NWS Office Facebook and Twitter Update

Every six months or so I take a look at the social media presence of the 118 National Weather Service forecast office (not including San Juan, Guam, and American Samoa). The previous analysis of both the Facebook and Twitter followers is located here ( http://us-climate.blogspot.com/2015/03/nws-social.html ). Please note that the three Alaska NWS Office share a single Facebook page, yet have separate Twitter pages. In lieu of a complete replication of the March 28th post, I want to show a snapshot of where things stand as of now.

A note of caution. While rankings are labeled on the maps, and while it is human nature to make comparisons, keep in mind that the needs and demographics of each WFO are different.

Facebook:

The three Facebook maps below show the raw number of Likes (Figure 1), the rate of Likes per 100,000 population (Figure 2), and the change in Likes over the last six months (Figure 3).

Figure 1. Facebook Likes for 118 National Weather Service forecast offices as of September 28, 2015. Rank label (highest to lowest) added to each WFO.

Figure 2. Facebook Likes per 100,000 population for 118 National Weather Service forecast offices as of September 28, 2015. Rank label (highest to lowest) added to each WFO.

Figure 3. Rate of change of Facebook Likes for 118 National Weather Service between March 28, 2015, and September 28, 2015. Percent change labelled for each WFO.


Twitter

Twitter functions quite a bit differently than Facebook. While Facebook is a more casual source of information, Twitter is much more like a news aggregation site. The information is short, sweet, and timely. My experience has been that Twitter is much more of a specialized type of social media where the participants are in the media, celebrities, actors, athletes, and subject matter specialists.

So, here are my two maps for Twitter. The first (Figure 4, shows the number of Twitter followers for each NWS page. The second show the average number of Tweets sent out by those offices since their Twitter account was activated (Figure 5).

Figure 4. Twitter followers for 120 National Weather Service forecast offices as of September 28, 2015. Rank label (highest to lowest) added to each WFO. (Alaska offices separated out).

Figure 5. Twitter followers per 100,000 population for 120 National Weather Service forecast offices as of September 28, 2015. Rank label (highest to lowest) added to each WFO. (Alaska offices separated out).

Figure 6. Number of Tweets per month sent by the 118 National Weather Service forecast offices as of September 28, 2015. Rank label (highest to lowest) added to each WFO. Value computed by dividing number of tweets by number of months since page's inception. (Alaska offices combined).


Table 1. Raw count of Facebook Likes, Twitter Followers, Tweets, and rate of Tweets by National Weather Service Forecast Office (WFO). Separate entries made for the Alaska WFOs so that Twitter statistics are identifiable. They share identical Facebook values.

WFO	FB Likes	Twitter Followers	Total Tweets	Tweets / Month
Aberdeen	9,463	3,425	2,285	58.0
Albany	28,481	5,789	5,438	141.6
Albuquerque	9,165	5,279	11,200	258.3
Amarillo	39,288	7,792	3,984	101.2
Anchorage	43,699*	4,951	2,501	65.8
Austin/San Antonio	16,336	10,300	8,812	218.1
Baltimore/Washington	38,992	10,700	4,705	133.0
Billings	14,186	3,464	4,207	104.1
Binghamton	85,197	7,283	9,264	254.8
Birmingham	78,009	17,100	12,000	297.0
Bismarck	20,592	3,312	4,601	113.9
Blacksburg	22,607	5,785	8,358	223.6
Boise	15,107	3,918	2,752	77.8
Boston	72,275	32,700	17,200	447.9
Brownsville	23,898	4,147	6,063	150.1
Buffalo	10,694	7,841	1,900	56.9
Burlington	20,503	5,827	6,228	162.2
Caribou	28,668	2,421	8,889	274.8
Charleston (SC)	8,401	10,200	9,228	176.0
Charleston (WV)	29,944	5,118	1,871	47.5
Cheyenne	11,719	2,924	4,455	113.1
Chicago	112,556	20,800	11,400	296.9
Cleveland	15,679	5,954	2,655	67.4
Columbia	12,292	3,848	5,199	109.7
Corpus Christi	10,188	4,535	3,763	93.1
Dallas/Fort Worth	102,441	39,000	12,800	316.8
Davenport/Quad Cities	13,380	4,816	3,289	83.5
Denver/Boulder	16,615	11,400	6,047	153.5
Des Moines	29,778	9,979	7,981	202.6
Detroit/Pontiac	12,259	6,026	3,273	83.1
Dodge City	9,820	4,291	6,904	170.9
Duluth	8,838	2,204	2,594	75.5
El Paso	5,971	4,317	9,264	213.6
Elko	6,943	2,285	7,386	182.8
Eureka	11,549	3,051	5,909	146.2
Fairbanks	43,699*	2,600	1,967	51.8
Flagstaff	11,660	6,286	4,175	103.3
Gaylord	9,822	2,808	9,675	245.6
Glasgow	10,538	2,507	2,097	51.9
Goodland	7,509	3,446	6,866	174.3
Grand Forks	10,075	3,124	3,334	84.6
Grand Junction	5,631	2,249	5,073	143.4
Grand Rapids	19,710	5,550	3,772	95.8
Gray/Portland	13,048	4,679	6,611	172.2
Great Falls	7,913	3,458	9,199	227.7
Green Bay	8,392	3,916	2,409	59.6
Greenville-Spartanburg	15,602	4,995	5,253	133.4
Hastings	19,210	4,365	7,453	189.2
Honolulu	17,248	8,050	1,722	32.8
Houston/Galveston	18,024	11,400	6,037	153.3
Huntsville	27,619	10,900	4,687	116.0
Indianapolis	20,747	14,900	8,245	209.3
Jackson (KY)	12,760	3,855	7,853	199.4
Jackson (MS)	75,295	13,200	14,400	356.4
Jacksonville	8,456	4,052	4,787	118.5
Juneau	43,699*	2,109	7,627	200.7
Kansas City/Pleasant Hill	82,302	24,100	9,481	180.8
Key West	28,081	7,863	4,226	104.6
La Crosse	12,270	4,006	6,022	152.9
Lake Charles	11,115	3,338	5,264	130.3
Las Vegas	10,593	8,844	9,026	223.4
Lincoln	30,948	6,647	4,746	120.5
Los Angeles/Oxnard	18,945	15,800	8,195	208.1
Louisville	21,372	8,868	8,821	224.0
Lubbock	8,155	5,680	5,272	130.5
Marquette	16,743	4,003	3,722	92.1
Medford	12,430	2,739	1,427	35.3
Melbourne	13,314	5,086	4,148	102.7
Memphis	22,413	9,820	9,628	238.3
Miami	15,069	13,300	8,395	207.8
Midland/Odessa	12,836	3,416	8,470	209.6
Milwaukee/Sullivan	25,652	5,169	4,797	121.8
Minneapolis/Twin Cities	23,455	15,400	10,300	261.5
Missoula	8,757	3,586	9,657	239.0
Mobile/Pensacola	25,052	14,400	12,200	301.9
Morristown/Knoxville	18,615	3,874	3,198	79.1
Mount Holly/Philadelphia	67,482	17,300	2,637	67.0
Nashville	55,707	13,600	14,200	360.5
New Orleans/Baton Rouge	32,068	7,587	6,010	148.7
New York	46,268	24,600	8,252	209.5
Newport/Morehead City	12,505	2,941	5,538	152.3
Norman/Oklahoma City	173,839	53,000	14,200	390.5
North Little Rock	64,215	19,100	2,861	70.8
North Platte	7,658	3,677	7,308	185.5
Northern Indiana	40,180	10,100	11,200	284.4
Omaha/Valley	13,283	8,778	4,701	119.4
Paducah	44,807	10,100	8,107	205.8
Peachtree City/Atlanta	26,782	10,300	4,377	108.3
Pendleton	9,201	2,461	2,052	52.1
Phoenix	10,567	10,500	6,978	172.7
Pittsburgh	48,433	13,500	5,404	144.6
Pocatello	4,014	2,465	9,095	225.1
Portland	19,633	6,893	5,465	142.3
Pueblo	5,939	3,805	5,377	133.1
Raleigh	12,545	7,873	3,471	92.9
Rapid City	11,519	3,452	1,306	33.2
Reno	14,657	4,833	5,024	124.3
Riverton	10,841	2,004	7,613	193.3
Sacramento	18,599	10,200	11,900	302.1
Salt Lake City	8,052	10,200	10,400	257.4
San Angelo	16,336	4,636	5,234	129.5
San Diego	13,427	6,355	12,500	288.3
San Francisco Bay Area/Monterey	18,989	17,100	16,800	426.5
San Joaquin Valley/Hanford	14,739	4,204	11,500	284.6
Seattle	12,839	7,670	2,812	73.2
Shreveport	29,533	4,069	11,500	292.0
Sioux Falls	21,464	10,300	7,619	188.6
Spokane	12,318	4,618	5,673	70.6
Springfield	27,260	8,964	4,706	119.5
St. Louis	36,715	8,564	3,002	74.3
State College	20,811	5,744	8,544	216.9
Tallahassee	8,964	6,782	6,249	154.7
Tampa	14,837	7,810	12,800	316.8
Topeka	13,285	7,880	4,356	107.8
Tucson	6,378	5,493	6,606	163.5
Tulsa	41,225	11,100	7,171	139.5
Wakefield	8,864	2,531	635	21.6
Wichita	33,731	14,000	7,561	192.0
Wilmington (NC)	8,780	5,200	2,116	59.8
Wilmington (OH)	39,268	11,500	5,548	140.9

Annual Temperature Extremes

A frequently used metric for describing the continental or marine characteristic of the climate for a location is the annual temperature range. Stations near an ocean have their temperatures significantly modefied by the air over the adjacent water. Ocean water heats up and cools down much slower than land. When a landmass is close to an ocean, modified air from the ocean can quickly replace continental air as it heats up or cools down; thereby minimizing the extremes in temperature. Conversely, stations in the interior of a continent are sufficiently far removed from the moderating influence of the water and their temperatures vary greatly.

In this blog post, I pulled daily temperature readings for the 1981-2014 time period and computed the coldest temperature each winter (not calendar year) and the hottest temperature each summer. The data were obtained from the National Center for Environmental Information's (NCEI) Global Historical Climatology Network (GHCN) database. Only WBAN ("first order") stations were used due to computational and computer memory limitations (note: in Alaska, Cooperative stations were also utilized). 

To be included in the analysis, a complete year of data must be present (no more than 10 missing observations). In addition, a station must have a minimum of 10 complete years since 1981. A total of 1,049 stations met the inclusion criteria. The following sets of maps show the typical warmest temperature of the year, the typical coldest temperature of the year, and typical annual temperature range. Importantly, the median value for each station was used – not the average. In some cases, extreme outliers or even bad data make it into the GHCN database. Using median values prevents those suspect values from overwhelming a station's true value.

U.S. Maps

Figures 1, 2, and 3 show the annual warmest temperature, annual coldest temperature, and annual temperature range. These are not all time extremes. They represent the median value from the list of annual extremes.

While no one should be surprised that it is hottest in the desert southwest and coldest in Alaska, what might surprise some people is that interior Alaska has the greatest annual temperature extremes in the U.S. The town of Eagle, Alaska, near the Canadian border, leads the way with a 145°F temperature swing from winter to summer in an average year. At the other end of the spectrum, several stations in Hawaii only have a 27°F change in temperatures from the low in winter to the max in summer!

Figure 1. Median value for the hottest temperature in a given year. 

Figure 2. Median value for the coldest temperature in a given year. 

Figure 3. Median value for the range in temperature between the coldest winter temperature and the hottest temperature the following summer.

Alaska-Centered Maps

For my Alaska readers, here are the same figures but reconfigured so that Alaska is the focus of the maps.

Figure 4. Median value for the hottest temperature in a given year (Alaska-centered)

Figure 5. Median value for the coldest temperature in a given year (AlasKa-centered)

Figure 6. Median value for the range in temperature between the coldest winter temperature and the hottest temperature the following summer (Alaska-centered).