Did the land cover around the station changed since the 80's? The change in temperature could be related to the change in land cover around the station (e.g. from agricultural land to urban and built-up). In my paper I showed that changes in land cover classes in a grid cell impact the temperature (due to the changes in the heat fluxes), wind speed and precipitation (De Meij and Vinuesa, Impact of SRTM and Corine Land Cover data on meteorological parameters using WRF, Atmos. Res., 2014). I guess that such impact would be visible in a sharp increase in temperature trend line.

Thank you Dr. Towe for your answer and interesting stats about US.

However I didn't mean to ask if it is possible in general, but if anybody observed such an air temperature increase somewhere in central Italy in the same period.

As I am not an expert of climate changes and global warming, I wonder if what I found was correct.

Thank you Dr de Meij.

I will read the paper you attached.

In effect I also found a remarkable decrease of precipitation (from 90 to 75 mm monthly mean)  and  wind speed increase from 4.5 to 7 km/h.

actually, you must consider a scale factor, passing from the very local temperature regime in a single observatory to the world scale ! I ignore if other researchers have already recognised what you discovered in the concerned meteo station in Southern Lazio, but everybody pointed out a general warming in last decade. You can address your request to our colleague Maurizio Sciortino, climatologist.

Grazie Sergio

As a matter of fact there are some factors to be taken into consideration.

First of all if there was a change in the land use, as already pointed out, which modified the climate at micro/local scale. Second: did they relocate  the station during the period considered? Did they change the sensors? Did they calibrated the sensors properly? Finally you could compare this series with a nearby series and see how large is the difference between them, if any, and then try to understand why.

Marina... it is a station of the Italian Meteorological Service (an organizational unit of the Italian Air Force and member of the World Meteorological Organization - WMO) so I feel quite confident that acquired data are reliable and consistent with highest standards and codes. I'll check other nearby stations, as you suggested, anyway.

very interesting Dr Towe, I will check this too.

i think the increase may be due to urban effects or other human activity otherwise please be careful

That change, if not produced by urbanization, can be produced by deforestation. It would be necessary to take into account Kenneth comments, and also to see if there was or there is any land use change during that time.

Dear Ivan, the local air temperature depends very much on the heat exchanges with the surrounding area. If this has been transformed in some way, the local temperature is not the same. However, before to say that the air temperature has changed, you must verify whether the thermometer, the screen against the solar radiation, its exposure

Kenneth is right that the situation is complex and that many series exist of uncertain quality, that have been largely used in performing climate statistics. To control, validate and correct a long series requires years of work, and this is not a popular task. I do not only refer to Italian series.

For instance, with a EU project we have revised the logest daily series in Europe

Camuffo, D., and Jones, P. (editors), 2002: Improved Understanding of Past Climatic Variability from Early Daily European Instrumental Sources. Kluwer Academic Publishers, Dordrecht, Boston, London, 392 pp. (also available as a special issue of Climate Change and the Padua series papers downloadable from my Research Gate page, year 2002).

We have done something as suggested in the nice article by Brooks (1921) but much more, checking and correcting day by day, 

A key problem of easy use of long series  is that long series followed national or international protocols that changed over time, and the bias due to the change appears in the same way and simultaneously in all the series. This is not a random error that might be mitigated by increasing the number of input series. The consequence is that the change is the same in all the series and interpreted as a climate signal. 

Personally, I prefer to use a limited number of accurate series of known quality. Not always a large number is a guarantee of better quality.  A large number of uncontrolled series may increase uncertainty or bias.

However, as Kenneth says, this has been discussed for years, but only a few scientists have applied to check and improve the quality of long series because this is extremely time consuming.

Dario

I appreciate all your comments and I checked that my station is operating (apparently unchanged) since 1986: same position, same instrument.

I've studied hot and heat wave days in South Europe (Portugal, Spain, Italy and Grecee) and they increased significantly in a high percentage of the study area. 

You can download the paper (it is in my researchgate profile).

Extreme temperature days and their potential impacts on southern Europe

Adrián Cardil, D. M. Molina, L. N. Kobziar

thank you Adrian.

Kenneth, you are right. There are many more examples of heat waves. Portugal 2003, Greece 2007, Rusia 2010, Australia in last years, Spain 1994, 2003 and others. Of course we had more heat waves 50 or 100 years ago...historical perspective is importante, I agree with you but It seems that heat wave days increased significantly in last 30 or 40 years. 

It is important to consider these days. For instance, they had consequences in large wildland fires, causing huge fires (Portugal 2003, Greece 2007, Australia, US, France.....). Although our suppression agencies decrease burned area in many areas, they did not the same under extreme temperature conditions.

Additionally, in some areas (Grecee), the annual number of high temperature days increased in a high percentage in the study period.

Thank you for your answer Kenneth! Feedback is always useful!

We have some studies about causes. However, we have a lot of ignition points every year. The point is that we need to know the conditions in which fires are able to grow with extreme fire behavior. We have thousends of ignition points every year but a small perfentage of fires burned more than 80% of the total burned area. Heat wave days and wind conditions are important in terms of burned area and fire size as you can see in my works in Spain, Italy..

We need to know all these factors and create fire tipologies as firefighters did in Catalonia (Spain). In this way, we will manage fires better!