Keeping in mind that our financial resources for harmonized measurements of common parameter sets for field testing in EnvEurope are quite limited, we need to think about if we could make use of remote sensing data. I could imagine (which may be far from reality) to use these data

• to substitute time- and money consuming measurement of e.g. primary productivity
• to get more or less consistent information about e.g. land use changes that took place at / around LTER sites over the last decades
  
• to test novel approaches like CO2 fixation rates or some other parameters we are usually not thinking of

This topic is just intended as a collection of conventional and maybe even crazy ideas, best with some background about the feasibility of what you propose.  I am really no expert in this area, but e.g. Ricardo Diaz-Delgado is quite experienced (as far as I remember) and I think some other people of the EnvEurope community too.

 

If there emerge any good ideas about the use of remote sensing data, we could look for sources / satellite data that could be used. We (UFZ) may offer some - perhaps free - access to satellite data hosted by the German Aerospace Center.

 

So, any ideas about it?   

Remote sensing is a quite promising technique for large scale data aquisition but it is not an

all-round solution for everything.

 

Remote sensing can be used to gather data on pressures (e.g. land use), state (e.g. biomass) or impacts (only in multitemporal way to indicate changes in state). But remote sensing is limited to drivers and processes working at a special scale (depending on sensors spatial resolution and temporal frequency), this never should be forgottn! In this context the need or usage of remote sensing should be target oriented following a predefined protocol for its interpretation.

 

Many attempts have been made to extrapolate or translate remote sensing data to larger scales (mostly because this is a quite convenient technique or the only available data source), but in my opinion this ever go ahead with more or less losses in indicatory power. Unfortunately these losses are hardtly described or analysed. Land management is a good example to describe the problem. Since remote sensing can only identify the distrubution of main crops and most of the environmental effect depends on the timing and intensity of measures on single plots (e.g. plant protection, kind of soil tillage), it would be unpossible to catch many of the most important land management impacts with the single remote sensing measure: crop diversity/crop allocation. Only with an intelligent combination of plot measurements and remote sensing data you can meet many of the sensitive land use practices.

 

All in all i suggest to relate the need for remote sensing data to the ecosystem functions or the indicators for them and to test very accurately if they provide data for the appropriate scales and process drivers. The usage of remote sensing should follow a detailed description about their scale reference and reference to what they indicating. I would like to raise again the question of "instruction leaflets" for the single ESF and indicators.

 

We first have to decide, what drivers do we need to address and only afterwards how. The remote sensing data than can help us to make the different sites more comparable.

 

 

#Previously Mark Frenzel wrote:

Keeping in mind that our financial resources for harmonized measurements of common parameter sets for field testing in EnvEurope are quite limited, we need to think about if we could make use of remote sensing data. I could imagine (which may be far from reality) to use these data

• to substitute time- and money consuming measurement of e.g. primary productivity
• to get more or less consistent information about e.g. land use changes that took place at / around LTER sites over the last decades
  
• to test novel approaches like CO2 fixation rates or some other parameters we are usually not thinking of

This topic is just intended as a collection of conventional and maybe even crazy ideas, best with some background about the feasibility of what you propose.  I am really no expert in this area, but e.g. Ricardo Diaz-Delgado is quite experienced (as far as I remember) and I think some other people of the EnvEurope community too.

 

If there emerge any good ideas about the use of remote sensing data, we could look for sources / satellite data that could be used. We (UFZ) may offer some - perhaps free - access to satellite data hosted by the German Aerospace Center.

 

So, any ideas about it?   

 

I“dare” to recall that the project EnvEurope is also expected to have strict relations with GMES and to contribute to the “in situ” component of GMES.

 

Most services and products of GMES (some also available on line: see the GMES website) concern, actually, the Earth Observation from space: we should start to consider them in the way that Mark and Michael are suggesting. In the indicator choice it could be important to take into consideration, as much as possible, also the contribution that we can obtain from the GMES services and products and vice versa (very simply said: which are the parameters that may be “added” and suggested from our side to GMES and the interactions, and so on, among “our” and “their” observations).

To this respect, it is also important to know which, if any, of the EnvEurope (and LTER-Europe) sites/institutions are already involved, somehow, in GMES.

Last but not least: one of the members of the EnvEurope Advisory Committee, Dr. Nadia Pinardi, has been chosen also for her involvement in the Scientific Board of GMES and she could give us suggestions and advices about all this.

 

Previously Mark Frenzel wrote:

Keeping in mind that our financial resources for harmonized measurements of common parameter sets for field testing in EnvEurope are quite limited, we need to think about if we could make use of remote sensing data. I could imagine (which may be far from reality) to use these data

• to substitute time- and money consuming measurement of e.g. primary productivity
• to get more or less consistent information about e.g. land use changes that took place at / around LTER sites over the last decades
  
• to test novel approaches like CO2 fixation rates or some other parameters we are usually not thinking of

This topic is just intended as a collection of conventional and maybe even crazy ideas, best with some background about the feasibility of what you propose.  I am really no expert in this area, but e.g. Ricardo Diaz-Delgado is quite experienced (as far as I remember) and I think some other people of the EnvEurope community too.

 

If there emerge any good ideas about the use of remote sensing data, we could look for sources / satellite data that could be used. We (UFZ) may offer some - perhaps free - access to satellite data hosted by the German Aerospace Center.

 

So, any ideas about it?   

 

Previously Alessandra Pugnetti wrote:

I “dare” to recall that the project EnvEurope is also expected to have strict relations with GMES and to contribute to the “in situ” component of GMES.

I can inform that it exists a specific project, called GISC, started in Jan 2010, i support of the GMES Bureau and based on SEIS. I'm reading the documentation and I think it will be important to link with them. It will be an issue of Action 4, after a previosly, necessary, meeting with Dr. Nadia Pinardi.

 

Information about GISC are here: http://gisc.ew.eea.europa.eu/gisc-project

 

 

Alessandro

just a few statements refering to Michael and Alessandra:

• we should of course be aware of all limitations of remote sensing. 
• we need an overview about what kind of data are available (for us; free or for little money) and could be meaningful (for us), the frequency of data aquisition, the time span from the past till today covered by the data, the resolution etc. I am sure this information could be at hand soon. I will ask colleagues at the UFZ and take a look at the literature. It would be nice if this can be done by other people too to get a complete picture.
• before we go more into detail (perhaps one issue of the Halle workshop) what kind of remote sensing data we would like to include to I propose to keep it either at the level of collecting options or to strictly refer to the conceptual framework and the gaps we need to fill
• our plots / simple sites / complex sites are embedded in landscapes and in the long run they are supposed to be affected by land use change at the landscape level with implications on regional species pools etc. Therefore I make an argument for including the surrounding in this way.

I propose to prepare a document (table) at the google docs space of EnvEurope where everyone can contribute for an overview. Tomorrow I will prepare a template and come back to this issue again.     

Remote sensing is complementary to in situ field monitoring and would contribute with overview information. However, it should be controlled in the field and by chance I got a question on tree age last Friday when remote satellite information gave an age of 30 years for one site stand actually being with trees over 100 years old. Be aware of quality control. However, we have also carried out Lidar scanning to improve primarily the ground elevation map and hope for much better information. Here field work would no be feasible or competitive. The scanning and interpretation is though somewhat costly.

Regards Lars lundin

I think that the use of remote data can be a valuable approach for the harmonization of data assessment, and thus comparability of the data. 

However, in some ecosystem types, this method will probably work better than in others: while I see a large potential in most terrestrial ecosystems and lakes, it might be more difficult in streams and rivers. Looking from above, you often do not see their surfaces at all, or only partly. Therefore, river ecosystem variables like estimation of their productivity (if there is any) or CO2 fixation/release can hardly be assessed remotely. Remote sensing data can only be assessed in the landscape context, in which river ecosystems are embedded. 

Thus, I think we also need to consider the different spatial scales of the ecosystems on which we are working.

 

I won't hide this is an attractive topic here posted and quite irresistible for me to reply. I agree with most of the added comments and just wanted to invite you to consider a pragmatic view based on my experience.

Following Michael's comment there are several relevant points to keep in mind when using RS tools to support ecological long term research and monitoring:

1. Ground-truth is essential to validate RS proxy values: biophysical parameters and thematic attributes have to be measured at plot scale to be used for RS products validation. Without a continuous and coherent sampling effort at plot scale any help from RS algorithms will be really valuable for any long-term site.

2. RS offer valuable help for long-term monitoring at landscape and plant community scale: so our indicators have to be identified at such scales and not much more detailed. Any attempt to assess at coarser scales will require specialization both in analysis and remote sensing sources.

3. RS has been largely proved to be useful as a tool to reconstruct historical trajectories, so it can be used to set up a reference trend for many ecological processes against which evidence anomalies.

4. There is the imperative need to employ RS specialists at site level in order to easily get the data systematically integrated into the long-term monitoring database allowing time analysis of RS data and being aware of caveats and advantages as well as of new satellites and sensors.

Having this in mind, one interesting offer is the set of products offered by MODIS which apply validated algorithms to produce freely accesible periodic images of any point in the globe such as LAI, land use cover, Fire products, Evapotranspiration, among many others.

So, in conclusion, I think there is an interesting open ground in EnvEurope to start with by introducing a new working group on RS as a tool to complement long-term monitoring of certain parameters looking forward to simplify our day-to-day work and not the opposite.

Cheers

 

 

Previously Mark Frenzel wrote:

Keeping in mind that our financial resources for harmonized measurements of common parameter sets for field testing in EnvEurope are quite limited, we need to think about if we could make use of remote sensing data. I could imagine (which may be far from reality) to use these data

• to substitute time- and money consuming measurement of e.g. primary productivity
• to get more or less consistent information about e.g. land use changes that took place at / around LTER sites over the last decades
  
• to test novel approaches like CO2 fixation rates or some other parameters we are usually not thinking of

This topic is just intended as a collection of conventional and maybe even crazy ideas, best with some background about the feasibility of what you propose.  I am really no expert in this area, but e.g. Ricardo Diaz-Delgado is quite experienced (as far as I remember) and I think some other people of the EnvEurope community too.

 

If there emerge any good ideas about the use of remote sensing data, we could look for sources / satellite data that could be used. We (UFZ) may offer some - perhaps free - access to satellite data hosted by the German Aerospace Center.

 

So, any ideas about it?   

 

I hoped that Ricardo would not resist in commenting to one of his favored issues. I will reserve a time slot for potential support of remote sensing in LTER monitoring at the Halle workshop (anybody able to give an incentive presentaion about it?)!

At the Google Docs for EnvEurope (login necessary) area I created a table in the Action2 folder where I started to collect some information about data gathered by satellites. Please take a look at it and add items you are aware of.

What I find quite attractive is the the potential of RS for reconstructing changes in land use which could be interesting to create a common database for all sites covering the past and the future. We already have some ideas in mind to connect this with ecosystem services.

A task force group of people experienced with RS would really be necessary. As advisors as well as people looking for, processing of and interpreting RS data.

How to make these resources available in terms of experts and funding (should be done by EnvEurope money)? Is it wise to establish this as a central service or in a distributed way?   

Ciao, Mark

... here comes the missing link to the GoogleDocs document:

https://spreadsheets.google.com/ccc?key=0AjVUzeeAkBy8dHBkLU8wbG1MSWxNYkYyOUJoYWxEMEE&hl=en_GB&authkey=CLrdsrQM

Dear Mark,

 I do believe that you are traveling in the right direction. We started discussing RS and its application to meet the objectives of our project, however I miss real concrete suggestions. First of all information of RS relly gives us an opportunity to analise historical data and create basis for the analysis in future. Therefore first of all we should decide which parameters are the best representing our LTER sites. For example land cover should be accepted for everybody. But even in this case the best salutation would be that only one performer collects and analyzes the data for all sites by the same methodology and only afterwards every country could check it. United methods and RS sources are the basis for the correct and comparable information. For this purpose we can use US GS Global Land Series http://edcsns17.cr.usgs.gov/EarthExplorer/ or even CORINE IMAGE. Some good reference projects are FOREST by JRC and WFRA 2010 or Remote Sensing survey by FAO. Our experience indicated that GMES is far from being the best of RS info.

For other parameters more detailed information is needed. Therefore there we should decide which parameter is the best and cheapest to be used. For example in forest ecosystem stand or even tree crown parameters could be used for the analysis, but in this case additional financial sources are needed. Airborne imaging based on standard fixed wing aircrafts becomes economically realistic. Airborne imaging using ultralight aviation and UAV (unmanned aviation vehicle) becomes widely available and capable of delivering high quality data. Hiperspectral cameras are coming in the nearest future. Lidar scanning of course in this series of information is most acceptable but due to a lot of knowledge gaps and expenses their application to our purpose remains problematic.

 

Therefore I would like to suggest first of all to decide which of the parameters could represent our sites and their surrounding best of all (from 1×1 km to 50×50 km). It should be estimated using the most acceptable and free available RS sources. The best way, as I have already mentioned, it should be performed by one country and checked by every country.

Second, we should decide which parameters we need. In this case each of participants could present their own experience in this field, the sources of information, methodology and prices. In this case some questionnaire would be necessary for collection such information.

Algirdas

Although I am really not an expert in this field, it seems that remote sensing has been also used in aquatic systems (lakes, large rivers) e.g. to measure water turbidity (suspended solids), dissolved organic carbon concentration, phytoplankton productivity (chlorophyll-a),  to map blooms of blue-green algae and abundance of macrophytes. It seems that the usage of remote sensing in aquatic systems is under development, although the usage of it is not yet as advanced as in terrestrial systems.

 

Kimmo