Aug. 30-Sept. 3).
• The most consecutive years without any "all-70" days is four, from 1962 to 1965 (and six out of seven years from 1962 to 1968).  By contrast, there were 20 of these days in the three-year period 1905, 1906 and 1907; they had nine, five and six of these days, respectively.
• The first decade of the 20th century had more "stuck in the 70s" days than the next three decades combined (38 vs. 37).  The 1890s had the second greatest number - 34. The current decade, so far, has had 23 occurrences (thru Sept. 2018).
• The latest occurrences for a high/low in the 70s were in 2017, on Oct. 8 and 9 (the 9th is also the latest date for a low in the 70s).  The earliest occurrences were on June 4 and 5 in 1960.
• The dates most likely to have experienced a day entirely in the 70s are 7/23, 7/29, 8/9 and 8/12, all of which have had seven such occurrences.
• One out of four of these days either had no measurable rain (half were before 1935) while another one out of four received more than a half-inch of rain.  The greatest rainfall to occur was in 1990 when 4.64" fell on Aug. 10 (the high/low was 76/70).  Two other days with substantial rainfall were associated with Hurricanes Connie and Irene, in Aug. 1955 and Aug. 2011, respectively.

 

 

July 2018 Weather Recap: Hot Temperatures, Followed by Torrents of Rain

If July 2018 were a cocktail its ingredients would be one part heat, three parts rain.  After a hot start, with temperatures nine degrees above average during the month's first five days (average high/low was 92/78), the rest of July was seasonably warm, which moderated the month's average temperature to 1.1 degree above average.  As for the rain, after a dry start, with just 0.24" measured through July 11, the skies opened up and the rest of the month was very rainy, with 7.21" measured.  With 7.45" in total this was the rainiest month in more than four years, since April 2014, and the rainiest July in fourteen (and the fourteenth wettest July on record).

 

In late July there were seven days in a row (July 21-27) in which measurable rain fell.  This was the fourth streak of six days or more this year, joining 1989 as the only two years since 1900 with four streaks of six days or longer.  (Besides the persistent rain, dew points throughout this week-long period were in the uncomfortable 72°-74° range.)

 

 

The seven-day streak with rain was the first since 2012, but this year's had considerably more rain (2.89" vs. 0.86").  The last streak of seven days or more with more rain was in May 2009, when 3.81" fell.

 

Downpours on 7/12 and 7/30 were confined to upper Manhattan; Central Park had 0.71" and 0.37", respectively, but my neighborhood, Greenwich Village, four miles south of the park, was dry.  And during the afternoon of 7/17 Manhattan was the bulls-eye for torrential rain from a severe thunderstorm, with 2.24" pouring down in little more than an hour.  Conversely, on 7/27 thunderstorms during late afternoon thru evening produced more than an inch of rain in most parts of the metro area (and 3-5" in parts of east-central NJ) but Central Park had only 0.25". 

 

July's rainfall in Central Park was considerably more than the three other reporting sites in the metro area:

 

 

 

Other July observations: 

• Four of the first five days in July had highs in the 90s; the only one that didn't was the 4th of July, but its high of 86° was the warmest reading on this holiday in five years.  However, due to its very warm low of 77°, the mean temperature was the warmest since 2010.

 

 

• Thirteen days in a row (July 8-20) had highs of 80° or warmer, the longest such streak since one of 19 days in August 2016.  (The record is 62 days in a row, set three years ago.)

• What was most impressive about the hot beginning to July were the low temperatures, all of which were 76° or warmer.  This happens almost as infrequently as five-day streaks with highs of 95°+.

• This was the ninth July in a row with no lows in the 50s, the longest such streak in the years since 1900.  Before 1980 more than half of Julys had at least one low in the 50s, but since 1980 that portion has fallen to 25%.

• By July 25 the month had picked up more rain than the combined amount in May and June (6.80" vs 6.64").

 

Here are links to previous July recaps:

2017

2016

2015

2014

 

Comparing Central Park's Weather Observations With Those of New York's Three Airports

New York City's official reporting site for weather conditions is located in Central Park, but LaGuardia and Kennedy Airports also collect data, as well as Newark Liberty Airport in New Jersey (the airports are 7, 16 and 27 miles away from Central Park, respectively).  Central Park is one of the few reporting sites in the US not located at an airport.  Although the park is surrounded by the "heat island" of Manhattan, its temperatures are influenced by the grass and trees, which hold humidity more than the concrete surroundings of airports.  As a result, afternoon temperatures in the summertime don't rise as much as they do at the three airports, and nighttime temperatures don't fall as much during the winter or summer. 

 

I've looked at five statistics, which cover temperatures and precipitation for the 2000-2017 period.  Of the four weather stations, Central Park averages the most days with highs of 32° or colder and receives the most precipitation while Newark is tops in the number of 90-degree days, lows of 32° or colder and snowfall. 

 

HIGHS OF 90°+

Central Park: Averages 15.6 days (the biggest difference between CPK and Newark was in 2006, when CPK had only eight days, 50% below average, while Newark had 27, an average number) 

Newark: 27.1 (it reported the most in all but one year)

LaGuardia: 21.2 (reported the most in 2007)

Kennedy: 11.5

 

HIGHS OF 32° OR COLDER

Central Park: Averages 17.9 days (it had the most of the four stations in ten years, and three first-place ties)

Newark: 16.2 (reported the most in one year, and one first-place tie)

LaGuardia: 16.4 (reported the most in two years, and one first-place tie)

Kennedy: 16.2 (reported the most in two years, and one first-place tie)

 

LOWS OF 32° OR COLDER

Central Park: An average of 68.8 days

Newark: 81.1 (had the most in all but two years)

LaGuardia: 63.6

Kennedy: 75.3 (had the most in two years)

 

ANNUAL PRECIPITATION

Central Park: 50.58" (half of the years had 50"+; reported the most in every year but two)

Newark: 46.41" (four years had 50"+; reported the most in one year, 2017)

LaGuardia: 45.68" (six years had 50"+; never reported the most)

Kennedy: 43.76" (three years had 50"+; reported the most in one year)

 

ANNUAL SNOWFALL

Central Park: 33.6" (50"+ fell in three of the years; had the most of all the sites in two years)

Newark: 35.2" (four years had 50"+; had the most in half of the years of the period, and tied with CPK in 2010)

LaGuardia: 32.8" (two years had 50"+; had the most in six years, including the last three)

Kennedy: 29.4" (one year had 50"+; no year had the most of the four stations)

 

 

 

Save

June 2018 Weather Recap for New York

Following an April that was significantly colder than average, by 3.6 degrees, and a May that was 4.5 degrees milder than average, June was "just right" as it was close to average, 0.2 degrees above average.  This was the closest to average for any month since June 2015 (0.2 below average).  However, despite the month being average overall, the first half was cooler (and drier) than average while the second half was warmer (five days had highs of 88° or hotter) and wetter (relative to the first half, since the amount that fell was average).  Here are a few other observations of interest: 

 

• Father's Day had a high of 91°, which was 37 degrees hotter than Mother's Day (typical difference between the two holidays is thirteen degrees).  It joined Father's Day 2010 as having the greatest difference in high temperature from Mother's Day (going back to 1960).  Also, this was the first Father's Day with a high in the 90s in twelve years (and the seventh since 1960).

• The hottest day of the first half of the year occurred on the last day of the period as the high reached 93° on 6/30. (The second half of the year began with a high of 96° on 7/1).

• Rainfall was below average, 3.11" vs. a norm of 4.41".  Three-fourths of this amount fell in the last two weeks of the month.  And much of it, 1.36", fell on 6/28, highlighted by the downpour during the evening rush hour, which saw 0.93" pour down in about an hour.

• This storm brought to mind the severe thunderstorm of 5/15 that snarled rush hour traffic.  However, despite producing less rain (0.58") it was characterized by destructive winds that killed five people who were struck by falling trees.

• June was the closest to average a month has been since June 2015, which was 0.2 degrees below average.  Of the twelve months that have been 0.2 degrees or closer to average since 2000, five of them have been in June.

 

 

 

 

 

No Thunderstorms in Central Park for the Past 15 Years? So Says the National Weather Service

I've lived in Manhattan since 1981 so I can attest to the fact that we get our fair share of thunderstorms; through 1995 the average number each year was nineteen.  However, while doing research on thunderstorms, I was very surprised to discover that since 1996 only three were reported in Central Park (and none since 2003).  This is according to NOAA's annual Local Climatological report for CPK (which is the official reporting site for New York City).  Yet, during these same years the reporting stations at LaGuardia, Kennedy and Newark airports each averaged 25 per year.  I thought, perhaps, this was just an egregious oversight, so I contacted NOAA and I was told that it wasn't an error but, rather, the result of "automation and a lack of human augmentation since the mid 1990s."

 

I was taken aback by this explanation.  So, the most populous city in the nation no longer reports the incidence of thunderstorms because a human can't be found to do it?   What added to my consternation was that the "Reference Notices" section found in the monthly and annual reports makes no mention that thunderstorms are no longer reported, nor is there an asterisk in the report's month-by-month line for thunderstorms explaining why the number is '0' every month.  Confounding me further is the fact that an annual average is still reported - a number that grows smaller each year since all of the years with no thunderstorms reported are factored in.  So the most current average listed, 11.1, is completely bogus. 

 

 

Of course, in the grand scheme of things a statistic indicating how often thunderstorms occur is not a crucial one (and it should be noted that any rainfall from a thunderstorm is reported).  However, what troubles me is that there is no explanation that thunderstorms are no longer reported.  Someone not familiar with the NYC climate may think the City actually gets no thunderstorms (and not be aware measurement sites at the three airports collect such information).  And as for the "lack of human augmentation" - isn't that what interns are for?  

 

 

Although thunderstorms are no longer reported in Central Park, the way it is displayed on NOAA's annual summary leaves the impression that they are, but there just weren't any storms in 2017.

 

Curiously, the end of tracking thunderstorms in Central Park coincides with the reporting of days of Heavy Fog for the first time.  In other words, before 1996 the average number of days each year with heavy fog reported (1/4 mile or less) was zero, but since 1996 there have been, on average, 14 such days.  It seems an odd coincidence that the jump in the number of foggy days somewhat coincides with the number of thunderstorms before 1996 while the number of thunderstorms in the present day is what the number of foggy days used to be before 1996.