Satellite image reveals the floristic variation of the Amazonas region

In her dissertation, Principal Research Scientist Sirpa Thessler from MTT (Agrifood Research Finland) combined field observations made in the lowland rainforests of the Amazonas region with satellite image analysis and was also able to predict the floristic variation of unexplored areas.

By knowing the differences in species composition between areas, conservation and land use can be planned and carried out effeciently. When the floristic variaton of the region is known, new conservation areas can be planned effectively to complement the existing ones.

INDICATOR SPECIES WERE OBSERVED IN THE FIELD

In her dissertation, Ms Thessler examined the field observations made by the University of Turku Amazon research group in the rainforests of Ecuador, Peru and Costa Rica and combined these with Landsat satellite images taken from the same areas.

The species richness in the rainforests is immense, for which reason observations in the field are practically always limited in certain indicator species. They provide information about the environmental variation of the region, such as the nutrient content and humidity of soil, thus also describe the general patterns of flora .

In Thessler’s study, the indicator species included grass- and shrub-like Melastomataceae species and ferns in the understorey as well as tree and palm species in the canopy layer.

PIXEL VALUES REVEAL VARIATION

Field observations on hundreds of plant species were statistically summarised in three dimensions, in order to be able to present the satellite image based predictions of compositional variation in a single map. In addition various vegetation types were classified from the satellite images.

“I found a statistical association between the field observations and the pixel values of the satellite images, which led me to conclude that the images are indicative of the floristic patterns,” Ms Thessler states.

The researcher was equipped with field observations from a combined area of 47 square kilometres: based on these, she was able to predict the flora of an area of 4,100 square kilometres. Thessler points out that by satellite images we do not achieve a level of accuracy equal to field observations, but in rainforests with practically impenetrable forest they constitute one possible method.

“Surveying the flora of rainforests by means of field observations is difficult, time consuming and expensive. Even a rough estimate made with the aid of satellite images is better than nothing,” Ms Thessler sums up.

NEIGHBOURS PROVIDE RELIABLE INFORMATION

Sirpa Thessler’s dissertation also revealed that the “k Nearest Neighbour” method used for forest inventories in Finland and several other European countries also provides reliable information when it comes to tropical rainforests.

The method involves the examination of a single pixel in a satellite image and searching the pixels, named neighbours, that are closest in their spectral values and from which field observations are available. The letter “k” refers to the number of neighbours. Based on the known neighbours, the floristic class, for example, or another indicator of the species composition can be predicted for the pixels over the study area.

“The method seems quite promising. It produced good accuracies in predicting the floristic variation of a rainforest,” Thessler relates.

Media Contact

Ulla Jauhiainen alfa

More Information:

http://www.mtt.fi

All latest news from the category: Earth Sciences

Earth Sciences (also referred to as Geosciences), which deals with basic issues surrounding our planet, plays a vital role in the area of energy and raw materials supply.

Earth Sciences comprises subjects such as geology, geography, geological informatics, paleontology, mineralogy, petrography, crystallography, geophysics, geodesy, glaciology, cartography, photogrammetry, meteorology and seismology, early-warning systems, earthquake research and polar research.

Back to home

Comments (0)

Write a comment

Newest articles

Superradiant atoms could push the boundaries of how precisely time can be measured

Superradiant atoms can help us measure time more precisely than ever. In a new study, researchers from the University of Copenhagen present a new method for measuring the time interval,…

Ion thermoelectric conversion devices for near room temperature

The electrode sheet of the thermoelectric device consists of ionic hydrogel, which is sandwiched between the electrodes to form, and the Prussian blue on the electrode undergoes a redox reaction…

Zap Energy achieves 37-million-degree temperatures in a compact device

New publication reports record electron temperatures for a small-scale, sheared-flow-stabilized Z-pinch fusion device. In the nine decades since humans first produced fusion reactions, only a few fusion technologies have demonstrated…

Partners & Sponsors