Publications

Versión en español

Publications of the team

Articles of the MonitorEO team that answered questions like yours, with what they did and how to ask MonitorEO for it. The assistant looks for them when you do not know how to ask, or ask it for examples. Grouped by the theme of MonitorEO of their main variable (Concepts); the questions are examples, and sometimes MonitorEO gives only part of the analysis.

Vegetation

What explains where and how ecosystem functional diversity, measured with NDVI-based functional types, changes on the Yamal Peninsula?

Morgan Tassone, Howard E Epstein, Amanda H Armstrong et al. (2026). Drivers of spatial and temporal heterogeneity in ecosystem functional diversity on the Yamal Peninsula, Siberia, Russia. Environmental Research: Ecology. https://doi.org/10.1088/2752-664x/aeaf5d

The study mapped Arctic Ecosystem Functional Types from seasonal MODIS NDVI metrics on the Yamal Peninsula (2001-2018) and used their number in small windows as an index of functional diversity. Landscape heterogeneity (elevation, hydrology, substrate) and continentality shaped its spatial pattern, while summer warming, earlier snowmelt and more moisture explained its increase over about a third of the peninsula.

With MonitorEO: NDVI (Modis), TMEAN (ERA5-Land), ELEVATION (Copernicus); trends, map of the trend, map, time series; period 2001-01-01 / 2018-12-31; region a drawn polygon over the Yamal Peninsula (big region: MODIS, annual step). Ask the assistant: "Annual trend of the maximum MODIS NDVI over the Yamal Peninsula (drawn polygon) between 2001 and 2018"

MonitorEO gives the NDVI metrics (annual mean, maximum), their trends and the climate drivers; the EFT classification and its richness are computed outside.

Do Ecosystem Functional Types derived from satellite EVI explain the differences in net CO2 exchange measured at flux towers?

Beatriz P. Cazorla, Ana Meijide, Javier Cabello et al. (2026). Satellite-derived Ecosystem Functional Types capture ecosystem functional heterogeneity at regional scale. Biogeosciences. https://doi.org/10.5194/bg-23-1223-2026

Ecosystem Functional Types were built from the 2001-2014 MODIS EVI series (annual productivity, seasonality and phenology) and compared with the net ecosystem exchange of CO2 measured by eddy covariance at 50 European sites. Different EFTs had distinct NEE magnitudes and dynamics, and EFTs explained NEE slightly better than plant functional types, with the advantage that they can be recomputed every year.

With MonitorEO: EVI (Modis); map, time series, trends; period 2001-01-01 / 2014-12-31; region {"point": ["<lon>", "<lat>"], "buffer": 1000}. Ask the assistant: "Monthly series of MODIS EVI from 2001 to 2014 within 1 km of the flux tower at [lon, lat]"

MonitorEO gives the EVI annual mean (productivity), maximum and minimum and their series; the seasonality (CV), the date of maximum and the EFT classification are computed outside (e.g. with the GEE code ajpelu/eftsn_gee). The flux data are not in MonitorEO.

How are the ecosystem functional types of the Arctic tundra, based on the seasonal dynamics of NDVI, distributed, and do they match vegetation and bioclimate maps?

Amanda Armstrong, Domingo Alcaraz-Segura, Martha Raynolds et al. (2024). Ecosystem functional types of the circumpolar Arctic tundra based on the seasonal dynamics of vegetation productivity. Environmental Research: Ecology. https://doi.org/10.1088/2752-664x/ad4beb

Ecosystem functional types were derived from the seasonal MODIS NDVI dynamics at 250 m over the whole circumpolar Arctic tundra and compared with the bioclimate subzones and vegetation units of the Circumpolar Arctic Vegetation Map. They broadly agree with those maps but reveal much functional variability within each unit, with richness increasing southwards and high heterogeneity in mountains of Alaska and Siberia.

With MonitorEO: NDVI (Modis); map, map of the trend; period 2001-01-01 / 2020-12-31; region {"ecoregion_at": [-149.6, 68.63]}. Ask the assistant: "Map of the mean of the annual maximum MODIS NDVI from 2001 to 2020 in the tundra ecoregion that contains the point [-149.6, 68.63] (Toolik Lake, Alaska)"

temporal "annual", metric "max", interannual "mean". The EFT classification is outside MonitorEO; snow leaves few valid pixels outside summer.

How are the ecosystem functional attributes and types of Sierra Nevada computed in Google Earth Engine?

BP Cazorla, D Alcaraz-Segura (2023). ajpelu/eftsn_gee: first release. Zenodo. https://doi.org/10.5281/zenodo.7524973

First release of the Google Earth Engine code behind the ecosystem-functioning dataset of Sierra Nevada. It complements the published maps of EVI-based functional attributes and types.

With MonitorEO: EVI (Modis); map, time series; period 2001-01-01 / 2018-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Monthly series of MODIS EVI from 2001 to 2018 in Sierra Nevada National Park"

MonitorEO gives the EVI series and annual metrics; this code adds the seasonality, the date of maximum and the EFT classification.

What are the primary production, its seasonality and phenology, and the functional diversity of the ecosystems of the Sierra Nevada Biosphere Reserve?

Beatriz P. Cazorla, Javier Cabello, Andrés Reyes et al. (2023). A remote-sensing-based dataset to characterize the ecosystem functioning and functional diversity in the Biosphere Reserve of the Sierra Nevada (southeastern Spain). Earth System Science Data. https://doi.org/10.5194/essd-15-1871-2023

This data paper releases maps of three Ecosystem Functional Attributes (annual EVI mean as a proxy of productivity, its seasonality and the date of the maximum) from MODIS MOD13Q1 EVI for the Sierra Nevada Biosphere Reserve since 2001, with the derived Ecosystem Functional Types and their diversity, rarity and interannual variability. It is a baseline to monitor ecosystem functioning in the protected area.

With MonitorEO: EVI (Modis); map, map of the trend, trends; period 2001-01-01 / 2018-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Map of the annual mean of MODIS EVI (mean between years) from 2001 to 2018 in Sierra Nevada National Park"

map_layer with temporal "annual", metric "mean", interannual "mean". For the whole Biosphere Reserve search "Sierra Nevada" with find_protected_areas. Seasonality, date of maximum and EFTs are in the dataset (PANGAEA) and its GEE code (ajpelu/eftsn_gee).

How diverse and irreplaceable are the habitats of protected areas worldwide, and where are similar areas outside them?

Javier Martínez-López, Bastian Bertzky, Marine Robuchon et al. (2023). Assessing habitat diversity and potential areas of similarity across protected areas globally. Ecological Informatics. https://doi.org/10.1016/j.ecoinf.2023.102090

Remote-sensing variables and modelling (eHabitat+) were used to divide protected areas into habitat functional types and to map areas with similar conditions in their ecoregions. New habitat diversity and irreplaceability indices allow ranking and comparing protected areas, illustrated with six cases.

With MonitorEO: NDVI (Sentinel-2), LSWI (Sentinel-2), SLOPE (Copernicus); download of variables (inputs of a model); period 2020-01-01 / 2023-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Download for eHabitat+ the Sentinel-2 NDVI (percentiles 5 and 95), the median LSWI from 2020 to 2023 and the slope of Sierra Nevada National Park and its ecoregions, on aligned grids"

download_variables builds exactly these inputs (study area and ecoregion grids); the segmentation into habitat types and the indices are computed in eHabitat+.

How much do the productivity, vegetation water, albedo and sensible heat of ecosystems resist a wildfire, and how fast do they recover?

Bruno Marcos, João Gonçalves, Domingo Alcaraz-Segura et al. (2023). Assessing the resilience of ecosystem functioning to wildfires using satellite-derived metrics of post-fire trajectories. Remote Sensing of Environment. https://doi.org/10.1016/j.rse.2022.113441

Using MODIS series from 2000 to 2018, the study followed the post-fire trajectories of four dimensions of ecosystem functioning after the 2005 wildfires in the north-west of the Iberian Peninsula. Primary production dropped abruptly but recovered within about two years, water content recovered more slowly, albedo changes sometimes lasted for years, and sensible heat effects faded after about one year.

With MonitorEO: EVI (Modis), NDWI (Modis), ALBEDO (Modis), LST (Modis); time series, comparison of periods, map of the comparison; period 2000-02-24 / 2018-12-31; region a drawn polygon over an area burned in 2005. Ask the assistant: "Monthly series of MODIS EVI, NDWI, albedo and LST between 2000 and 2018 in a drawn polygon over an area burned in 2005 in northern Portugal"

Sensible heat is approached with LST; the MODIS NDWI collection ends in February 2023. The resistance and recovery metrics are computed outside MonitorEO.

How do the sensor (MODIS or Landsat) and the scale influence ecosystem functional types and their diversity?

Lingling Liu, Jeffrey R. Smith, Amanda H. Armstrong et al. (2023). Influences of Satellite Sensor and Scale on Derivation of Ecosystem Functional Types and Diversity. Remote Sensing. https://doi.org/10.3390/rs15235593

Using Costa Rica as a case, EFTs and their diversity were derived from 250 m MODIS and 30 m Landsat with different methodological choices, at national and local extents. Landsat separates more EFTs thanks to its finer pixels, MODIS is better in very cloudy areas, and diversity measured at smaller extents is higher but nested within that of larger extents.

With MonitorEO: NDVI (Modis), NDVI (Landsat), CF (Modis); map, statistics of a map; period 2015-01-01 / 2020-12-31; region {"country": "CRI"}. Ask the assistant: "Map of the annual mean of MODIS NDVI from 2015 to 2020 in Costa Rica, and the MODIS cloud frequency (CF) of the same period"

Compare the same map with Landsat over a smaller area (a protected area) to see the effect of the sensor; CF shows where clouds limit Landsat. EFT classification outside MonitorEO.

What are the advantages of using satellite-derived ecosystem functioning attributes in ecological niche models?

Adrián Regos, João Gonçalves, Salvador Arenas‐Castro et al. (2022). Mainstreaming remotely sensed ecosystem functioning in ecological niche models. Remote Sensing in Ecology and Conservation. https://doi.org/10.1002/rse2.255

This review argues that ecosystem functioning attributes from satellites (productivity, seasonality and phenology of vegetation indices, surface temperature, albedo, evapotranspiration) improve ecological niche models by capturing habitat dynamics and disturbances such as fires, droughts or pests. It highlights cloud platforms such as Google Earth Engine as the way to compute them at large scale.

With MonitorEO: EVI (Modis), LST (Modis), ALBEDO (Modis), ET (Modis); download of variables (inputs of a model); period 2001-01-01 / 2020-12-31; region the study area of the species (a protected area, an administrative unit or a drawn polygon). Ask the assistant: "Download on aligned grids, for a species distribution model, the mean, percentile 5 and percentile 95 of MODIS EVI, the mean LST, albedo and ET from 2001 to 2020 in a drawn polygon"

download_variables produces the predictor grids; the niche model is fitted outside MonitorEO.

How does a wildfire affect the productivity, vegetation water, albedo and sensible heat of ecosystems over the following two years?

Bruno Marcos, João Gonçalves, Domingo Alcaraz-Segura et al. (2021). A framework for multi-dimensional assessment of wildfire disturbance severity from remotely sensed ecosystem functioning attributes. Remote Sensing. https://doi.org/10.3390/rs13040780

The authors propose a framework to measure wildfire severity on four dimensions of ecosystem functioning with interannual anomalies of satellite-derived functional attributes (MODIS, 2000-2018) in north-western Iberia. Random forests separated burned from unburned areas very well, and the effects were strongest on productivity and sensible heat, with both abrupt and lagged responses.

With MonitorEO: EVI (Modis), NDWI (Modis), ALBEDO (Modis), LST (Modis); comparison of periods, map of the comparison, burn severity; period 2000-02-24 / 2018-12-31; region a drawn polygon over a burned area. Ask the assistant: "Compare the MODIS EVI of the year after the fire with the mean of the three years before in a drawn polygon over a burned area in northern Portugal"

Repeat with NDWI, ALBEDO and LST (sensible heat approached with LST); burn_severity adds the dNBR of Landsat or Sentinel-2. The random forest is outside MonitorEO.

Can satellite-derived ecosystem functional types help set geographic conservation priorities?

B. P. Cazorla, J. Cabello, J. Peñas et al. (2021). Incorporating Ecosystem Functional Diversity into Geographic Conservation Priorities Using Remotely Sensed Ecosystem Functional Types. Ecosystems. https://doi.org/10.1007/s10021-020-00533-4

In the Baja California peninsula (Mexico), EFTs based on the dynamics of EVI-derived primary production were used to map functional heterogeneity and to rank areas by EFT richness and rarity. These functional priorities reinforced and complemented earlier priorities based on species composition and structure.

With MonitorEO: EVI (Modis); map, statistics of a map; period 2001-01-01 / 2014-12-31; region {"find_admin_units": "Baja California", "level": 1, "country": "MEX"}. Ask the assistant: "Map of the annual mean of MODIS EVI from 2001 to 2014 in Baja California and Baja California Sur (administrative units of level 1)"

MonitorEO gives the productivity attribute (annual mean EVI) and its maximum and minimum; seasonality, phenology, EFTs and priorities are computed outside.

Do biogeographic regionalisations based on species and structure match the patterns of ecosystem functioning seen from satellites?

Beatriz P. Cazorla, Pedro P. Garcillán, Javier Cabello et al. (2021). Patterns of ecosystem functioning as a tool for biological regionalization: the case of the Mediterranean-desert-tropical transition of Baja California. Mediterranean Botany. https://doi.org/10.5209/mbot.68529

Using the Mediterranean-desert-tropical transition of the Baja California peninsula as a case, the study compared existing ecological regionalisations with satellite-derived Ecosystem Functional Types. Functional patterns complemented and refined the regions based on composition and structure, offering a new view of the geography of life.

With MonitorEO: EVI (Modis); statistics of a map, map; period 2001-01-01 / 2014-12-31; region {"ecoregion_at": [-115.5, 30.5]}. Ask the assistant: "Statistics of the annual mean of MODIS EVI from 2001 to 2014 in the ecoregion that contains the point [-115.5, 30.5] (Baja California)"

Repeat for each RESOLVE ecoregion of the peninsula to compare them; EFTs are computed outside.

Which remote-sensing methods can characterise the biophysical features of all protected areas worldwide, and what are their challenges?

Javier Martínez-López, Bastian Bertzky, Simon Willcock et al. (2021). Remote sensing methods for the biophysical characterization of protected areas globally: Challenges and opportunities. ISPRS International Journal of Geo-Information. https://doi.org/10.3390/ijgi10060384

This review covers remote-sensing methods to characterise the habitats and ecosystems of terrestrial protected areas at global scale, stressing the need to combine structural and functional attributes and a broad set of variables. It discusses examples from local to global scale and gives recommendations for systematic, long-term monitoring.

With MonitorEO: NDVI (Modis), LST (Modis), ET (Modis), ELEVATION (Copernicus); statistics of a map, trends, download of variables (inputs of a model); period 2001-01-01 / 2020-12-31; region {"find_protected_areas": "<name of the protected area>"}. Ask the assistant: "Annual trend of MODIS NDVI between 2001 and 2020 in the protected area [name]"

MonitorEO covers the WDPA protected areas with functional variables (indices, LST, ET) and terrain.

How to obtain reliable functional attributes from Landsat and Sentinel-2 vegetation index series full of gaps caused by clouds and fires?

Saverio Vicario, Maria Adamo, Domingo Alcaraz-Segura et al. (2020). Bayesian harmonic modelling of sparse and irregular satellite remote sensing time series of vegetation indexes: A story of clouds and fires. Remote Sensing. https://doi.org/10.3390/rs12010083

A Bayesian harmonic model fitted year by year, with hierarchical priors that share information across years, was proposed to handle sparse and irregular Landsat and Sentinel-2 vegetation index series. It yields the yearly mean, standard deviation and date of peak with their uncertainty, and was tested with real cloud patterns of Peneda-Gerês National Park (Portugal) and its sensitivity to abrupt changes such as fires.

With MonitorEO: NDVI (Landsat), NDVI (Sentinel-2), CF (Sentinel-2); time series, map; period 2017-04-01 / 2019-12-31; region {"find_protected_areas": "Peneda-Gerês"}. Ask the assistant: "Monthly series of Sentinel-2 NDVI from April 2017 to 2019 in Peneda-Gerês National Park, with the number of images and valid pixels"

The series of MonitorEO report the valid pixels of each step and CF the cloud frequency; the Bayesian harmonic fit is outside.

Where are the data of ecosystem functional attributes and types of Sierra Nevada derived from MODIS EVI?

Cazorla, Beatriz P, Cabello, J, Reyes, A et al. (2020). Ecosystem functioning and functional diversity of Sierra Nevada (SE Spain), Version 2. PANGAEA. https://doi.org/10.1594/pangaea.924792

This PANGAEA dataset (version 2) provides the ecosystem functional attributes of the Sierra Nevada Biosphere Reserve (annual primary production, seasonality and phenology of MODIS MOD13Q1 EVI, 2001-2018) and the Ecosystem Functional Types and functional diversity derived from them. It is the data behind the ESSD data paper of the same team.

With MonitorEO: EVI (Modis); map, GeoTIFF; period 2001-01-01 / 2018-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Download as GeoTIFF the annual mean of MODIS EVI (mean between years) from 2001 to 2018 in Sierra Nevada National Park"

Use it to compare MonitorEO maps with the published attributes; seasonality, phenology and EFTs come from the dataset.

How can the functional dimension of biodiversity, seen from satellites, be incorporated into ecology and conservation?

B Pérez Cazorla (2020). Incorporating the functional dimension of biodiversity through remote sensing into ecology and conservation. Universidad de Almería (PhD thesis). https://repositorio.ual.es/handle/10835/10837

This doctoral thesis uses attributes of the seasonal curves of spectral vegetation indices (Ecosystem Functional Attributes, with the EVI as a proxy of primary production) as surrogates of ecosystem functions. It derives Ecosystem Functional Types and shows that they reinforce and complement conservation priorities based on composition and structure.

With MonitorEO: EVI (Modis); map, map of the trend; period 2001-01-01 / 2018-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Map of the annual mean of MODIS EVI from 2001 to 2018 in Sierra Nevada National Park"

MonitorEO gives the productivity attribute; seasonality, phenology and EFTs are computed outside.

How to characterise the heterogeneity of social-ecological systems by combining census data and remote sensing?

M. Vallejos, S. Aguiar, G. Baldi et al. (2020). Social-Ecological Functional Types: Connecting People and Ecosystems in the Argentine Chaco. Ecosystems. https://doi.org/10.1007/s10021-019-00415-4

The authors proposed social-ecological functional types, built by clustering census tracts of the Argentine Chaco with 17 human, ecological and interaction variables from census and remote sensing. Ecological variables, such as mean annual productivity, defined the main classes and human ones the subclasses, giving a tool for land-use planning in a deforestation hotspot.

With MonitorEO: EVI (Modis); statistics of a map, map; period 2001-01-01 / 2016-12-31; region {"find_admin_units": "<departments of the Argentine Chaco>", "level": 2, "country": "ARG"}. Ask the assistant: "Annual mean of MODIS EVI from 2001 to 2016 in each department of the Chaco province (administrative units of level 2)"

MonitorEO gives the productivity of each unit; census variables and the clustering are outside.

Which predictors (climate, land use or ecosystem functioning) make bird niche models more transferable in time?

Adrián Regos, Laura Gagne, Domingo Alcaraz-Segura et al. (2019). Effects of species traits and environmental predictors on performance and transferability of ecological niche models. Scientific Reports. https://doi.org/10.1038/s41598-019-40766-5

Niche models of 27 bird species were fitted with climate, land-use/land-cover and satellite-derived ecosystem functional attributes, alone and combined, and tested within and beyond the calibration period. Combining the three improved performance and transferability, but transferability stayed low for most species, so projections under global change need caution.

With MonitorEO: EVI (Modis), LST (Modis), PRECIP (TerraClimate), TMAX (TerraClimate); download of variables (inputs of a model); period 2001-01-01 / 2010-12-31; region the study area (an administrative unit or a drawn polygon). Ask the assistant: "Download on aligned grids the mean and percentiles 5 and 95 of MODIS EVI, the mean LST and the TerraClimate annual precipitation from 2001 to 2010 in a drawn polygon"

Repeat for a second period to test transferability; the niche models are outside MonitorEO.

Which satellite indicators best detect wildfire disturbances in space and time?

Bruno Marcos, João Gonçalves, Domingo Alcaraz-Segura et al. (2019). Improving the detection of wildfire disturbances in space and time based on indicators extracted from MODIS data: a case study in northern Portugal. International Journal of Applied Earth Observation and Geoinformation. https://doi.org/10.1016/j.jag.2018.12.003

A generic framework compared, ranked and combined several MODIS-derived indicators to discriminate burned areas, applying a simple change-point analysis to their time series in northern Portugal without external fire data. A consensus of indicators from different spectral regions mapped wildfire disturbances best and can complete incomplete fire databases.

With MonitorEO: NDVI (Modis), EVI (Modis), LST (Modis), ALBEDO (Modis); time series, map of the comparison; period 2001-01-01 / 2016-12-31; region a drawn polygon over a burned area in northern Portugal. Ask the assistant: "Monthly series of MODIS NDVI, LST and albedo from 2001 to 2016 in a drawn polygon over a burned area in northern Portugal"

MonitorEO gives the indicator series and before/after comparisons; the change-point detection is outside.

Do satellite-derived ecosystem functioning attributes predict the local abundance of a rare species?

Salvador Arenas-Castro, Adrián Regos, João F. Gonçalves et al. (2019). Remotely sensed variables of ecosystem functioning support robust predictions of abundance patterns for rare species. Remote Sensing. https://doi.org/10.3390/rs11182086

Species abundance models of the narrow Iberian endemic Gerês lily (Iris boissieri) were calibrated with climate, topography, land cover and satellite ecosystem functioning attributes. Models with functioning attributes alone explained the most variance (slightly more with topography), and predicted interannual abundance tracked climate fluctuations.

With MonitorEO: EVI (Modis), LST (Modis), ELEVATION (Copernicus), SLOPE (Copernicus); download of variables (inputs of a model); period 2001-01-01 / 2017-12-31; region {"find_protected_areas": "Peneda-Gerês"}. Ask the assistant: "Download on aligned grids the mean and percentiles 5 and 95 of MODIS EVI, the LST, the elevation and the slope from 2001 to 2017 in Peneda-Gerês National Park"

For interannual predictions use temporal "annual" (one grid per year); the abundance models are outside MonitorEO.

Are satellite ecosystem functional attributes good predictors of habitat suitability across scales?

Salvador Arenas-Castro, João Gonçalves, Paulo Alves et al. (2018). Assessing the multi-scale predictive ability of ecosystem functional attributes for species distribution modelling. PLOS ONE. https://doi.org/10.1371/journal.pone.0199292

Species distribution models of two plants of conservation concern (Iris boissieri and Taxus baccata) were fitted at several scales with climate, landscape, satellite ecosystem functional attributes, or combinations. Models with functional attributes performed as well as climate-based ones, found additional suitable areas and depended little on scale, supporting them as essential biodiversity variables.

With MonitorEO: EVI (Modis), LST (Modis), ALBEDO (Modis); download of variables (inputs of a model); period 2001-01-01 / 2015-12-31; region {"find_protected_areas": "Peneda-Gerês"}. Ask the assistant: "Download on aligned grids the annual mean, minimum and maximum of MODIS EVI from 2001 to 2015 in Peneda-Gerês National Park and its ecoregions"

The study area and ecoregion grids of download_variables give the two scales; the SDMs are outside.

How much of the ecosystem functional diversity of the conterminous US do the AmeriFlux and NEON flux networks represent?

Samuel Villarreal, Mario Guevara, Domingo Alcaraz-Segura et al. (2018). Ecosystem functional diversity and the representativeness of environmental networks across the conterminous United States. Agricultural and Forest Meteorology. https://doi.org/10.1016/j.agrformet.2018.07.016

Ecosystem functional types from MODIS carbon-uptake dynamics (2001-2014) were used to assess how representative the eddy-covariance sites of AmeriFlux and NEON are across the conterminous United States. AmeriFlux covered about half of the EFT categories and of the functional heterogeneity, NEON about a quarter, and desert, western mountain and plains domains were the least represented.

With MonitorEO: EVI (Modis); map, statistics of a map; period 2001-01-01 / 2014-12-31; region {"bbox": [-125, 24, -66, 50]}. Ask the assistant: "Map of the annual mean of MODIS EVI from 2001 to 2014 in the conterminous United States (bbox [-125, 24, -66, 50])"

Big region: MODIS and annual step. EFTs and the representativeness analysis are outside MonitorEO.

How can remote sensing of ecosystem functions help report the conservation status of habitats under the EU Habitats Directive?

J. Cabello, P. Mairota, D. Alcaraz-Segura et al. (2018). SATELLITE REMOTE SENSING OF ECOSYSTEM FUNCTIONS: OPPORTUNITIES AND CHALLENGES FOR REPORTING OBLIGATIONS OF THE EU HABITAT DIRECTIVE. IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium. https://doi.org/10.1109/igarss.2018.8517296

The paper discusses the prospects and issues of using satellite remote sensing of ecosystem functions in the six-yearly reporting of the Habitats Directive (Articles 11 and 17), which today relies on composition and structure. Monitoring functions is presented as a way to bring remote sensing into real-world biodiversity monitoring of Natura 2000.

With MonitorEO: EVI (Modis), LST (Modis), ET (Modis); trends, comparison of periods; period 2013-01-01 / 2018-12-31; region {"find_protected_areas": "<Natura 2000 site>"}. Ask the assistant: "Compare the MODIS EVI of the 2013-2018 reporting period with that of 2007-2012 in a Natura 2000 site"

Natura 2000 sites are in WDPA (find_protected_areas); MonitorEO computes the functional indicators, not the conservation status.

How did the seasonal greenness dynamics of protected areas in southern Spain and northern Morocco change between 2001 and 2012, and what controls those trends?

Patrícia Lourenço, Domingo Alcaraz-Segura, Andrés Reyes-Díez et al. (2018). Trends in vegetation greenness dynamics in protected areas across borders: what are the environmental controls?. International Journal of Remote Sensing. https://doi.org/10.1080/01431161.2018.1466080

Interannual trends of the seasonal dynamics of MODIS Terra EVI (January 2001 to December 2012) were computed for the protected areas of southern Spain and northern Morocco and grouped into four patterns. The trends appear mostly climate driven in both countries, although socio-ecological processes may also explain those of Moroccan protected areas and Spanish wetlands.

With MonitorEO: EVI (Modis); trends, map of the trend; period 2001-01-01 / 2012-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Monthly trend of MODIS EVI between 2001 and 2012 in Sierra Nevada National Park (repeat for each protected area)"

trend_table with step "monthly" gives the trend of each month of the seasonal curve (Mann-Kendall and Sen); a list of WDPA IDs computes several areas at once. Grouping into patterns is outside.

How is soil organic carbon distributed across Andalusia, and which environmental factors control it?

Daphne Armas, Mario Guevara, Domingo Alcaraz-Segura et al. (2017). Mapa digital del perfil del carbono orgánico en los suelos de Andalucía, España. Ecosistemas. https://doi.org/10.7818/ecos.2017.26-3.10

Maps of soil organic carbon percentage were produced for Andalusia from 1,500 legacy soil profiles and 20 environmental covariates of climate, topography and satellite functional attributes of primary production, with quantified uncertainty. The work identifies the controls of the spatial variability of this carbon pool.

With MonitorEO: EVI (Modis), ELEVATION (Copernicus), SLOPE (Copernicus), BIO01 (WorldClim), BIO12 (WorldClim); download of variables (inputs of a model); period 2001-01-01 / 2015-12-31; region {"find_admin_units": "Andalucía", "level": 1, "country": "ESP"}. Ask the assistant: "Download on aligned grids the annual mean of MODIS EVI from 2001 to 2015, the elevation, the slope, BIO01 and BIO12 of Andalusia at 250 m"

Covariates for digital soil mapping; the soil profiles and the model are outside MonitorEO.

Can satellite-derived ecosystem functional attributes anticipate species range shifts?

Domingo Alcaraz-Segura, Angela Lomba, Rita Sousa-Silva et al. (2017). Potential of satellite-derived ecosystem functional attributes to anticipate species range shifts. International Journal of Applied Earth Observation and Geoinformation. https://doi.org/10.1016/j.jag.2016.12.009

The study used satellite-derived ecosystem functional attributes as predictors in species distribution models of several species. They worked as meaningful essential biodiversity variables, giving early warnings of range shifts and predicting short-term fluctuations in suitable conditions.

With MonitorEO: EVI (Modis), NDVI (Modis); download of variables (inputs of a model); period 2001-01-01 / 2015-12-31; region the study area of the species. Ask the assistant: "Download on aligned grids, one per year, the mean, minimum and maximum of MODIS EVI from 2001 to 2015 in the study area"

Annual grids (temporal "annual") allow following yearly fluctuations of suitability; the SDMs are outside.

How will habitat quality for the European badger in arid south-eastern Spain change under climate change when satellite ecosystem functioning is included?

Juan M. Requena-Mullor, Enrique López, Antonio J. Castro et al. (2017). Remote-sensing based approach to forecast habitat quality under climate change scenarios. PLOS ONE. https://doi.org/10.1371/journal.pone.0172107

Global climate-based models of the badger range were combined with regional models built with EVI-derived variables of ecosystem functioning in an arid region of south-eastern Spain, and habitat was forecast for 2071-2099. Including these variables made the forecast less favourable for the badger and changed the spatial pattern of suitability more than climate-only models.

With MonitorEO: EVI (Modis), PRECIP (TerraClimate), TMAX (TerraClimate); download of variables (inputs of a model); period 2001-01-01 / 2014-12-31; region {"find_admin_units": "Almería", "level": 2, "country": "ESP"}. Ask the assistant: "Download on aligned grids the annual mean, minimum and maximum of MODIS EVI from 2001 to 2014 and the TerraClimate annual precipitation of the province of Almería"

Predictors for the distribution models; future scenarios are outside MonitorEO.

How can every protected area in the world be systematically divided into habitat types from satellite data?

Javier Martínez-López, Bastian Bertzky, Francisco Bonet-García et al. (2016). Biophysical characterization of protected areas globally through optimized image segmentation and classification. Remote Sensing. https://doi.org/10.3390/rs8090780

The paper introduces eHabitat+, a habitat modelling service of the EU Digital Observatory for Protected Areas, and its component that segments protected areas into habitat functional types with an optimised automatic segmentation of several environmental variables. It allows producing comparable indicators for protected areas worldwide, as illustrated with case studies.

With MonitorEO: NDVI (Modis), LSWI (Landsat), SLOPE (Copernicus), ELEVATION (Copernicus); download of variables (inputs of a model); period 2015-01-01 / 2020-12-31; region {"find_protected_areas": "<name of the protected area>"}. Ask the assistant: "Download for eHabitat+ the NDVI (percentiles 5 and 95), the LSWI, the elevation and the slope of a protected area and its ecoregions, on aligned grids"

download_variables was designed for these eHabitat+ inputs; the segmentation is done in eHabitat+.

How can satellites monitor the ecosystem functioning of Spain's National Parks and detect its trends and anomalies?

J. Cabello, D. Alcaraz-Segura, A. Reyes et al. (2016). Sistema para el Seguimiento del funcionamiento de ecosistemas en la Red de Parques Nacionales de España mediante Teledetección. Revista de Teledetección. https://doi.org/10.4995/raet.2016.5731

REMOTE, a monitoring system developed with the Spanish National Parks agency (OAPN), uses MODIS Terra EVI and NDVI time series to describe each park with three attributes of carbon gains: annual production, seasonality and phenology. It sets reference conditions, detects trends and anomalies and runs on open-source software to inform managers about the health of the parks.

With MonitorEO: EVI (Modis), NDVI (Modis); time series, trends, comparison of periods; period 2001-01-01 / 2024-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Compare the monthly MODIS EVI of 2024 with the 2001-2015 reference in Sierra Nevada National Park"

compare_series gives anomalies against a reference period and trend_table the trends; for every national park pass the list of their WDPA IDs (find_protected_areas).

What is the ecosystem functional diversity of the Mediterranean-desert-tropical transition of Baja California, and how can it be used for conservation?

B Pérez Cazorla (2015). Ecología y conservación de la diversidad funcional de ecosistemas en la transición mediterráneo-desierto-tropical de la Península de Baja California. Universidad de Granada (thesis). https://doi.org/10.30827/digibug.38511

This thesis work used MODIS time series to compute ecosystem functional types, the means of their functional attributes, functional diversity, interannual variability and similarity across the Baja California peninsula. It lays the ground for the later articles on functional regionalisation and conservation priorities in the peninsula.

With MonitorEO: EVI (Modis); map, statistics of a map; period 2001-01-01 / 2013-12-31; region {"find_admin_units": "Baja California", "level": 1, "country": "MEX"}. Ask the assistant: "Map of the annual mean of MODIS EVI from 2001 to 2013 in Baja California and Baja California Sur"

EFT classification and diversity are computed outside MonitorEO.

How does quality filtering of MODIS EVI affect the monitoring of ecosystem functioning?

A. Reyes Díez, D. Alcaraz-Segura, J. Cabello Piñar (2015). Implicaciones del filtrado de calidad del índice de vegetación EVI para el seguimiento funcional de ecosistemas. Revista de Teledetección. https://doi.org/10.4995/raet.2015.3316

The study examined the implications of using the quality-assessment layer of the MODIS MOD13Q1 vegetation index product to filter EVI values when characterising ecosystem functioning. It shows that the choice of filter changes the derived functional attributes and gives recommendations for monitoring.

With MonitorEO: EVI (Modis); time series, map; period 2001-01-01 / 2014-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Monthly series of MODIS EVI from 2001 to 2014 in Sierra Nevada National Park, with and without the snow filter"

MonitorEO applies pixel filters (clouds, water, snow, aerosols) that can be switched with quality; each series reports its valid pixels.

How can satellites detect which ecosystems depend on groundwater?

Contreras, S., Alcaraz-Segura, D., Scanlon, B. et al. (2013). Detecting ecosystem reliance on groundwater based on satellite-derived greenness anomalies and temporal dynamics. Earth Observation of Ecosystem Services. https://doi.org/10.1201/b15628-18

The chapter detects groundwater-dependent ecosystems from MODIS EVI: areas that stay greener than expected for their rainfall (a regional EVI-precipitation function) and that keep their greenness through dry periods. It illustrates the approach along a gradient of groundwater dependence in the Telteca area of the Monte desert (Argentina).

With MonitorEO: EVI (Modis), PRECIP (CHIRPS); map, time series; period 2001-01-01 / 2012-12-31; region a drawn polygon. Ask the assistant: "Map of the annual mean of MODIS EVI and the CHIRPS annual precipitation from 2001 to 2012 in a drawn polygon of the Monte desert"

The EVI-precipitation residuals are computed outside MonitorEO.

How can satellites assess the functional diversity and carbon gains of a national park network to guide conservation strategies?

Cabello, J., Lourenço, P., Reyes, A. et al. (2013). Ecosystem services assessment of national parks networks for functional diversity and carbon conservation strategies using remote sensing. Earth Observation of Ecosystem Services. https://doi.org/10.1201/b15628-14

The chapter uses satellite vegetation indices to assess ecosystem services of a national parks network, focusing on the functional diversity of ecosystems and on carbon gains as a basis for conservation strategies. It links biodiversity conservation with the supply of services to human welfare.

With MonitorEO: EVI (Modis); statistics of a map, map; period 2001-01-01 / 2012-12-31; region the Spanish National Parks: a list of their WDPA IDs (find_protected_areas for each; Sierra Nevada is 389011). Ask the assistant: "Statistics of the annual mean of MODIS EVI from 2001 to 2012 in each National Park of Spain"

Pass the list of WDPA IDs; EFT diversity is computed outside.

Which environmental and human factors control ecosystem functional diversity in temperate South America?

Domingo Alcaraz-Segura, José Paruelo, Howard Epstein et al. (2013). Environmental and human controls of ecosystem functional diversity in temperate South America. Remote Sensing. https://doi.org/10.3390/rs5010127

EFTs were derived from three attributes of the seasonal MODIS EVI curve (annual mean, seasonal coefficient of variation and date of maximum) across temperate South America, and EFT richness was related to environmental and human variables. Water availability, not energy, was the main climatic driver in natural areas, while human influence first increased and then reduced richness.

With MonitorEO: EVI (Modis), PRECIP (TerraClimate), PET (TerraClimate); map, download of variables (inputs of a model); period 2001-01-01 / 2010-12-31; region {"country": "URY"}. Ask the assistant: "Map of the annual mean of MODIS EVI from 2001 to 2010 in Uruguay"

MonitorEO gives the productivity attribute and the water/energy drivers; the CV, date of maximum and EFTs are computed outside.

How do productivity and its seasonality change when forest is cleared for agriculture and ranching?

J.N. Volante, D. Alcaraz-Segura, M.J. Mosciaro et al. (2012). Ecosystem functional changes associated with land clearing in NW Argentina. Agriculture, Ecosystems & Environment. https://doi.org/10.1016/j.agee.2011.08.012

Land clearing in subtropical NW Argentina (5.9% of the region in 2000-2007) was mapped with Landsat, and its effect on two functional attributes was measured with MODIS EVI: the annual mean (a proxy of aboveground productivity) and its seasonal coefficient of variation. Clearing changed total productivity little but made cleared parcels much more seasonal and more variable between years than the natural vegetation.

With MonitorEO: EVI (Modis); map of the comparison, comparison of periods, map; period 2006-01-01 / 2007-12-31; region {"find_admin_units": "Salta", "level": 1, "country": "ARG"}. Ask the assistant: "Compare the MODIS EVI of 2006-2007 with that of 2000-2001 in the province of Salta (Argentina)"

Annual mean with temporal "annual"; the seasonal CV is not a MonitorEO metric (approach it with the annual minimum and maximum). Deforestation mapping is outside.

How can remote sensing bring the ecosystem functioning dimension into conservation?

Javier Cabello, Néstor Fernández, Domingo Alcaraz-Segura et al. (2012). The ecosystem functioning dimension in conservation: Insights from remote sensing. Biodiversity and Conservation. https://doi.org/10.1007/s10531-012-0370-7

This review identifies the fields where a satellite characterisation of ecosystem functioning (carbon gains, energy and water exchanges) can help conservation science and practice. It illustrates practical ways to use satellite-derived information for that purpose.

With MonitorEO: EVI (Modis), LST (Modis), ET (Modis); trends, comparison of periods; period 2001-01-01 / 2024-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Annual trend of MODIS EVI, LST and ET between 2001 and 2024 in Sierra Nevada National Park"

Is the North American boreal forest greening or browning, and does the answer depend on the satellite dataset?

DOMINGO ALCARAZ‐SEGURA, EMILIO CHUVIECO, HOWARD E. EPSTEIN et al. (2010). Debating the greening vs. browning of the North American boreal forest: Differences between satellite datasets. Global Change Biology. https://doi.org/10.1111/j.1365-2486.2009.01956.x

NDVI trends of recently burned and unburned boreal forest in central Canada were computed with the seasonal Mann-Kendall test from the 8 km GIMMS dataset and a 1 km dataset of the Canada Centre for Remote Sensing. The 1 km data showed recovery after fire and greening, while GIMMS missed most recoveries and showed widespread declines, a bias with broad implications for carbon assessments.

With MonitorEO: NDVI (Landsat), NDVI (Modis); trends, map of the trend; period 1985-01-01 / 2024-12-31; region a drawn polygon over the burned or unburned forest. Ask the assistant: "Monthly trend of Landsat NDVI between 1985 and 2024 in a drawn polygon over a burned boreal forest in Canada"

trend_table with step "monthly" gives the trend of each month (as the seasonal Mann-Kendall); comparing Landsat with MODIS shows the sensitivity to the dataset.

Do NDVI trends in the Iberian Peninsula agree across different time series derived from the same AVHRR sensor?

Domingo Alcaraz-Segura, Elisa Liras, Siham Tabik et al. (2010). Evaluating the consistency of the 1982-1999 NDVI trends in the Iberian Peninsula across four time-series derived from the AVHRR sensor: LTDR, GIMMS, FASIR, and PAL-II. Sensors. https://doi.org/10.3390/s100201291

Trends of NDVI between 1982 and 1999 were computed with the seasonal Mann-Kendall test from four AVHRR datasets (LTDR, GIMMS, FASIR, PAL-II) and their spatial agreement was tested. Positive trends dominated the north, east and central mountains and were absent in the south-west, but agreement across all four datasets covered only about half of Iberia, so carbon-gain trends should be checked with several datasets.

With MonitorEO: NDVI (Landsat), NDVI (Modis); map of the trend, trends; period 1985-01-01 / 1999-12-31; region {"bbox": [-9.6, 35.9, 3.4, 43.9]}. Ask the assistant: "Map of the annual trend of Landsat NDVI between 1985 and 1999 in the Iberian Peninsula (bbox [-9.6, 35.9, 3.4, 43.9]) at a coarse resolution"

AVHRR datasets are not in MonitorEO; Landsat covers the period since 1984 (heavy for all Iberia: export to Google Drive) and MODIS since 2000.

What is the reference functional signature of the major Iberian vegetation types according to NDVI dynamics, and what controls it?

Domingo Alcaraz-Segura, Javier Cabello, José Paruelo (2009). Baseline characterization of major Iberian vegetation types based on the NDVI dynamics. Plant Ecology. https://doi.org/10.1007/s11258-008-9555-2

Baseline NDVI dynamics (1982-1999) were described for the largest homogeneous patches of natural vegetation of the Iberian Peninsula using attributes such as annual mean, relative range, maximum, minimum and their months. Eurosiberian types were more productive with summer peaks, Mediterranean ones had summer minima; precipitation drove productivity in Mediterranean areas and temperature in Eurosiberian ones.

With MonitorEO: NDVI (Modis), PRECIP (ERA5-Land), TMEAN (ERA5-Land); time series, map; period 2001-01-01 / 2020-12-31; region a drawn polygon over a homogeneous vegetation patch. Ask the assistant: "Monthly series of MODIS NDVI from 2001 to 2020 in a drawn polygon over a homogeneous holm-oak forest, with ERA5-Land precipitation and temperature"

The monthly series gives the seasonal curve (maximum, minimum and their months); AVHRR is not in MonitorEO.

How has ecosystem functioning of the Spanish National Parks changed, and how does it respond to environmental changes?

Domingo Alcaraz-Segura, Javier Cabello, José M. Paruelo et al. (2009). Use of descriptors of ecosystem functioning for monitoring a national park network: A remote sensing approach. Environmental Management. https://doi.org/10.1007/s00267-008-9154-y

The spatial and temporal patterns of ecosystem functioning of the Spanish National Parks between 1982 and 2006 were described with GIMMS AVHRR NDVI, a linear estimator of the radiation intercepted by vegetation. The study proposes these descriptors to monitor a national park network.

With MonitorEO: NDVI (Modis), NDVI (Landsat); trends, time series; period 2001-01-01 / 2024-12-31; region the Spanish National Parks: a list of their WDPA IDs (find_protected_areas for each; Sierra Nevada is 389011). Ask the assistant: "Monthly trend of MODIS NDVI between 2001 and 2024 in each National Park of Spain"

Same analysis with MODIS (since 2000) or Landsat (since 1984) instead of AVHRR.

How has vegetation greenness changed in Spain's National Parks, and are they equally affected by the European greening?

Domingo Alcaraz‐Segura, Javier Cabello, José M. Paruelo et al. (2008). Trends in the surface vegetation dynamics of the national parks of Spain as observed by satellite sensors. Applied Vegetation Science. https://doi.org/10.3170/2008-7-18522

GIMMS AVHRR NDVI was used to give a baseline of the vegetation dynamics of the Iberian National Parks of Spain and to analyse their trends between 1981 and 2003. Trends differed between parks and vegetation types, so the greening detected across Europe did not affect all protected areas in the same way.

With MonitorEO: NDVI (Modis), NDVI (Landsat); trends, map of the trend; period 2001-01-01 / 2024-12-31; region the Spanish National Parks: a list of their WDPA IDs (find_protected_areas for each; Sierra Nevada is 389011). Ask the assistant: "Monthly trend of MODIS NDVI between 2001 and 2024 in each National Park of mainland Spain"

With Landsat the series can start in 1985 (but not in 1981).

Which ecosystem functional types exist in the Iberian Peninsula according to the annual dynamics of NDVI?

Domingo Alcaraz, José Paruelo, Javier Cabello (2006). Identification of current ecosystem functional types in the Iberian Peninsula. Global Ecology and Biogeography. https://doi.org/10.1111/j.1466-822x.2006.00215.x

Ecosystem functional types of the Iberian Peninsula were defined from three AVHRR NDVI attributes: the annual integral (productivity), the relative annual range and the month of maximum (seasonality). Productivity decreased towards the south-east and was highest in the Eurosiberian region and the high Mediterranean mountains, giving a functional map complementary to vegetation maps.

With MonitorEO: NDVI (Modis); map, statistics of a map; period 2001-01-01 / 2024-12-31; region {"bbox": [-9.6, 35.9, 3.4, 43.9]}. Ask the assistant: "Map of the annual mean of MODIS NDVI from 2001 to 2024 in the Iberian Peninsula (bbox)"

MonitorEO gives the annual mean (productivity), maximum and minimum; the month of maximum and the EFT classification are computed outside.

How does the radiation absorbed by vegetation in an arid protected area vary in space and time, and how does it depend on rainfall?

José M. Paruelo, G. Piñeiro, P. Escribano et al. (2005). Temporal and spatial patterns of ecosystem functioning in protected arid areas in southeastern Spain. Applied Vegetation Science. https://doi.org/10.1111/j.1654-109x.2005.tb00633.x

The photosynthetically active radiation absorbed by the vegetation of Cabo de Gata-Níjar Natural Park was characterised with AVHRR NDVI for five landscape classes, and a 19-year NDVI series was related to precipitation inside and outside the park. Vegetation absorbed less than 20% of the incoming radiation, mountains more than hills and plains, with a single growing-season peak earlier in the plains.

With MonitorEO: NDVI (Modis), PRECIP (CHIRPS); time series, statistics of a map; period 2001-01-01 / 2024-12-31; region {"find_protected_areas": "Cabo de Gata-Níjar"}. Ask the assistant: "Monthly series of MODIS NDVI and CHIRPS precipitation from 2001 to 2024 in Cabo de Gata-Níjar Natural Park"

By landscape class: drawn polygons, or where with ELEVATION/SLOPE. APAR is not in MonitorEO (NDVI is its linear estimator).

Water

Can chlorophyll-a be estimated in small high-mountain lakes from satellite images, and which sensor works best?

J. Llodrà‐Llabrés, J. C. Pérez‐Girón, T. Postma et al. (2026). Satellite Solutions: Facing Chlorophyll-a Retrieval in Small Mountain Lakes in the Sierra Nevada, Spain. Water Resources Research. https://doi.org/10.1029/2026wr043523

Five glacial lakes of Sierra Nevada were sampled in 2020, 2021 and 2023, and about 100 chlorophyll-a measurements were matched with Sentinel-2, PlanetScope and WorldView-3 images. Multi-band models (green and near-infrared above all) did as well as models with spectral indices; Sentinel-2 and Planet beat the finer WorldView-3, and Planet stayed reliable close to the shore, where the adjacency effect is strongest.

With MonitorEO: Chl-a (Sentinel-2), NDCI (Sentinel-2); time series, map, statistics of a map; period 2020-06-01 / 2023-10-31; region a drawn polygon over each lake (small lakes: Sentinel-2, 10-20 m pixels). Ask the assistant: "Monthly series of Sentinel-2 chlorophyll-a (Chl-a) in Laguna de la Caldera (drawn polygon), June to October of 2020 to 2023, with the automatic water filter"

The Sentinel-2 Chl-a formula of the MonitorEO catalog comes from this article (reference llodrallabres2026). Pass quality {"water": "auto"} and a season of the ice-free months; Planet and WorldView-3 are not in MonitorEO; pixels next to the shore suffer the adjacency effect.

Which indices, models and atmospheric corrections work to estimate chlorophyll-a in inland waters with Sentinel-2?

Joana Llodrà-Llabrés, Javier Martínez-López, Thedmer Postma et al. (2023). Retrieving water chlorophyll-a concentration in inland waters from Sentinel-2 imagery: Review of operability, performance and ways forward. International Journal of Applied Earth Observation and Geoinformation. https://doi.org/10.1016/j.jag.2023.103605

This review synthesises the studies that retrieved chlorophyll-a, an indicator of trophic state, from Sentinel-2 in lakes, reservoirs and rivers, comparing indices, models, atmospheric corrections and field sampling. It assesses which choices matter most for accuracy and how operational the methods are, and proposes ways forward.

With MonitorEO: NDCI (Sentinel-2), Chl-a (Sentinel-2); time series, map; period 2018-01-01 / 2024-12-31; region a drawn polygon over a lake or reservoir. Ask the assistant: "Monthly series of Sentinel-2 NDCI between 2018 and 2024 in the Cubillas reservoir (drawn polygon), with the automatic water filter"

Pass quality {"water": "auto"}; MonitorEO uses the Sentinel-2 L2A surface reflectance, not the alternative atmospheric corrections compared in the review.

Fire

Can burned areas in a protected area be detected with Sentinel-2 spectral indices such as NDVI and RBR?

Ederson Steven Cobo-Muelas, Pablito Marcelo López-Serrano, Christian Wehenkel et al. (2026). Wildfire Detection in the Iztaccíhuatl-Popocatépetl Protected Natural Area Using Spectral Indices and Logistic Regression. Fire. https://doi.org/10.3390/fire9020050

Burned areas in the Iztaccíhuatl-Popocatépetl protected area (central Mexico) were detected with a logistic regression fed with Sentinel-2 variables: NDVI, a corrected Relativized Burn Ratio and the SWIR2 band. Agreement with the observed burned areas reached about 0.79 (kappa and overall accuracy), and the results depended on the probability threshold chosen.

With MonitorEO: RBR (Sentinel-2), NBR (Sentinel-2), NDVI (Sentinel-2); burn severity, map of the comparison; region {"find_protected_areas": "Iztaccíhuatl-Popocatépetl"}. Ask the assistant: "Burn severity with the Sentinel-2 RBR in Iztaccíhuatl-Popocatépetl National Park: period before [3 months before the fire] and after [2-3 months after the fire]"

burn_severity with index "RBR" (continuous, no classes) or the default dNBR with USGS classes; NDVI change with compare_map. The logistic regression and the SWIR2 band alone are outside MonitorEO.

Energy and temperature

Which areas of Granada combine the highest heat exposure and vulnerability, and how does the pattern of urban vegetation mitigate heat?

Carson Silveira, Nuria Pistón, Javier Martínez-López et al. (2025). Identifying priority heat-risk areas in Granada, Spain, using InVEST and landscape metrics. Urban Forestry & Urban Greening. https://doi.org/10.1016/j.ufug.2025.128794

Urban heat in Granada and its metropolitan area was modelled with the InVEST urban cooling model and crossed with heat exposure and social vulnerability indices to find priority areas. Boosted regression trees showed that green cover above about 35% with patches larger than about 200 m² gave the most cooling, so the authors recommend connecting small green spaces in dense neighbourhoods.

With MonitorEO: LST (Landsat), NDVI (Sentinel-2), ALBEDO (Sentinel-2); map, statistics of a map, GeoTIFF; period 2020-06-01 / 2024-08-31; region {"point": [-3.6, 37.18], "buffer": 8000}. Ask the assistant: "Map of the Landsat land surface temperature (LST) of the summers (June-August) of 2020 to 2024 in Granada (point with an 8 km radius)"

InVEST and the vulnerability indices are outside MonitorEO; it gives the observed surface temperature, vegetation and albedo (season ["06-01", "08-31"]) to compare with the model.

How to characterise and monitor with satellites the climate regulation service of ecosystems, including its biophysical effects?

Alcaraz-Segura, D., Berbery, E.H., Müller, O.V. et al. (2013). Characterizing and monitoring climate regulation services. Earth Observation of Ecosystem Services. https://doi.org/10.1201/b15628-21

The chapter argues that climate regulation by ecosystems includes biophysical effects (water and energy balance) besides greenhouse gases, and discusses the climate regulation value index, whose biophysical part depends on net radiation and latent heat flux. It shows how satellite-derived land surface information can characterise and monitor this service.

With MonitorEO: LE (Modis), ALBEDO (Modis), LST (Modis); time series, trends; period 2001-01-01 / 2020-12-31; region a drawn polygon or a protected area. Ask the assistant: "Monthly series of MODIS latent heat (LE), albedo and LST from 2001 to 2020 in a drawn polygon"

MonitorEO has latent heat (LE), albedo and LST but not net radiation.

Atmosphere

How has aerosol optical depth been distributed in space and time over Sierra Nevada in the last 20 years, and how does it relate to ecosystem types?

Ana del Águila, Domingo Alcaraz-Segura, Javier Martínez-López et al. (2024). Two decades of high-resolution aerosol product over the Sierra Nevada Mountain region (SE Spain): Spatio-temporal distribution and impact on ecosystems. Atmospheric Research. https://doi.org/10.1016/j.atmosres.2024.107515

The 1 km MODIS MAIAC aerosol optical depth (2001-2022) was validated against three AERONET stations at different altitudes and then used to map aerosols over the Sierra Nevada mountains. AOD was higher near valleys and towns and fell with altitude except over snow, and its monthly trends were analysed by ecosystem type and elevation belt in the protected area.

With MonitorEO: ARSL (Modis); time series, trends, map of the trend, map; period 2001-01-01 / 2022-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Monthly trend of MODIS aerosol optical depth (ARSL) in Sierra Nevada National Park between 2001 and 2022"

ARSL is the same MCD19A2 (MAIAC) product at 1 km; trend_table with step "monthly". By elevation belt: where with ELEVATION. The AERONET validation is outside MonitorEO.

Terrain

How will climate change affect an endemic rock plant of the Sierra Nevada high mountain, and does it differ between subpopulations at different altitudes?

Antonio J. Mendoza-Fernández, Ángel Fernández-Ceular, Domingo Alcaraz-Segura et al. (2022). The Fate of Endemic Species Specialized in Island Habitat under Climate Change in a Mediterranean High Mountain. Plants. https://doi.org/10.3390/plants11233193

For Moehringia fontqueri, a threatened endemic of north-facing siliceous rocks in eastern Sierra Nevada, the authors measured seed set and modelled current and future habitat suitability with Maxent using topographic, satellite and climate predictors. They assessed separately the subpopulations by altitude to guide conservation against climate change.

With MonitorEO: ELEVATION (Copernicus), SLOPE (Copernicus), BIO01 (WorldClim), BIO12 (WorldClim), SCF (Landsat), NDVI (Sentinel-2); download of variables (inputs of a model); period 2018-01-01 / 2022-12-31; region {"wdpa_id": 389011}. Ask the assistant: "Download on aligned grids the elevation, slope, WorldClim BIO01 and BIO12, the Landsat snow cover frequency and the Sentinel-2 NDVI of 2018-2022 in Sierra Nevada National Park above 2,000 m"

Predictors for the species distribution model; the model and the future climate scenarios are outside MonitorEO.