Attività antropiche e pressioni di mercato stanno rapidamente riducendo la biodiversità. Per questa ragione, conservare il patrimonio ecosistemico, tassonomico e genetico risulta fondamentale al fine di garantire potenziale adattativo alle specie, e, in ultima analisi, un futuro sostenibile per il pianeta. Al fine di minimizzare la perdita di biodiversità, numerosi metodi sono stati proposti per priorizzare ecosistemi, specie e popolazioni. Il presente lavoro di tesi fornisce in primo luogo una revisione di tali approcci, proponendo un albero decisionale volto a favorirne un corretto utilizzo. Secondariamente, la variabilità genomica neutrale del bufalo d’acqua (Bubalus bubalis L.) è investigata per mezzo di un pannello di marcatori SNP a media densità, rivelando due centri di domesticazione (India Nord-occidentale, Cina-Indocina) e possibili rotte di migrazione per gli ecotipi ‘river’ e ‘swamp’. L’adattamento locale ad East Coast Fever, patologia endemica delle popolazioni bovine in Africa Sub-sahariana, è stato inoltre studiato in bovini autoctoni Ugandesi (Bos taurus L.) combinando tecniche di modellizzazione delle nicchie ecologiche e di genomica del paesaggio. L’approccio ha portato ad indentificare PRKG1 e SLA2 come possibili geni di adattamento. I risultati sono discussi alla luce delle possibili implicazioni nella conservazione del bufalo e nella gestione delle risorse genetiche animali Ugandesi. / Biodiversity is quickly disappearing due to human impact on the biosphere, and to market pressure. Consequently, the protection of both wild and domestic species needs to become a priority in order to preserve their evolutionary potential and, ultimately, guarantee a sustainable future for coming human generations. To date, tens of methods have been proposed to prioritize biodiversity for conservation purposes. Here, an ontology for priority setting in conservation biology is provided with the aim of supporting the selection of the most opportune methodologies given specific conservation goals. Further, two case studies are presented characterizing neutral and adaptive genomic diversity in water buffalo (Bubalus bubalis L.) and indigenous Ugandan cattle (Bos taurus L.), respectively. In particular, two independent domestication centres (North-western India and Indochina) and separate migration routes are suggested for the ‘river’ and ‘swamp’ water buffalo types. In the case of indigenous Ugandan cattle, the integration of species distribution modelling and landscape genomics techniques allowed the identification of PRKG1 and SLA2 as candidate genes for local adaptation to East Coast Fever, a vector-borne disease affecting bovine populations of Sub-Saharan Africa. Results are discussed for their implications in water buffalo conservation and Ugandan cattle adaptive management.

L’Afrique centrale atlantique (ACA) englobe l’ensemble du domaine bas-guinéen, les îles du Golfe de Guinée et une partie de l’archipel afromontagnard. Plusieurs centres d’endémisme ont été identifiés en son sein et sont généralement considérés comme liés à la présence de refuges forestiers durant les périodes glaciaires. Cependant, l’origine de cet endémisme, sa localisation et les méthodes permettant d’identifier ces centres restent controversées. La localisation de ces zones d’endémisme et des plantes rares qu’elles abritent, est pourtant un prérequis indispensable pour la mise en place de politiques cohérentes de conservation et demeure une priorité pour les organisations privées, institutionnelles ou gouvernementales actives dans la gestion et le maintien durable de la biodiversité.<p><p>Cette étude phytogéographique porte sur la famille des Orchidaceae et est basée sur l’analyse de la distribution des taxons endémiques de l’ACA. Elle s’appuie sur un jeu de données original résultant d’un effort d’échantillonnage important au Cameroun et d’un travail d’identification et de localisation de spécimens dans les principaux herbaria européens abritant des collections d’ACA. Durant cette étude, (i) nous avons tout d’abord identifié ces taxons endémiques et documenté leur distribution au travers de plusieurs contributions taxonomiques et floristiques, (ii) nous nous sommes ensuite intéressé aux nouvelles méthodes permettant d’analyser ces données d’herbier de plantes rares et donc pauvrement documentées, testant aussi l’intérêt des Orchidaceae comme marqueurs chorologiques, et finalement, appliquant ces méthodes à notre jeu de données, (iii) nous avons délimité des centres d’endémisme et identifié les territoires phytogéographiques des Orchidaceae en ACA.<p><p>(i) Une révision taxonomique des genres Chamaeangis Schltr. et Stolzia Schltr. a été réalisée respectivement. Sept nouveaux taxons ont été décrits: Angraecum atlanticum Stévart & Droissart, Chamaeangis spiralis Stévart & Droissart, Chamaeangis lecomtei (Finet) Schltr. var. tenuicalcar Stévart & Droissart, Polystachya engogensis Stévart & Droissart, Polystachya reticulata Stévart & Droissart, Stolzia repens (Rolfe) Summerh var. cleistogama Stévart, Droissart & Simo et Stolzia grandiflora P.J.Cribb subsp. lejolyana Stévart, Droissart & Simo. Plusieurs notes taxonomiques, phytogéographiques et écologiques supplémentaires ont également été redigées. Au total, nous avons identifié 203 taxons d’Orchidaceae endémiques d’ACA parmi lesquels 193 sont pris en compte pour l’étude des patrons d’endémisme.<p><p>(ii) Au Cameroun, les patrons de distribution des Orchidaceae et des Rubiaceae endémiques d’ACA ont été étudiés conjointement. Des méthodes de rééchantillonnage des données (raréfaction) ont été appliquées pour calculer des indices de diversité et de similarité. Elles ont permis de corriger les biais liés à la variation de l’effort d’échantillonnage. Un gradient de continentalité a été observé, les parties côtières étant les plus riches en taxons endémiques d’ACA. Contrairement à la région du Mont Cameroun et aux massifs de Kupe/Bakossi qui ont connu une attention particulière des politiques et des scientifiques, la partie côtière du sud Cameroun, presque aussi riche, reste mal inventoriée pour plusieurs familles végétales.<p><p>Cette analyse à l’échelle du Cameroun a également permis de comparer les patrons d’endémisme des Orchidaceae et des Rubiaceae. Les différences observées seraient principalement dues à la présence d’Orchidaceae terrestres dans les végétations basses et les prairies montagnardes de la dorsale camerounaise alors que les Rubiaceae sont généralement peu représentées dans ces habitats. Au sein des habitats forestiers, la concordance entre les patrons d’endémisme des Orchidaceae et des Rubiaceae remet en question l’utilisation des capacités de dispersion des espèces comme critère pour choisir les familles permettant l’identification des refuges forestiers et semble ainsi confirmer la pertinence de l’utilisation des Orchidaceae comme marqueur chorologique.<p><p>La distribution potentielle a été utilisée pour étudier en détail l’écologie, la distribution et le statut de conservation de Diceratostele gabonensis Summerh. une Orchidaceae endémique de la région guinéo-congolaise uniquement connue d’un faible nombre d’échantillons. Cette méthodologie semble appropriée pour compléter nos connaissances sur la distribution des espèces rares et guider les futurs inventaires en Afrique tropicale.<p><p>(iii) En ACA, les Orchidaceae permettent d’identifier plusieurs centres d’endémisme qui coïncident généralement avec ceux identifiés précédemment pour d’autres familles végétales. Ces constats supportent aussi l’utilisation des Orchidaceae comme marqueur chorologique. La délimitation des aires d’endémisme des Orchidaceae a ainsi permis de proposer une nouvelle carte phytogéographique de l’ACA. Les éléments phytogéographiques propres à chacune des dix phytochories décrites ont été identifiés et leurs affinités floristiques discutées. Les résultats phytogéographiques obtenus (a) soutiennent l’existence d’une barrière phytogéographique matérialisée par la rivière Sanaga entre les deux principaux centres et aires d’endémisme de l’ACA, (b) étendent l’archipel afromontagnard situé principalement au Cameroun au plateau de Jos (Nigeria) et (c) montrent l’importance de la chaîne montagneuse morcelée Ngovayang-Mayombe pour la distribution de l’endémisme en ACA. Cette chaîne de montagne, qui s’étend le long des côtes de l’océan du sud Cameroun au Congo-Brazzaville et qui correspond à plusieurs refuges forestiers identifiés par de nombreux auteurs, est ici considérée comme une seule aire d’endémisme morcelée./<p>Atlantic central Africa (ACA) covers the Lower Guinean Domain, the four islands of the Gulf of Guinea and a part of the afromontane archipelago. Different centres of endemism have been identified into this area and are usually considered as related to glacial forest refuges. However, the origin of this endemism, the localization of the centres and the methods employed to identify these centres are subject to debate. Yet, the localization of these centres of endemism and the identification of the rare plants they harbor is an essential prerequisite to setting up rational conservation policies, and remains a priority for private, institutional and governmental organizations which are dealing with the sustainable management of biodiversity.<p><p>This phytogeographical study focuses on Orchidaceae and analyses the distribution of the taxa endemic to ACA. We use an original dataset resulting from an important sampling efforts and the identification of specimens coming from all the principal herbaria where collections from ACA are housed. During this study, (i) we first identified the taxa endemic to ACA and documented their distribution through several taxonomic and floristic contributions, (ii) we used and developed new methods allowing to correct for sampling bias associated with the use of rare and poorly documented taxa, testing at the same time the use of Orchidaceae as chorological markers, and finally, applying these methods to our dataset, (iii) we delimited the centres of endemism and identified the phytogeographical territories of Orchidaceae in ACA.<p><p>(i) A taxonomic revision of Chamaeangis Schltr. and Stolzia Schltr. respectively was carried out. Seven new taxa were described: Angraecum atlanticum Stévart & Droissart, Chamaeangis spiralis Stévart & Droissart, Chamaeangis lecomtei (Finet) Schltr. var. tenuicalcar Stévart & Droissart, Polystachya engogensis Stévart & Droissart, Polystachya reticulata Stévart & Droissart, Stolzia repens (Rolfe) Summerh var. cleistogama Stévart, Droissart & Simo and Stolzia grandiflora P.J.Cribb subsp. lejolyana Stévart, Droissart & Simo. Several additional taxonomic, phytogeographical and ecological notes were also published. We finally identified 203 Orchidaceae taxa endemic to ACA, among which 193 were used to study the patterns of endemism.<p><p>(ii) In Cameroon, the distribution patterns of both Orchidaceae and Rubiaceae endemic to ACA were studied. Subsampling methods (rarefaction) were applied to calculate diversity and similarity indices and to correct potential bias associated with heterogeneous sampling intensity. A gradient of continentality was confirmed in Cameroon, the coastal part being the richest in taxa endemic to ACA. The Cameroon Mountain and the Kupe/Bakossi mountain massifs have received a great consideration of politics and scientists. On the contrary, the Southern coastal part of Cameroon, though almost as rich as the Northern part, remains poorly known for several plant families.<p>This analysis also allowed us to compare patterns of endemism of Orchidaceae and Rubiaceae. The differences observed could be mainly due to the terrestrial habit of some Orchidaceae, which are only found in the grasslands of the highest part of the Cameroonian volcanic line where endemic Rubiaceae are rare. Within forest habitats, the concordance between the patterns of endemism of Orchidaceae and Rubiaceae question the widespread use of dispersal ability as a selection criterion for the families used to identify forest refuges. This also confirms the relevance of Orchidaceae as chorological marker.<p><p>Species distribution modelling was used of an in depth study of the ecology, the distribution and the conservation status of Diceratostele gabonensis Summerh. an Orchidaceae endemic to the Guineo-Congolian regional centre of endemism which is only known from very few collections. This method is proved to be appropriate to complete our knowledge on the distribution of rare plant species and to guide the future inventories in tropical Africa.<p><p>(iii) In ACA, an analysis of the distribution of endemic Orchidaceae confirmed the presence and location of several centres of endemism previously identified on the basis of other plant families. This result again supports the use of Orchidaceae as a chorological marker. The chorological study of the endemic Orchidaceae allowed us to propose a new phytogeographical map for ACA. Phytogeographical elements for each of the ten phytochoria described were identified and their floristic affinities were also discussed. Our results (a) support the existence of a phytogeographical barrier, materialized by the Sanaga River, between the two main centres and area of endemism of the ACA, (b) extend the limits of the afromontane archipelago to the Jos Plateau in Nigeria and (c) show the importance of the Ngovayang-Mayombe line to explain the distribution of endemism in ACA. This mountainous line, stretching along the ocean coast from Southern Cameroon to Congo-Brazzaville, corresponds to several forest refuges identified by many authors, and is here considered as an unique but discontinuous area of endemism.<p><p><p> / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

Urgent scientifically-informed action is needed to stabilise the Earth System amidst anthropogenic climate change. Particularly, the notable transgression of the ‘biosphere integrity’ Planetary Boundary needs to be addressed. Modern Earth System Models struggle to accurately represent regional to local-scale climate features and biodiversity aspects. Recent developments allow to tackle these issues using Artificial Intelligence. This dissertation focuses on two main aspects: (i) deriving high spatio-temporal resolution climate data from coarser models; and (ii) integrating high-temporal-resolution climate data into Species Distribution Models. Three specific objectives were defined: Obj1 Improving Perfect Prognosis – Statistical Downscaling methods through modern Deep Learning algorithms. Obj2 Downscaling a high-resolution multivariate climate ensemble. Obj3 Employ the resulting dataset to improve Species Distribution Models’ projections. The objectives are connected to the three articles that support this cumulative dissertation. Its scope is limited to the Free State of Saxony, Germany, where local high-resolution climate data and high-quality observations of endangered vascular plant species were employed. From a broader perspective, these efforts should contribute to the overarching goal of bridging the gap between the scales of species distribution and climate models while establishing open-source, reproducible, and scalable containerised frameworks. Recent Deep Learning algorithms were leveraged to accomplish (i). The proposed frameworks enhance previous performance of Perfect Prognosis – Statistical Downscaling approaches, while ensuring repeatability. The key near-surface variables considered are precipitation, water vapour pressure, radiation, wind speed, and, maximum, mean and minimum temperature. The assumptions that support the Perfect Prognosis approach were thoroughly examined, confirming the robustness of the methods. The downscaled ensemble exhibits a novel output resolution of daily 1 km, which can serve as input for multiple climate impact studies, especially for local-scale decision-making and in topographically complex regions. Considerable methodological implementations were proposed and thoroughly analysed to achieve (ii). Despite notable limitations, Species Distribution Models are frequently used in climate change conservation planning. Thus, recent developments in climate data resolution could improve their usefulness and reliability, which have been previously constraint to coarse temporal aggregates in the projection domain. The presented framework provides fine-grained species suitability projections and satisfactory spatio-temporal transferability, albeit worrying trends. These improved projections are a step forward towards tailored conservation efforts. Limitations of Machine Learning methods and Species Distribution Models are addressed. Substantial avenues for future improvements are thoroughly discussed. As results suggest further reduction of suitable habitats, yet another call for swift action towards low-carbon societies is made. This requires maximising climate change mitigation and adaptation measures, along with a swift transition from short-term profit-driven policies to long-term sustainable development, but primarily, a collective shift in consciousness from anthropocentric positions to ecocentric policies and societies.:Contents Declaration of conformity........................................................ I Abstract....................................................................... III Zusammenfassung.................................................................. V Resumen........................................................................ VII Acknowledgments................................................................. IX List of Figures................................................................. XV List of Tables................................................................. XIX Symbols and Acronyms........................................................... XXI I Prelude & Foundations 1 1 Introduction................................................................... 3 1.1 Motivation – Human Impact on Earth....................................... 3 1.2 Earth System Modelling and Downscaling................................... 5 1.3 Biosphere’s Response to Recent Changes................................... 8 1.4 Species Distribution Models.............................................. 9 1.5 Objectives.............................................................. 10 1.6 Scope................................................................... 10 1.7 Outline................................................................. 10 2 Methodological Basis.......................................................... 13 2.1 Introduction to Artificial Intelligence Methods......................... 13 2.1.1 Artificial Intelligence........................................... 13 2.1.2 Machine Learning.................................................. 14 2.1.3 Deep Learning..................................................... 14 2.2 Downscaling Techniques.................................................. 15 2.2.1 Dynamical Downscaling............................................. 15 2.2.2 Statistical Downscaling........................................... 15 2.2.2.1 Model Output Statistics................................... 16 2.2.2.2 Perfect Prognosis......................................... 16 2.3 Species Distribution Models: Temporal Aspects........................... 17 2.4 Computational Framework................................................. 18 2.4.1 High-Performance Computing........................................ 18 2.4.2 Containers........................................................ 18 2.5 Remarks on Reproducibility.............................................. 19 II Articles’ Synthesis 21 3 Data.......................................................................... 23 3.1 Study Area.............................................................. 23 3.2 ReKIS................................................................... 24 3.3 ERA5.................................................................... 24 3.4 CORDEX.................................................................. 24 3.5 Species Occurrences..................................................... 25 3.6 WorldClim............................................................... 26 4 Methodological Implementations................................................ 27 4.1 Advancing Statistical Downscaling....................................... 27 4.1.1 Transfer Function Calibration.................................... 27 4.1.2 Evaluation....................................................... 29 4.1.3 Repeatability.................................................... 29 4.2 Downscaling a Multivariate Ensemble..................................... 30 4.2.1 Transfer Function Adaptations.................................... 30 4.2.2 Validation....................................................... 30 4.2.3 Perfect Prognosis Assumptions Evaluation......................... 31 4.3 Integrating High-Temporal-Resolution into SDMs.......................... 32 4.3.1 Climate Data..................................................... 32 4.3.1.1 Predictor Sets.......................................... 32 4.3.1.2 Temporal Approaches..................................... 33 4.3.2 SDM Implementation............................................... 33 4.3.3 Spatio-Temporal Thinning & Trimming.............................. 33 4.3.4 Meta-analysis.................................................... 34 4.3.5 Pseudo-Reality Assessment........................................ 34 4.3.6 Spatio-Temporal Transferability.................................. 34 5 Results & Discussions......................................................... 35 5.1 Advancing Statistical Downscaling....................................... 35 5.1.1 Performance Improvement.......................................... 35 5.1.2 Repeatability.................................................... 36 5.1.3 Transfer Function Suitability.................................... 38 5.2 Downscaling a Multivariate Ensemble..................................... 39 5.2.1 Transfer Function performance.................................... 39 5.2.2 Bias-Correction.................................................. 40 5.2.3 Pseudo-Reality................................................... 42 5.2.4 Projections...................................................... 43 5.3 Integrating High-Temporal-Resolution into SDMs.......................... 45 5.3.1 Predictor Set Evaluation for H2k................................. 45 5.3.2 Temporal Approach Comparison..................................... 46 5.3.3 Spatio-Temporal Transferability.................................. 47 5.3.4 Suitability Projections.......................................... 47 III Insights 51 6 Summary....................................................................... 53 6.1 Article A1.............................................................. 53 6.2 Article A2.............................................................. 54 6.3 Article A3.............................................................. 56 7 Conclusions and Outlook....................................................... 59 References 65 Articles 81 A1 Repeatable high-resolution statistical downscaling through deep learning..... 83 A2 Downscaling CORDEX Through Deep Learning to Daily 1 km Multivariate Ensemble in Complex Terrain............................................................. 103 A3 Integrating High-Temporal-Resolution Climate Projections into Species Distribu- tion Model..................................................................... 127 / Um das Erdsystem angesichts des anthropogenen Klimawandels zu stabilisieren, sind Maßnahmen auf Basis wissenschaftlicher Erkenntnisse dringend erforderlich. Insbesondere muss die drastisch Überschreitung der planetaren Grenze ‘Integrität der Biosphäre’ angegangen werden. Bisher haben aber Modelle des Erdsystems Schwierigkeiten, regionale bis lokale Klimamerkmale und Aspekte der Biodiversität genau abzubilden. Aktuelle Entwicklungen ermöglichen es, diese Herausforderungen mithilfe von Künstlicher Intelligenz anzugehen. Diese Dissertation konzentriert sich auf zwei Hauptaspekte: (i) die Ableitung von Klimadaten mit hoher räumlicher und zeitlicher Auflösung aus groberen Modellen und (ii) die Integration von Klimadaten mit hoher zeitlicher Auflösung in Modelle zur Artverbreitung. Es wurden drei konkrete Ziele definiert: Ziel1 Verbesserung von Perfect Prognosis – Statistische Downscaling-Methoden durch moderne Deep Learning-Algorithmen Ziel2 Downscaling eines hochauflösenden multivariaten Klimaensembles Ziel3 Verwendung des resultierenden Datensatzes zur Verbesserung von Prognosen in Modellen zur Artverbreitung Diese Ziele werden in drei wissenschaftlichen Artikeln beantwortet, auf die diese kumulative Dissertation sich stützt. Der Anwendungsbereich erstreckt sich auf den Freistaat Sachsen, Deutschland, wo lokale hochauflösende Klimadaten und hochwertige Beobachtungen gefährdeter Gefäßpflanzenarten verwendet wurden. In einer breiteren Perspektive tragen diese Bemühungen dazu bei, die Kluft zwischen regionalen sowie zeitlichen Skalen der Artverbreitung und Klimamodellen zu überbrücken und gleichzeitig Open-Source-, reproduzierbare und skalierbare containerisierte Frameworks zu etablieren. Aktuelle Deep Learning-Algorithmen wurden eingesetzt, um Hauptaspekt (i) zu erreichen. Die vorgeschlagenen Frameworks verbessern die bisherige Leistung von Perfect Prognosis – Statistische Downscaling-Ansätzen und gewährleisten gleichzeitig die Wiederholbarkeit. Die wichtigsten bodennahen Variablen, die berücksichtigt werden, sind Niederschlag, Wasserdampfdruck, Strahlung, Windgeschwindigkeit sowie Maximal-, Durchschnitts- und Minimaltemperatur. Die Annahmen, die den Perfect Prognosis-Ansatz unterstützen, wurden analysiert und bestätigen die Robustheit der Methoden. Das downscaled Ensemble weist eine neuartige Auflösung von 1 km auf Tagesbasis auf, welches als Grundlage für mehrere Studien zu den Auswirkungen des Klimawandels dienen kann, insbesondere für Entscheidungsfindung auf lokaler Ebene und in topografisch komplexen Regionen. Es wurden umfassende methodische Implementierungen vorgeschlagen und analysiert, um Hauptaspekt (ii) zu erreichen. Trotz großer Einschränkungen werden Modelle zur Artverbreitung häufig in der Klimaschutzplanung eingesetzt. Daher könnten aktuelle Entwicklungen in der Klimadatenauflösung deren Nützlichkeit und Zuverlässigkeit verbessern, die bisher auf grobe zeitliche Aggregatformen im Projektionsbereich beschränkt waren. Das vorgestellte Framework bietet feingliedrige Prognosen zur Eignung von Arten und zufriedenstellende räumlich-zeitliche Übertragbarkeit, trotz besorgniserregender Trends. Diese verbesserten Prognosen sind ein Schritt in Richtung maßgeschneiderter Naturschutzmaßnahmen. Einschränkungen von Machine Learning-Methoden und Modellen zur Artverbreitung werden untersucht. Substanzielle Möglichkeiten zur zukünftigen Verbesserung werden ausführlich erörtert. Da die Ergebnisse darauf hinweisen, dass geeignete Lebensräume weiter abnehmen, wird erneut zum schnellen Handeln in Richtung kohlenstoffarmer Gesellschaften aufgerufen. Dies erfordert die Maximierung von Maßnahmen zur Bekämpfung des Klimawandels und zur Anpassung, zusammen mit einem raschen Übergang von kurzfristig Profitorientierten Politiken zu langfristiger nachhaltiger Entwicklung, aber vor allem zu einem kollektiven Bewusstseinswandel von anthropozentrischen Positionen zu ökozentrischen Politiken und Gesellschaften.:Contents Declaration of conformity........................................................ I Abstract....................................................................... III Zusammenfassung.................................................................. V Resumen........................................................................ VII Acknowledgments................................................................. IX List of Figures................................................................. XV List of Tables................................................................. XIX Symbols and Acronyms........................................................... XXI I Prelude & Foundations 1 1 Introduction................................................................... 3 1.1 Motivation – Human Impact on Earth....................................... 3 1.2 Earth System Modelling and Downscaling................................... 5 1.3 Biosphere’s Response to Recent Changes................................... 8 1.4 Species Distribution Models.............................................. 9 1.5 Objectives.............................................................. 10 1.6 Scope................................................................... 10 1.7 Outline................................................................. 10 2 Methodological Basis.......................................................... 13 2.1 Introduction to Artificial Intelligence Methods......................... 13 2.1.1 Artificial Intelligence........................................... 13 2.1.2 Machine Learning.................................................. 14 2.1.3 Deep Learning..................................................... 14 2.2 Downscaling Techniques.................................................. 15 2.2.1 Dynamical Downscaling............................................. 15 2.2.2 Statistical Downscaling........................................... 15 2.2.2.1 Model Output Statistics................................... 16 2.2.2.2 Perfect Prognosis......................................... 16 2.3 Species Distribution Models: Temporal Aspects........................... 17 2.4 Computational Framework................................................. 18 2.4.1 High-Performance Computing........................................ 18 2.4.2 Containers........................................................ 18 2.5 Remarks on Reproducibility.............................................. 19 II Articles’ Synthesis 21 3 Data.......................................................................... 23 3.1 Study Area.............................................................. 23 3.2 ReKIS................................................................... 24 3.3 ERA5.................................................................... 24 3.4 CORDEX.................................................................. 24 3.5 Species Occurrences..................................................... 25 3.6 WorldClim............................................................... 26 4 Methodological Implementations................................................ 27 4.1 Advancing Statistical Downscaling....................................... 27 4.1.1 Transfer Function Calibration.................................... 27 4.1.2 Evaluation....................................................... 29 4.1.3 Repeatability.................................................... 29 4.2 Downscaling a Multivariate Ensemble..................................... 30 4.2.1 Transfer Function Adaptations.................................... 30 4.2.2 Validation....................................................... 30 4.2.3 Perfect Prognosis Assumptions Evaluation......................... 31 4.3 Integrating High-Temporal-Resolution into SDMs.......................... 32 4.3.1 Climate Data..................................................... 32 4.3.1.1 Predictor Sets.......................................... 32 4.3.1.2 Temporal Approaches..................................... 33 4.3.2 SDM Implementation............................................... 33 4.3.3 Spatio-Temporal Thinning & Trimming.............................. 33 4.3.4 Meta-analysis.................................................... 34 4.3.5 Pseudo-Reality Assessment........................................ 34 4.3.6 Spatio-Temporal Transferability.................................. 34 5 Results & Discussions......................................................... 35 5.1 Advancing Statistical Downscaling....................................... 35 5.1.1 Performance Improvement.......................................... 35 5.1.2 Repeatability.................................................... 36 5.1.3 Transfer Function Suitability.................................... 38 5.2 Downscaling a Multivariate Ensemble..................................... 39 5.2.1 Transfer Function performance.................................... 39 5.2.2 Bias-Correction.................................................. 40 5.2.3 Pseudo-Reality................................................... 42 5.2.4 Projections...................................................... 43 5.3 Integrating High-Temporal-Resolution into SDMs.......................... 45 5.3.1 Predictor Set Evaluation for H2k................................. 45 5.3.2 Temporal Approach Comparison..................................... 46 5.3.3 Spatio-Temporal Transferability.................................. 47 5.3.4 Suitability Projections.......................................... 47 III Insights 51 6 Summary....................................................................... 53 6.1 Article A1.............................................................. 53 6.2 Article A2.............................................................. 54 6.3 Article A3.............................................................. 56 7 Conclusions and Outlook....................................................... 59 References 65 Articles 81 A1 Repeatable high-resolution statistical downscaling through deep learning..... 83 A2 Downscaling CORDEX Through Deep Learning to Daily 1 km Multivariate Ensemble in Complex Terrain............................................................. 103 A3 Integrating High-Temporal-Resolution Climate Projections into Species Distribu- tion Model..................................................................... 127 / Acción urgente científicamente informada es necesaria para estabilizar el sistema terrestre en medio del cambio climático antropogénico. En particular, la notable transgresión del límite planetario de ’integridad de la biosfera’ debe abordarse. Los modernos modelos del sistema terrestre tienen dificultades para representar con precisión las características climáticas a escala regional y local, así como los aspectos de la biodiversidad. Desarrollos recientes permiten abordar estos problemas mediante la inteligencia artificial. Esta disertación se enfoca en dos aspectos principales: (i) derivar datos climáticos de alta resolución espacio-temporal a partir de modelos más gruesos; y (ii) integrar datos climáticos de alta resolución temporal en modelos de distribución de especies. Se definieron tres objetivos específicos: Obj1 Mejorar los métodos de pronóstico perfecto – reducción de escala estadística mediante algoritmos modernos de aprendizaje profundo. Obj2 Generar un conjunto climático multivariado de alta resolución. Obj3 Emplear el conjunto de datos resultante para mejorar las proyecciones de los modelos de distribución de especies. Los objetivos están vinculados a los tres artículos que respaldan esta disertación acumulativa. Su alcance se limita al Estado Libre de Sajonia, Alemania, donde se emplearon datos climáticos locales de alta resolución y observaciones de alta calidad de especies de plantas vasculares en peligro de extinción. Desde una perspectiva más amplia, estos esfuerzos deberían contribuir a la meta general de cerrar la brecha entre las escalas de la distribución de especies y los modelos climáticos, mientras que se establecen marcos de trabajo contenedorizados de código abierto, reproducibles y escalables. Algoritmos recientes de aprendizaje profundo fueron aprovechados para lograr (i). Los marcos de trabajo propuestos mejoran el rendimiento previo de los métodos de pronóstico perfecto – reducción de escala estadística, al tiempo que garantizan la repetibilidad. Las variables clave de la superficie cercana consideradas son precipitación, presión de vapor de agua, radiación, velocidad del viento, así como la temperatura máxima, media y mínima. Se examinaron meticulosamente las suposiciones que respaldan el método de pronóstico perfecto, confirmando la robustez de las propuestas. El conjunto reducido de escala exhibe una novedosa resolución diaria de 1 km, el cual puede servir como insumo para múltiples estudios de impacto climático, especialmente para la toma de decisiones a nivel local y en regiones topográficamente complejas. Se propusieron y analizaron minuciosamente considerables implementaciones metodológicas para lograr (ii). A pesar de sus notables limitaciones, los modelos de distribución de especies son utilizados con frecuencia en la planificación de la conservación debido al cambio climático. Por lo tanto, los desarrollos recientes en la resolución de datos climáticos podrían mejorar su utilidad y confiabilidad, ya que antes se limitaban a agregados temporales gruesos en el caso de las proyecciones. El marco de trabajo presentado proporciona proyecciones de idoneidad de especies detalladas y una transferibilidad espacio-temporal satisfactoria, aunque con tendencias preocupantes. Estas proyecciones mejoradas son un paso adelante en los esfuerzos de conservación a la medida. Se abordan las limitaciones de los métodos de aprendizaje automático y de los modelos de distribución de especies. Se discuten a fondo posibilidades sustanciales para futuras mejoras. Dado que los resultados sugieren una mayor reducción de hábitats adecuados, se hace otro llamado a la acción rápida hacia sociedades bajas en carbono. Esto requiere maximizar las medidas de mitigación y adaptación al cambio climático, junto con una transición rápida de políticas orientadas a beneficios a corto plazo hacia un desarrollo sostenible a largo plazo, pero principalmente, un cambio colectivo de conciencia, desde posiciones antropocéntricas hacia políticas y sociedades ecocéntricas.:Contents Declaration of conformity........................................................ I Abstract....................................................................... III Zusammenfassung.................................................................. V Resumen........................................................................ VII Acknowledgments................................................................. IX List of Figures................................................................. XV List of Tables................................................................. XIX Symbols and Acronyms........................................................... XXI I Prelude & Foundations 1 1 Introduction................................................................... 3 1.1 Motivation – Human Impact on Earth....................................... 3 1.2 Earth System Modelling and Downscaling................................... 5 1.3 Biosphere’s Response to Recent Changes................................... 8 1.4 Species Distribution Models.............................................. 9 1.5 Objectives.............................................................. 10 1.6 Scope................................................................... 10 1.7 Outline................................................................. 10 2 Methodological Basis.......................................................... 13 2.1 Introduction to Artificial Intelligence Methods......................... 13 2.1.1 Artificial Intelligence........................................... 13 2.1.2 Machine Learning.................................................. 14 2.1.3 Deep Learning..................................................... 14 2.2 Downscaling Techniques.................................................. 15 2.2.1 Dynamical Downscaling............................................. 15 2.2.2 Statistical Downscaling........................................... 15 2.2.2.1 Model Output Statistics................................... 16 2.2.2.2 Perfect Prognosis......................................... 16 2.3 Species Distribution Models: Temporal Aspects........................... 17 2.4 Computational Framework................................................. 18 2.4.1 High-Performance Computing........................................ 18 2.4.2 Containers........................................................ 18 2.5 Remarks on Reproducibility.............................................. 19 II Articles’ Synthesis 21 3 Data.......................................................................... 23 3.1 Study Area.............................................................. 23 3.2 ReKIS................................................................... 24 3.3 ERA5.................................................................... 24 3.4 CORDEX.................................................................. 24 3.5 Species Occurrences..................................................... 25 3.6 WorldClim............................................................... 26 4 Methodological Implementations................................................ 27 4.1 Advancing Statistical Downscaling....................................... 27 4.1.1 Transfer Function Calibration.................................... 27 4.1.2 Evaluation....................................................... 29 4.1.3 Repeatability.................................................... 29 4.2 Downscaling a Multivariate Ensemble..................................... 30 4.2.1 Transfer Function Adaptations.................................... 30 4.2.2 Validation....................................................... 30 4.2.3 Perfect Prognosis Assumptions Evaluation......................... 31 4.3 Integrating High-Temporal-Resolution into SDMs.......................... 32 4.3.1 Climate Data..................................................... 32 4.3.1.1 Predictor Sets.......................................... 32 4.3.1.2 Temporal Approaches..................................... 33 4.3.2 SDM Implementation............................................... 33 4.3.3 Spatio-Temporal Thinning & Trimming.............................. 33 4.3.4 Meta-analysis.................................................... 34 4.3.5 Pseudo-Reality Assessment........................................ 34 4.3.6 Spatio-Temporal Transferability.................................. 34 5 Results & Discussions......................................................... 35 5.1 Advancing Statistical Downscaling....................................... 35 5.1.1 Performance Improvement.......................................... 35 5.1.2 Repeatability.................................................... 36 5.1.3 Transfer Function Suitability.................................... 38 5.2 Downscaling a Multivariate Ensemble..................................... 39 5.2.1 Transfer Function performance.................................... 39 5.2.2 Bias-Correction.................................................. 40 5.2.3 Pseudo-Reality................................................... 42 5.2.4 Projections...................................................... 43 5.3 Integrating High-Temporal-Resolution into SDMs.......................... 45 5.3.1 Predictor Set Evaluation for H2k................................. 45 5.3.2 Temporal Approach Comparison..................................... 46 5.3.3 Spatio-Temporal Transferability.................................. 47 5.3.4 Suitability Projections.......................................... 47 III Insights 51 6 Summary....................................................................... 53 6.1 Article A1.............................................................. 53 6.2 Article A2.............................................................. 54 6.3 Article A3.............................................................. 56 7 Conclusions and Outlook....................................................... 59 References 65 Articles 81 A1 Repeatable high-resolution statistical downscaling through deep learning..... 83 A2 Downscaling CORDEX Through Deep Learning to Daily 1 km Multivariate Ensemble in Complex Terrain............................................................. 103 A3 Integrating High-Temporal-Resolution Climate Projections into Species Distribu- tion Model..................................................................... 127