Xavier Rotllan-PuigASTER-Projects
Xavier Rotllan-Puig
Master of Science
Using ML for modelling the distribution of plants and their pollinators across Europe
About
23
Publications
7,774
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
199
Citations
Citations since 2017
Introduction
My research interests are in the area of ecology and biodiversity conservation, mainly focussed on the Mediterranean endemic and threatened flora. I have been collaborating in the study of reproductive processes of plants, as well as their relationships with other species and the environment. In this sense, I have expertise in field and laboratory works, and also in using GIS and ecological modelling techniques. Lately, my work is focused on ecological modelling and (spatial) data analysis.
Additional affiliations
June 2021 - present
Position
- Project officer
Description
- Flora Fauna Green Rural Corridor Connect: Localizing and modelling wildflower-pollinator interactions in rural landscapes with citizen science and deep learning
January 2017 - present
ASTER-Projects
Position
- Analyst
Description
- Ecology. Biodiversity. Conservation. Ecological modelling. Analysis of (spatial) data. - Main client: Joint Research Centre (European Commission)
June 2016 - December 2016
Piksel (Working for the Joint Research Centre - European Commission)
Position
- GIS Developer
Description
- Disaggregation of statistical data for the evaluation of the environmental impact of agriculture
Education
September 2009 - September 2010
September 1997 - January 2004
Publications
Publications (23)
Biodiversity loss, including the decrease of diversity of ecological interactions, is known to reduce the capacity of ecosystems to cope with the effects of global change. Here we assessed whether the distribution and survival of two declining relict plant species, Daphne rodriguezii and Cneorum tricoccon, were affected by the mutualism disruption...
Pollination is a valuable ecosystem service, and plant–pollinator interactions in particular are known to play a crucial role in conservation and ecosystem functioning. These mutualisms, like other ecological interactions, are currently threatened by different drivers of global change, mainly habitat loss, fragmentation, or modification of its qual...
Insect pollinators are a key component of biodiversity; they also play a major role in the reproduction of many species of wild plants and crops.
It is widely acknowledged that insect pollinators are threatened by many environmental pressures, mostly of anthropogenic nature. Their decline is a global phenomenon. A better understanding of their dist...
One of the crucial choices when modelling species distributions using pseudo-absences and background approaches is the delineation of the background area to fit the model. We hypothesise that there is a minimum background area around the geographical centre of the species distribution that characterises well enough the range of environmental condit...
The United Nations Sustainable Development Goal 15 (Life on Land), adopted the indicator 15.3.1 to measure the Land Degradation Neutrality. This indicator is based on three sub-indicators: (1) Trends in Land Cover, (2) Land Productivity and (3) Carbon Stocks. The Land Productivity sub-indicator refers to the total above-ground Net
Primary Productio...
We present a new application to recognize 218 species of cultivated crops on geo-tagged photos, ‘Pl@ntNet Crops’. The application and underlying algorithms are developed using more than 750k photos voluntarily collected by Pl@ntNet users. The app is then enriched by data and photos coming from the European Union’s (EU) Land Use and Coverage Area fr...
This report is a narrative review of the effects reported in the scientific literature (peer-reviewed and indexed scientific publications showing quantitative evidence) about the relationships between a selection of nature restoration practices and food production. The reported effects are extracted from systematic reviews, meta-analyses and indivi...
As part of the UN Sustainable Development Goal 15 (Life on Land), the indicator 15.3.1 is adopted to measure the Land Degradation Neutrality. Land Degradation Neutrality is addressed as stable —or increasing— state in the amount and quality of land resources required to support ecosystem functions and services and enhance food security during a cer...
LPDynR uses 'phenological' and productivity-related variables derived from time series of vegetation indexes, such as the Normalized Difference Vegetation Index, to assess ecosystem dynamics and change, which eventually might drive to land degradation. The final result of the Land Productivity Dynamics indicator is a categorical map with 5 classes...
A versatile tool that aims at (1) defining the minimum background extent necessary to fit Species Distribution Models reliable enough to extract ecologically relevant conclusions from them and (2) optimizing the modelling process in terms of computation demands
Biodiversity is suffering dramatic declines across the globe, threatening the ability of ecosystems to provide the services on which humanity depends. Mainstreaming biodiversity into the plans, strategies and policies of different economic sectors is key to reversing these declines. The importance of this mainstreaming is recognized by the Conventi...
One of the crucial choices when modelling species distributions using pseudo-absences approaches is the delineation of the background area to fit the model. We hypothesise that there is a minimum background area around the centre of the species distribution that characterizes well enough the range of environmental conditions needed by the species t...
Insect pollinators are a key component of biodiversity; they also play a major role in the reproduction of many species of wild plants and crops.
It is widely acknowledged that insect pollinators are threatened by many environmental pressures, mostly of anthropogenic nature. Their decline is a global phenomenon. A better understanding of their dist...
The aim of this package is to download species presences (occurrences) data from public repositories. For now it is implemented for Bioatles, which has data from the Balearic Islands (http://bioatles.caib.es), and for GBIF (https://www.gbif.org/).
This document shows the methodology used to assess the accuracy of the new Land Area Prediction Model (LAPM), as well as the results of the assessment.
LAPM aims at predicting land-use areas within each fine-scale Homogenous Spatial Units
(HSU). It has been evaluated its accuracy comparing its predictions with LPIS data aggregated to HSU level for...
Supplementary material of:
Declining relict plants: Climate effect or seed dispersal disruption? A landscape-scale approach. 2016. Basic and Applied Ecology 17(1):81-91
Narrow endemics constitute the cornerstone of Mediterranean plant diversity. Naufraga balearica (Apiaceae) is a critically endangered, extremely narrow endemic plant from the western Mediterranean island of Majorca. Because the species belongs to a monotypic genus, N. balearica was hypothesized to be a palaeoendemism. Here we conducted phylogenetic...
en las masas tratadas y se estudió la dispersión de CHV1 dentro y fuera de las parcelas tratadas. Los resultados muestran que la diversidad de GCV en Cataluña es baja, con tan solo 13 GCV descritos. Asimismo, el GCV EU-2 es en la mayoría de casos el dominante. La presencia de cepas hipovirulentas anteriores a las inoculaciones era baja (<4%) y en l...
Questions
Questions (2)
I've shared some R code to perform ISODATA clustering:
I would very much appreciate if it can be tested and some comments made.
A data set can also be found in the repository to ease the tests.
Thanks in advance!!
Hi,
In the multinom function manual (nnet R-package) it is advised to "roughly scale" the variables to [0,1]. How necessary is that? Which is the best way (method) to do it?
Thanks in advance.
xavi.
Projects
Project (1)
To support European agricultural and environmental policies, the Digital Rural Landscape Lab uses analytics and novel data capturing methods - combining insights from smartphones, farm sensors, street level cameras, crowdsourcing and satellites. Innovative integration of these data flows will improve farm management and refine rural landscape and biodiversity monitoring.
