I have never measured light in the forest, but in the chamber with the cenosis of plants we used photometer and luxmeter. So in the forest, i think you can just measure with such device on the same level/on any level which you are interested in. Now there are many modern device to measure and connect to the computer.... Also you can measure spectrum, which is also different and can influence flowering and propagation. 

The most common method to measuring relative light in the forest is to use hemispherical photography. you can use a Fishey lens with a view angle of 180 degree  and take photos from canopy. then you can obtain relative light in light analyzer softwares like Gap Light Analyzer 2. if you search "Hemispherical photography" in different forestry journals you will be find so many articles that help you to achieve your goal.

good luck

Luis, colleagues above just answered your question better than I could. In fact you can use a portable radiometer. Some models may even use the channels you might want from remote sensing images so that you can correlate field and aerial information. I myself I don´t do that kind of measurement. Best wishes.

Jorge

Have a look at Decagon's ACCUPAR LP-80, an affordable ceptometer, that allows PAR and LAI measurements.

A first description of the canopy with hemispherical photographs and Gap Light Analizer (GLA) will allow you an easy and extensive description of the light that is trasmitted to the understory. However, in my own experience, this must be compelted with light measurement with light sensors (this will allow you to calibrate the values of GLA). Onset offers a cheap devices that registers light and temperature in a small size.

I think the easiest way to do it remains the hemispherical photography. It is easy, fast and requires no spectacular or expansive equipment. But you should evaluate the correct exposure in order to avoid errors. The photos with automatic exposure are not reliable, but there are some known protocols to use - http://www.sciencedirect.com/science/article/pii/S0168192305001917.

Generally measuring radiation under forest canopy (dishomogeneity in canopy characteristics, closure and gaps, different species...) needs different approaches from measuring in controlled situation like greenhouse...There are various methods and instruments with different costs and accuracy related to filed conditions. On the basis of elements you have given I suggest to use linear light sensor with multiple measurements (ceptometers) with a proper sampling method. At the same time you can couple this approach with digital hemispherical photo with canopy image analysis.

I add some references

ceptometer

http://www.afs-journal.org/index.php?Itemid=129&option=com_article&access=dkey&dkey=10.1051/forest:19960238&lang=en

http://www.tandfonline.com/doi/abs/10.1080/11263504.1997.10654167#.VH7dWTGG9c4

http://www.sciencedirect.com/science/article/pii/S0378112797002697

digital

http://www.sciencedirect.com/science/article/pii/S0168192312002766

http://www.sciencedirect.com/science/article/pii/S016819231400210X

Agricultural and Forest Meteorology (Impact Factor: 3.89). 01/2013; 168(2013):130-139. DOI: 10.1016/j.agrformet.2012.09.002

I've used the Licor LAI-2200, with two arms, and that's a good thing, because the first arm stay out the forest and the other come together measuring the LAI on the forest understory. I think this is the easiest method.

http://www.licor.com/env/products/leaf_area/LAI-2200/

I suggest hemispherical photography also, because with proper programs, you can calculate amount of radiation for given periods, such as vegetation season, or only for the life span of the flover you interested in.

Method which you may use depends on needed accuracy of light amount estimation and your budget. Most of the previous answers suggest equipment with high accuracy but it is costly. If you just want to estimate amount of light to compare it between some sites then spherical densiometer is enough (costs about 100 USD). It is possible to buy it via amazon: http://www.amazon.com/Forestry-Suppliers-Spherical-Densiometer-Convex/dp/B0096EOF20

If you want to develop cheap and rapid method for estimation of the canopy closure (opposite to amount of light), then there are several easy self-made instruments that you may prepare and use, e.g.: http://www.google.pl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&ved=0CEYQFjAE&url=http%3A%2F%2Fwww.rbnc.org%2Fschoolunits%2Fschlphoto%2Fterreco%2FDensi_inst.pdf&ei=glqBVKv7IcPmUpi3gOgP&usg=AFQjCNFRooHBwzxOU3zzx7ER066P38hV3g&bvm=bv.80642063,d.d24

Dear Luis,

i would agree with Ardalan. Using a hemiphoto is the simplest and now with modern technology the easiest way to calculate light interception on the forest floor which is important for regeneration of any seedling of tree or shrub species. If you can get a nikon coolpix 990 with a fish eye lens then the best software is developed by a japanese university and available on line for free..http://www.agr.nagoya-u.ac.jp/~shinkan/LIA32/index-e.html

There are also other expensive models but basically they measure the same thing.

Good luck.

Do NOT use instruments that measure solar radiation such as photometers when in forests because these measurements are very sensitive to changes in ambient solar radiation, due to time of day, weather and season. 

You should use either a measure of canopy closedness or canopy openness to do this. 2 options you might consider:

Hemispherical photos are an accurate and precise method. You can account for things like slope aspect and solar aspect to get good measures of how much light is reaching the forest floor. 

Spherical densiometer is an efficient and affordable tool to measure canopy closedness. They will not tell you how much solar radiation passes through the canopy, but it is useful to determine the amount of solar radiation at one site relative to another site. Efficiency is often a valuable asset when working in the field.  

Dear Dr Luís Quinta-Nova,

I would like to thank you for your question. The best method for measure the solar radiation is the canopy temperature using a machine you can test the temperature and according to that you can identify the genotypes tolerance to heat or not. The erect leaf or not

with best regards

Khaled

My experience agrees with Sutherland's recommendations: use hemispherical photographs (with Gap Light Analyzer free software) or a densiometer. Light meters are too sensitive to spatial and temporal variation unless you have huge numbers of sensors with continuous recording.  Two small pieces of advice: measure the light at the height of the plant, which may not be the most comfortable height for people, and stay away from commercial 'fish-eye' lens because they do not provide a full hemispherical image.

Here is an instrument that would be good for your purposes, it measures both LAI and PAR and is great for using in the field. 

http://www.cid-inc.com/products/leaf-area-lai/plant-canopy-imager

After exploring my data, I'd like to add to my first answer: Whatever method you use, take multiple measurements of canopy openness for each plot. I measured trees in 11.28 m large radius plots. For each large radius plot I took three canopy openness measurements at 6 m from plot center and at a random azimuth. At each of these locations I measured the biomass of an understory plant of interest. In data exploration I observe that the relationship between understory biomass and canopy openness is very noisy, but when I average the three measurements from each plot together a clear pattern begins to emerge. The canopy structure of the forests I study is highly variable (coastal temperate forest) so next time I will take even more canopy openness measures (probably five). This is very easy to do with a spherical densiometer because it takes just a few minutes to take a measurement. 

 We investigated the illumination under the canopy of the forest using a portable luxometer (Люксметр Ю-116). The variability of the indicators is very large. But we increased the repeatability to 200 and measurements are made very quickly at the site (about 30 min). Measurements were done by envelope method (diagonals and perimeter) in 1-2 meters. Simultaneously we made measurements in the open area. And it was possible to determine the percentage of passage of visible solar radiation through the forest canopy and the conditions of illumination of the lower phyto- horizons.

Regards