It will depend on your needs... (resolution, revisiting frequency, etc...)

You can check for details about several Free Satellite Imagery Data Sources at this website: http://gisgeography.com/free-satellite-imagery-data-list

Sentinel-2 from Copernicus program could be a good one..

Really, there is no better satellite data to everything. This inevitably depends on how the primary objective is defined: Forest monitoring versus forest cover classificatio, time scale, spatial resolution, etc. I suggest you: Sentinel (Stand level modelling), Landsat, Modis, and AVHRR (Biome level modelling). However, if yo want to study the forest based on tree level modelling you will needd a very high resolution data.

There are no silver bullets.

The best choice can be done based on the specific aim.

Thank you all, in fact my aim is to study a specific species in the forest, is satellite data like Landsat or sentenel would be useful. so what do think?

How big are the trees you intend to explore?

What is the spectral resolution that you need?

Are there areas with specific species or is there a mix of trees and the aim is to allow detection down to the tree level?

What is to be studied? biodiversity detection? fire hazards? nutrients assessment? .......

Answers to questions like the above mentioned might provide the information regarding the spatial, spectral and temporal resolutions needed and the most potentially suitable satellite for your research.

Thank you for help,

How big, about 20 m high, circumference 04 to 06 m. we have areas with specific species, so i need to map the distribution of population over 5 million hectares.

So we have:

- The aim is mapping species distribution.

- ~50,000 square Km to cover.

- Trees with diameter of 1.3-2m (What tree is it to have - 20m high and up to 2m canopy diameter?).

You know the area therefore the homogeneity of species in it. Are there areas with one tree species? are there bushes and grasses between the trees?

How many species? do you care about specific species or just trees in general?

You can get the answers for these and more so you will know the pixel size you want to work with, the spectral bands that you need to capture what you look for and how often do you need to obtain data.

Will 30m pixel size would work for your land cover? would the landsat band formation be good to obtain the phenomena you are looking for? are there images of your area obtained by landsat (probably yes).

Sentinel2 will probably have better spatial and spectral resolutions but might be more difficult to find images.

As said there is no silver bullet - as you have a defined aim and based on your familiarity with the landscape and its cover, the pixel size needed can be obtained. Based on the aim the needed spectral bands are to be considered and based on that you should look for the satellite that will fit you best.

Thank you for help

The main challenge is if you want to identify one specific species among others..

In that case you will probably need a satellite with higher spatial resolution coupled with some field work...

Good luck!

Thank you Mr Joaquin, i will take a look

Thank you Mr Ayman, i agree with you, i thought may be some tried to do this with Landsat image. you spoke about VHR data, how can i get it please?

Hi all, Indeed, according to the current availability, the best choice for your study would be Sentinel 2 (HR of 10m) unlike Landsat (15/30 m). But fieldwork will always be necessary for identifications and corrections. In some cases, in-depth work on spectral signatures (with available spectrometer) may be required to refine the classification.