We now give the subject of site and instrument exposure considerable attention as these factors can create great variability in the accuracy, continuity, and representativeness of measurements. They are also factors over which we can exert considerable control to minimize unwanted impacts.

Station exposure includes general siting characteristics (urban, suburban, rural, etc) and topographic setting, and instrument exposure. Instrument exposure relates to the specific placement of the sensors (height above ground, distance to surrounding objects, etc.)

The specific topographic and environmental setting of a station is an extremely important variable in determining the measured values of the climate data collected. In some situations, the topographic exposure of a station can result in much more variability in temperature and precipitation values than differences of say, 5 horizontal miles and 100 vertically feet (NWS policy for renumbering relocated stations). Changes of even a few feet (5 or 10) in elevation, or less than 50 feet horizontally, can affect readings. Relocating a station nearer to hills or hollows, in and out of nocturnal drainage channels, or near the influence of other subtle features, can change the climate record. This may sound unlikely, but it has been documented many times, even in apparently homogeneous "flat" land. Snow can have a major effect on temperature, and the way that snow accumulates and melts in the vicinity of the station can change with time. Also note that temperature may be affected significantly by a site change, and precipitation barely so, or vice versa. The impacting factors are typically element specific.

The type of topographic exposure of a location greatly influences spot climate. Figure 7 illustrates a classic example of the different types of topographic settings in the high Alleghenies of north-central West Virginia. They are:

a) elevation (above sea level, and locally, small variations of a few feet),
b) slope (north, south, east, west facing, or level), and,
c) topographic setting (ridge top, valley, etc.).

We are all familiar with the important role that elevation plays in determining the climate and weather of a locale. Therefore we concentrate our discussion on the other two, less known factors of slope and topographic setting.

The slope orientation of a site changes the site's exposure to solar radiation and therefore the basic energy balance that drives temperature. The north slope of a hill or building does not receive as much radiation as the sides facing south (in the northern hemisphere). Also, while the radiation exposure is highest to the south, the highest temperatures usually occur on southwest facing slopes. The point to make here is that extremes of any exposure are to be avoided in site selection whenever possible. If such exposures are unavoidable, it is again critical that the metadata (especially digital photography) reflect the specifics of the site's orientation.

In addition to slope orientation, there are four basic topographic settings that can introduce significant variability into the spot climate of a site. These settings are:

- Crest (summit)
- Slope (sides of a hill or ridge)
- Valley bottom (frost hollows, flood plains, etc. surrounded by higher ground)
- Plain (level, flat surface without higher or lower ground close by)

Each of the settings, at similar elevations, can exhibit large differences in climate, including maximum and minimum temperatures, temperature range, shape of the diurnal temperature curve, etc. In general, elevation for elevation, crest, or hill tops, exhibit reduced temperature range (lower maximum and higher minimum) as compared to valley bottom locations. Slopes exhibit features in-between crests and valleys.

Topographic setting is extremely important factor in deciding where to install a station and with relocations, whether to determine if climate compatibility exists between the old (or current site) and the new location (section 5: data continuity contains a detailed discussion of station relocations).

Topographic Setting is Important in Determining Climate Compatibility for Station Relocations (Allegheny Mountains of West Virginia)
Figure 7: Topographic Setting is Important in Determining Climate Compatibility for Station Relocations (Allegheny Mountains of West Virginia).

Now let's shift our focus to instrumentation exposure and its potential impact on the climate record. As discussed earlier, standards exist to insure compatibility and accuracy of measurements. Standards for how to set up instruments at climate stations are documented (NWS Observing Handbook No. 2: Cooperative Station Observations - July 1989-in process of being updated).

Temperature is often affected by factors which govern the local energy budget including:
- Sensible heat fluxes (type of surface)
- Latent heating (evaporative cooling from vegetation, water bodies, etc.)
- Radiative fluxes (solar and infrared, vertical & horizontal)
- Obstructions that restrict wind movement and ventilation
- Snow cover (which consumes heat in the melting process and reflects radiation)

Precipitation is most affected by factors that influence aerodynamics
- all precipitation gauges under catch
- shielded gages under catch less than non-shielded gages
- under catch is a much bigger factor when precipitation is frozen
- natural vegetation slows the wind and increases the catch

Figure 8 depicts a non-standard COOP MMTS exposure. To make matters worse, this is a Historical Climate Network station, which implies that the station is one of the best climatologically in the U.S. network. The rain gauge is situated only a couple of feet from a power-plant building, over a non-standard gravel surface (grassland is the surrounding environment) and within two feet of an exhaust vent and two metal stacks. Adherence to exposure standards decreases the risk of collecting data representative for only a very small local spot or mico-climate such as shown here.

Non-compliant Temperature Sensor at a Published HCN COOP Station
Figure 8: Non-compliant Temperature Sensor at a Published HCN COOP Station.

In the real world, non-standard instrument exposures of one type or another are common. It is difficult to meet all instrument exposure guidelines in the natural environment. There are space, power, vegetation, topographic considerations, and aesthetic constraints, especially when private property and homes are involved. However, some non-compliant situations are worse than others.

Unrepresentative Temperature Siting in an Isolated Grove of Trees Without Adequate Instrument Exposure Metadata
Figure 9. Unrepresentative Temperature Siting in an Isolated Grove of Trees Without Adequate Instrument Exposure Metadata.

Figure 9 depicts another example of a non-standard instrument siting. Standards call for the sensor to be located over ground representative of the surrounding environment. These young trees have grown a great deal since this NWS MMTS system was installed at this cooperative station. The presence of the trees is not recorded in the metadata.

We can take measures to avoid or mitigate non-compliant exposure situations. Understanding the purpose of the data you are collecting is a big step in the correct direction. Is the goal to obtain data for a very localized spot climate (a mountain pass with a highway or a frost hollow for agricultural growing season applications)? Assessing "spot", or extremely localized microclimatic and weather conditions are a valid need under certain circumstances (e.g., mountain pass data for transportation forecasts, research projects, etc.). Most frequently however, operational data are used for multiple purposes, many of which are not immediately known or understood and require that the station describe and represent the climate of a broader area. For these applications, you want to avoid locations that harbor non-representative spot mico-climates. Spot climates, by definition, are real and interesting, but may be representative for only a very limited locale. And indeed, on occasion they are justified, to help us understand the degree of fine-scale spatial variation.

Figure 10 illustrates an example of a poorly sited rain gauge. The gauge is situated only 2 feet from a 30 + foot high building. Siting guidelines call for obstructions to the gauge opening to be no more than half the height of the obstruction's distance to the gauge. In this case, the gauge, optimally, should have been no closer than 60 feet from the structure.

Dinosaur National Park (that is a natural wall, of rock, and this is an artificial wall, of a house). In the Dinosaur case, the site tied or set four new all-time monthly records almost immediately. Both exposures are likely to yield data with a high degree of uncertainty in their accuracy, especially under certain synoptic conditions, and even more with snow swirling and drifting or blowing from the roof.

Extremely Poorly Situated Non-Compliant Rain Gauge Exposure
Figure 10. Extremely Poorly Situated Non-Compliant Rain Gauge Exposure.

There are many examples of stations that push the boundaries of non-compliance to the limit. In fact, many climate data users argue that bad data is worse than no data. Keep this thought in mind when locating climate stations! Are you be willing to defend the site you selected for a climate station (new or relocated) in front of an audience of climate data users, customers, and scientists? If the answer is yes chances are good it's acceptable. If the answer is no, keep looking for another, more compliant location.

It is important to note that non-standard instrument exposures may not necessarily translate into bogus data with every single observation. Non-standard exposure means that data quality may vary, or put another way, the uncertainty in the quality and representativeness of the data increases. The validity of the data may vary with many meteorological variables such as wind speed and direction, amount of sunshine, time of year, etc. Some sites may produce high quality, representative data, while others may provide very poor quality, unrepresentative information.

Instrument exposure standards are really compromised with rooftop locations (Figure 11).

An Early Rooftop Meteorological Station
Figure 11: An Early Rooftop Meteorological Station.

While the number of NOAA rooftop climate stations has remained at about 40 for the last decade, the number of private rooftop stations has grown during that period into the thousands. Rooftop exposures have an advantage of increased instrument security and good exposure for wind sensors (standard height is about 33 feet). However, there are also drawbacks. Access for maintenance can be difficult and exposure for precipitation and temperature instrumentation is clearly non-compliant, being elevated to high above the ground. Additionally the instrument exposure is usually over environmentally non-representative surfaces (metal, black tar, shingle, stone etc.), while at the same time being close to a wide variety of roof surfaces which are subject to change.

The unrepresentative-ness of rooftop temperature and precipitation data was discovered long ago after studies quantified the biases. The late Professor Helmut Landsburg, considered the "father of climatology", stated in his 1942 "bible of climatology" textbook, "Physical Climatology" that:

"Climate derived from records of roof stations may by no means be representative of those at the ground level."

In another published paper 28 years later, Professor Landsburg again reiterated his concern about rooftop exposures with respect to the urban warming issue:

"They [rooftop stations] are certainly of little value in a full assessment of the climatic changes brought about by urbanization."

Rooftops make good observation sites if you live work, play, or grow your food on a roof. Unfortunately, few people do any of the above. Rooftop exposures have been shown to exhibit biases towards warm temperatures (both maximum and minimum) and lower precipitation when compared to ground based stations. The warm temperature biases likely result from extreme daytime heating of artificial rooftop surfaces, reduced cooling of the roof at night, and from heat flow from within the building, especially in winter. The biases can be substantial. One limited study indicates 5 to 10 degrees on summer days with bright sun and light winds. Biases have been found to vary significantly, depending on many factors (location on the roof, color of roof, type of roof surface (rock, metal, etc.) time of year, etc when compared to standard ground-based sensors.

On the flip side, if a station has been on a non-changing roof for decades, the site may have good continuity (value) for tracking climate change and variability. For some climate applications, consistency with a long record can be more important than accuracy with a shorter record. The best of both worlds is to have an exposure compliant, long-term station. Balancing these considerations for a long-established station requires some judgment and perhaps a difficult decision. In any case, proper metadata documentation will greatly assist future assessments of the station's usefulness and quality.

Let's review a summary of data from a recently relocated, published rooftop COOP climate station. There are still several dozen other rooftop COOP stations the NWS operates.

Figure 12 depicts the NOAA published COOP climate station on the roof of the Baltimore, Maryland Custom House in downtown Baltimore city.

The old Baltimore, Maryland Custom House rooftop COOP
Maxima>100 F
Maxima>90 F
Minima > 80 F
Baltimore (roof)
Baltimore (ground)
10 COOP mean
Figure 12. The old Baltimore, Maryland Custom House rooftop COOP and summary of comparison of overlapping data with surrounding ground-based stations.

The table to its right summarizes a comparison of 12 months of overlapping data that was collected on the rooftop and at the new relocated site (for data continuity), relocated several blocks away at ground level with other nearby standard, ground based stations.

A combination of the rooftop and downtown urban siting explain the regular occurrence of extremely warm temperatures. Compared to nearby ground-level instruments and nearby airports and surrounding COOPs, it is clear that a strong warm bias exists, partially because of the rooftop location.

Maximum and minimum temperatures are elevated, especially in the summer. The number of 80 plus minimum temperatures during the one-year of data overlap was 13 on the roof and zero at three surrounding LCD airports, the close by ground-based inner Baltimore harbor site, and all 10 COOPs in the same NCDC climate zone. Eighty-degree minimum are luckily, an extremely rare occurrence in the mid-Atlantic region at standard ground-based stations, urban or otherwise.

Temperatures can be elevated on roofs due to the higher solar radiation absorption and re-radiation associated with many roof surfaces including black tar, shingles, stone, and metal. During the colder months, ongoing upward heat transfer through the roof from the heated interior of the building also can contribute to the warm bias although stronger winter winds tend to create better mixing and minimize this impact.

Let's look at another example of the importance of instrument exposure on data representativeness and accuracy. It has long been understood that the height of a rain gauge above the ground affects the measured rainfall totals. Basically, studies show that the higher the gauge orifice, the lower the reported amount (Figure 13).

Variation of gauge catch with height for a given set of wind conditions as reported by Symons (1881)
Figure 13: Variation of gauge catch with height for a given set of wind conditions as reported by Symons (1881) (reprinted with permission from HydroLynx Systems, Inc. , 2002)

The quote below summarizes the results of an early study on this relationship:

"a rain gauge located on the top of a 30-foot tall house caught just 80 percent of the amount measured in a ground-level rain gauge. Similarly, a gauge on top of a 150 foot abbey tower caught just over 50 percent of the ground level catch."

Can you guess the year of the study?

a) 1950
b) 1971
c) 1899
d) 1815
e) 1769

If you chose (e), 1769, you are correct! Much of what we know about observational bias has been known for a very long time … it's more a matter of applying what we know.

Elevated Structures, Including Towers, Do Not Meet Basic Exposure Standards for Temperature and Precipitation Instrumentation
Figure 14: Elevated Structures, Including Towers, Do Not Meet Basic Exposure Standards for Temperature and Precipitation Instrumentation.

We now look at one final example of the importance of instrument exposure on temperature data as a paper submitted in 2003 for publication in the Bulletin of the AMS (Microclimate Exposures of Surface-Based Weather Stations - Implications for the Assessment of Long-Term Temperature Trends, Davey et. Al. CSU) notes inconsistencies in the vertical placement of sensors at existing published COOP stations (differences of several feet in some instances).

Figure 15 depicts the importance of adherence to the standard height of 5 feet above the ground for the placement of the temperature sensor.

XYplot of Degrees (F) and Hours of the Day
Figure 15: Hourly Surface Temperature Variability as a Function of Sensor Height.

Each curve denotes average hourly temperatures for 5 spring days at seven different heights are plotted for Seabrook, NJ (data from Thornthwaite, 1948). The sensor height differences are small, ranging from 0.3 feet above the ground to 21.3 feet

There are two features to note in the graph: First, there is a difference of up to 5 degrees F at the time of the maximums and minimums between with only 21 feet difference in elevation. Second, the daily temperature range (difference between maximum and minimum) is different at each elevation. Differences of only a few feet in the sensor height can make a one degree F difference in extremes for the day. This difference is significant in relation to the magnitude of the tenths of a degree climate signal researchers look for. Therefore, it is important we adhere to the standard sensor height (1.5 meters above the ground) to maximize detection of the climate change signal over time.

Bottom-line…adhering to the standard temperature sensor height of about 5 feet reduces uncertainty in the measurements, increases our ability to make accurate comparisons between sites, and assures a level-playing field amongst observation sites, especially when relocations are involved.

In summary, non-standard instrument exposures minimally at best increase uncertainty in the accuracy and comparability of data. At worst, they pollute and blur information and result in poor assessments and costly, incorrect decision-making. The magnitude of the impacts on the record can be complex and difficult to quantify. That's why adherence to standards is so important!

Recommended Exposure-Related Action

The following actions are recommended to minimize unwanted discontinuities in the climate record:

1) Do not operate climate stations in grossly non-standard exposures (rooftops, in close proximity to heat exchangers, etc.). If you cannot locate a station reasonably close to exposure standards, don't install it there! The data could be misleading and do more harm that good when used for assessments of broader local or regional conditions where the users may not be familiar with the "small print" or spot climate of the station's non-compliance. Consider other, more conforming locations.

2) Coordinate with your RCSPM and other climate services partners on any related issues (NCDC, RCCs, SCs).