pith. machine review for the scientific record. sign in

arxiv: 2605.04286 · v1 · submitted 2026-05-05 · 💻 cs.LG · stat.AP· stat.CO

Recognition: 3 theorem links

· Lean Theorem

Probabilistic Classification and Uncertainty Quantification of Sahara Desert Climate Using Feedforward Neural Networks

Stephen Tivenan , Indranil Sahoo , Yanjun Qian
Authors on Pith no claims yet

Pith reviewed 2026-05-08 18:01 UTC · model grok-4.3

classification 💻 cs.LG stat.APstat.CO
keywords climate classificationneural networksprobabilistic modelingSahara Desertuncertainty quantificationtemporal evolutiondesertificationKöppen-Trewartha
0
0 comments X

The pith

A neural network outputs probabilities for climate classes instead of fixed labels, applied to the Sahara over thirty years.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper trains a feedforward neural network on climate data from 1960-1970 to assign probabilities to different climate categories at locations across the Sahara. This replaces the sharp boundaries of traditional classification systems with estimates that reflect uncertainty, especially useful in gradual transition zones. The model is then run on data through 1989 to check whether the probability assignments stay consistent and to track how they shift over time. Such shifts are analyzed to point toward areas of ongoing change, such as advancing or retreating desert conditions.

Core claim

A feedforward artificial neural network can map climate variables to probability distributions over climate classes, allowing classification that includes uncertainty estimates and enabling the detection of temporal changes in zone probabilities across the Sahara from 1960 to 1989.

What carries the argument

A feedforward neural network that takes climate data inputs and produces a vector of probabilities for each possible climate class.

Load-bearing premise

The relationships between climate variables and class probabilities learned from the 1960-1970 training data remain stable and unbiased when the network is applied to later years.

What would settle it

Large, systematic differences between the network's probability outputs and the climate classes that would be assigned by direct application of the deterministic rules to the same later-period data would undermine the claim of reliable uncertainty-aware classification.

Figures

Figures reproduced from arXiv: 2605.04286 by Indranil Sahoo, Stephen Tivenan, Yanjun Qian.

Figure 1
Figure 1. Figure 1: Spatial distribution of total precipitation and average temperature (in centimeters and Celsius) view at source ↗
Figure 2
Figure 2. Figure 2: Spatial maps of winsorized P/R values (left column) and corresponding KTC dry climate classifications (arid, semi-arid, and non-arid; right column) for the candidate years 1975, 1980, and 1985. The P/R values are winsorized between 0.001 and 2 to enhance visualization clarity. precipitation and temperature data to compute the KTC-based classification at each spatial location. This process results in each p… view at source ↗
Figure 3
Figure 3. Figure 3: Architecture of the feedforward neural network for spatio-temporal classification at location view at source ↗
Figure 4
Figure 4. Figure 4: Predicted probability maps for arid, semi-arid, and non-arid regions over the Sahara and Sahel, view at source ↗
Figure 5
Figure 5. Figure 5: Comparison of climate classifications obtained from the proposed neural network-based method view at source ↗
Figure 6
Figure 6. Figure 6: Spatial distribution of the coefficient of variation (CV) of the dominant predicted aridity view at source ↗
Figure 7
Figure 7. Figure 7: Country-specific coefficient of variation (CV) maps showing fluctuations in the dominant aridity view at source ↗
Figure 8
Figure 8. Figure 8: Classification performance metrics (precision, recall, and F1-score) over time for arid (blue), view at source ↗
read the original abstract

Climate classification plays a vital role in agricultural planning, hydrological studies, and climate science. One of the most widely used systems for classifying global climate zones is the K\"oppen-Trewartha (KT) classification. However, the KT classification is fundamentally deterministic, offering discrete labels to spatial locations without accounting for uncertainties in classification. In this paper, we provide a framework for probabilistic modeling of climatic zones. We implement a feedforward artificial neural network (ANN) for classification, allowing for efficient, uncertainty-aware categorization of climatic regions, thereby offering a more nuanced understanding of transitional climate zones compared to traditional deterministic methods. We apply this method to the Sahara Desert region over the 30-year period of 1960 - 1989, using data at more than 400,000 space-time locations from the first 11 years to train our model. We assess the model's short- and long-term classification capabilities to evaluate its stability and accuracy over time. We also compare the probabilistic classification from our model with the traditional KT classification. In addition, we use fluctuation analysis methods to highlight the temporal evolution of climatic zones across the Sahara region and identify areas undergoing significant flux of probabilities of their climate classes, providing insights into broader trends in desertification.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript proposes a feedforward neural network (ANN) to produce probabilistic classifications of Köppen-Trewartha (KT) climate zones over the Sahara Desert for 1960–1989. The model is trained on data from the first 11 years (1960–1970) at >400,000 space-time locations and is claimed to deliver uncertainty-aware outputs that better capture transitional zones than deterministic KT labels; short- and long-term stability is assessed, and fluctuation analysis is used to track temporal evolution of class probabilities and desertification signals.

Significance. If the stability and calibration claims hold, the work would supply a practical probabilistic alternative to deterministic climate classification, potentially improving identification of transitional zones and flux regions relevant to desertification studies. The temporal hold-out design and fluctuation analysis constitute a reasonable attempt to address non-stationarity, though the absence of quantitative validation metrics limits the immediate impact.

major comments (3)
  1. [Methods] Methods section: the feedforward ANN architecture (number of layers, hidden units, activation functions), loss function, and output probability mechanism (e.g., softmax) are not specified. These details are load-bearing for the central claim of “uncertainty-aware categorization,” because standard cross-entropy training on a deterministic KT target does not automatically yield well-calibrated probabilities or epistemic uncertainty without additional techniques such as temperature scaling or ensemble methods.
  2. [Results] Results / evaluation section: no quantitative performance numbers (accuracy, Brier score, expected calibration error, or reliability diagrams) are reported for the held-out period after 1970. Without these, it is impossible to verify the claim that the model maintains “stable and meaningful probability estimates” under potential distribution shift, which directly undermines the superiority assertion over deterministic KT labels.
  3. [Fluctuation analysis] Fluctuation analysis paragraph: the method used to quantify “significant flux of probabilities” and to identify desertification trends is not described (no equations, window sizes, or statistical thresholds). This step is central to the paper’s claim of providing “insights into broader trends,” yet remains non-reproducible and unvalidated against null models of interannual variability.
minor comments (2)
  1. [Abstract] The abstract states that “short- and long-term classification capabilities were assessed” but supplies no concrete temporal split or metric; this should be clarified with explicit year ranges and scores in the main text.
  2. [Introduction] Notation for climate variables and KT class labels is introduced without a table or appendix; a concise summary table would improve readability.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive and detailed feedback, which has identified key areas where the manuscript can be strengthened for clarity, reproducibility, and rigor. We address each major comment below and confirm that revisions will be made to incorporate the requested details and evaluations.

read point-by-point responses

  1. Referee: [Methods] Methods section: the feedforward ANN architecture (number of layers, hidden units, activation functions), loss function, and output probability mechanism (e.g., softmax) are not specified. These details are load-bearing for the central claim of “uncertainty-aware categorization,” because standard cross-entropy training on a deterministic KT target does not automatically yield well-calibrated probabilities or epistemic uncertainty without additional techniques such as temperature scaling or ensemble methods.

    Authors: We agree that the Methods section requires these specifications for full reproducibility and to support interpretation of the probabilistic outputs. The original submission omitted explicit details on the architecture and training procedure. In the revised manuscript we will expand the Methods section to describe the feedforward ANN (including layer counts, hidden unit sizes, activation functions, loss function, and softmax output layer) and clarify that the probabilities arise from the softmax normalization of the final layer. We acknowledge that this yields a predictive probability distribution over classes rather than calibrated probabilities or epistemic uncertainty (which would require ensembles or Bayesian approaches); the primary advantage remains the ability to represent transitional zones via soft assignments instead of hard deterministic labels. We will add this discussion of limitations. revision: yes

  2. Referee: [Results] Results / evaluation section: no quantitative performance numbers (accuracy, Brier score, expected calibration error, or reliability diagrams) are reported for the held-out period after 1970. Without these, it is impossible to verify the claim that the model maintains “stable and meaningful probability estimates” under potential distribution shift, which directly undermines the superiority assertion over deterministic KT labels.

    Authors: We accept that the absence of quantitative metrics limits the strength of the stability claims. The original manuscript relied on qualitative visual comparisons of probability maps and class transitions over the temporal hold-out to illustrate stability. In the revised version we will add quantitative evaluations on the post-1970 held-out data, including accuracy, Brier score, expected calibration error, and reliability diagrams, to directly assess calibration and performance under distribution shift and to provide a clearer comparison against deterministic KT labels. revision: yes

  3. Referee: [Fluctuation analysis] Fluctuation analysis paragraph: the method used to quantify “significant flux of probabilities” and to identify desertification trends is not described (no equations, window sizes, or statistical thresholds). This step is central to the paper’s claim of providing “insights into broader trends,” yet remains non-reproducible and unvalidated against null models of interannual variability.

    Authors: We agree that the fluctuation analysis requires a complete methodological description to ensure reproducibility and to substantiate the desertification insights. The original text provided only a high-level mention. In the revision we will supply the full details, including the equations for computing probability flux, the specific window sizes and aggregation steps, the statistical thresholds applied, and explicit comparisons against null models of interannual variability to validate the detected trends. revision: yes

Circularity Check

0 steps flagged

No circularity detected in the neural network classification approach

full rationale

The paper applies a standard feedforward ANN trained on 1960-1970 data to generate probabilistic climate classifications over 1960-1989, with temporal assessment for stability. No equations, derivations, or self-citations are shown that reduce any prediction or uniqueness claim to the inputs by construction. The probabilistic outputs follow directly from the trained model's softmax layer applied to held-out periods, constituting empirical modeling rather than a tautological chain. The framework remains self-contained against external benchmarks like KT labels without load-bearing self-referential steps.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim depends on the assumption that climate variables contain sufficient signal for a neural network to learn stable class probabilities and that temporal extrapolation from the training window remains valid.

free parameters (1)
  • neural network weights and biases
    All parameters are fitted to the 1960-1970 training data to minimize classification loss.
axioms (1)
  • domain assumption The deterministic Köppen-Trewartha labels can be meaningfully represented as probability distributions learned by a feedforward network.
    Invoked when the authors replace discrete labels with probabilistic outputs without proving that the underlying climate data support such a representation.

pith-pipeline@v0.9.0 · 5527 in / 1316 out tokens · 87549 ms · 2026-05-08T18:01:09.969460+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

129 extracted references · 4 canonical work pages · 1 internal anchor

  1. [1]

    Environmetrics , volume=

    An illustration of model agnostic explainability methods applied to environmental data , author=. Environmetrics , volume=. 2023 , publisher=

    2023

  2. [2]

    , citeulike-article-id =

    Bradski, G. , citeulike-article-id =. Dr. Dobb's Journal of Software Tools , keywords =

  3. [3]

    1962 , publisher=

    A note on the classification of dry climates in the Koppen system , author=. 1962 , publisher=

    1962

  4. [4]

    Algorithms for smoothing data with periodic and parametric splines , journal =

    P Dierckx , abstract =. Algorithms for smoothing data with periodic and parametric splines , journal =. 1982 , issn =. doi:https://doi.org/10.1016/0146-664X(82)90043-0 , url =

    work page doi:10.1016/0146-664x(82)90043-0 1982

  5. [5]

    and Haberland, Matt and Reddy, Tyler and Cournapeau, David and Burovski, Evgeni and Peterson, Pearu and Weckesser, Warren and Bright, Jonathan and

    Virtanen, Pauli and Gommers, Ralf and Oliphant, Travis E. and Haberland, Matt and Reddy, Tyler and Cournapeau, David and Burovski, Evgeni and Peterson, Pearu and Weckesser, Warren and Bright, Jonathan and. Nature Methods , year =

  6. [6]

    and Millman, K

    Harris, Charles R. and Millman, K. Jarrod and van der Walt, Stéfan J and Gommers, Ralf and Virtanen, Pauli and Cournapeau, David and Wieser, Eric and Taylor, Julian and Berg, Sebastian and Smith, Nathaniel J. and Kern, Robert and Picus, Matti and Hoyer, Stephan and van Kerkwijk, Marten H. and Brett, Matthew and Haldane, Allan and Fernández del Río, Jaime ...

  7. [7]

    Hunter, J. D. , Title =. Computing In Science & Engineering , Volume =

  8. [8]

    2017 , eprint=

    Adam: A Method for Stochastic Optimization , author=. 2017 , eprint=

    2017

  9. [9]

    and Jarvis, Claire H

    Rigol, Juan P. and Jarvis, Claire H. and Stuart, Neil , copyright =. Artificial neural networks as a tool for spatial interpolation , volume =. International Journal of Geographical Information Science : IJGIS , language =

  10. [10]

    Sensors (Basel, Switzerland) , keywords =

    Daily Estimation of Global Solar Irradiation and Temperatures Using Artificial Neural Networks through the Virtual Weather Station Concept in Castilla and León, Spain , volume =. Sensors (Basel, Switzerland) , keywords =

  11. [11]

    and Gopal, Sucharita and Kaufmann, Robert K

    Snell, Seth E. and Gopal, Sucharita and Kaufmann, Robert K. , address =. Spatial Interpolation of Surface Air Temperatures Using Artificial Neural Networks: Evaluating Their Use for Downscaling GCMs , volume =. Journal of Climate , keywords =

  12. [12]

    The role of artificial neural network and machine learning in utilizing spatial information , volume =

    Goel, Akash and Goel, Amit Kumar and Kumar, Adesh , address =. The role of artificial neural network and machine learning in utilizing spatial information , volume =. Spatial Information Research (Online) , keywords =

  13. [13]

    Isa, M. H. and Rahardjo, H. and Rezaur, R. B. and Mustafa, M. R. and Arif, A. and Katsanos, Dimitrios , address =. Artificial Neural Network Modeling for Spatial and Temporal Variations of Pore-Water Pressure Responses to Rainfall , volume =. Advances in Meteorology , keywords =

  14. [14]

    Programming an Artificial Neural Network Tool for Spatial Interpolation in GIS - A Case Study for Indoor Radio Wave Propagation of WLAN , volume =

    Sen, Alper and Gümüsay, M Umit and Kavas, Aktül and Bulucu, Umut , address =. Programming an Artificial Neural Network Tool for Spatial Interpolation in GIS - A Case Study for Indoor Radio Wave Propagation of WLAN , volume =. Sensors (Basel, Switzerland) , keywords =

  15. [15]

    Land Degradation & Development , volume=

    Global desertification vulnerability to climate change and human activities , author=. Land Degradation & Development , volume=. 2020 , publisher=

    2020

  16. [16]

    and Funk, Chris and Fink, Andreas H

    Nicholson, Sharon E. and Funk, Chris and Fink, Andreas H. , copyright =. Rainfall over the African continent from the 19th through the 21st century , volume =. Global and Planetary Change , keywords =

  17. [17]

    The perceptron: A probabilistic model for information storage and organization in the brain , volume =

    Rosenblatt, F , address =. The perceptron: A probabilistic model for information storage and organization in the brain , volume =. Psychological Review , language =

  18. [18]

    The Bulletin of Mathematical Biophysics , volume=

    A logical calculus of the ideas immanent in nervous activity , author=. The Bulletin of Mathematical Biophysics , volume=. 1943 , publisher=

    1943

  19. [19]

    and Munda, Josiah L

    Babatunde, Olubayo M. and Munda, Josiah L. and Hamam, Yskandar , address =. Exploring the Potentials of Artificial Neural Network Trained with Differential Evolution for Estimating Global Solar Radiation , volume =. Energies (Basel) , language =

  20. [20]

    Winter wheat yield prediction using convolutional neural networks from environmental and phenological data , volume =

    Srivastava, Amit Kumar and Safaei, Nima and Khaki, Saeed and Lopez, Gina and Zeng, Wenzhi and Ewert, Frank and Gaiser, Thomas and Rahimi, Jaber , address =. Winter wheat yield prediction using convolutional neural networks from environmental and phenological data , volume =. Scientific Reports , language =

  21. [21]

    and Philippopoulos, K

    Alimissis, A. and Philippopoulos, K. and Tzanis, C.G. and Deligiorgi, D. , copyright =. Spatial estimation of urban air pollution with the use of artificial neural network models , volume =. Atmospheric Environment (1994) , language =

    1994

  22. [22]

    and Booij, M.J

    Gholami, V. and Booij, M.J. and Nikzad Tehrani, E. and Hadian, M.A. , copyright =. Spatial soil erosion estimation using an artificial neural network (ANN) and field plot data , volume =. Catena (Giessen) , language =

  23. [23]

    , copyright =

    Nag, Pratik and Sun, Ying and Reich, Brian J. , copyright =. Spatio-temporal DeepKriging for interpolation and probabilistic forecasting , volume =. Spatial Statistics , keywords =

  24. [24]

    and Yang, Taibao and Yang, Xuemei and Wang, Congqiang , address =

    Ghebrezgabher, Mihretab G. and Yang, Taibao and Yang, Xuemei and Wang, Congqiang , address =. Assessment of desertification in Eritrea: land degradation based on Landsat images , volume =. Journal of Arid Land , keywords =

  25. [25]

    Analysis of spatiotemporal changes and driving factors of desertification in the Africa Sahel , volume =

    Yang, Zuowei and Gao, Xin and Lei, Jiaqiang and Meng, Xiaoyu and Zhou, Na , copyright =. Analysis of spatiotemporal changes and driving factors of desertification in the Africa Sahel , volume =. Catena (Giessen) , keywords =

  26. [26]

    Recent trends in the Central and Western Sahel rainfall regime (1990–2007) , volume =

    Lebel, Thierry and Ali, Abdou , copyright =. Recent trends in the Central and Western Sahel rainfall regime (1990–2007) , volume =. Journal of Hydrology (Amsterdam) , language =

    1990

  27. [27]

    Persistence and success of the Sahel desertification narrative , volume =

    Gangneron, Fabrice and Pierre, Caroline and Robert, Elodie and Kergoat, Laurent and Grippa, Manuela and Guichard, Françoise and Hiernaux, Pierre and Leauthaud, Crystele , address =. Persistence and success of the Sahel desertification narrative , volume =. Regional Environmental Change , language =

  28. [28]

    Desertification, land use, and the transformation of global drylands , volume =

    Bestelmeyer, Brandon T and Okin, Gregory S and Duniway, Michael C and Archer, Steven R and Sayre, Nathan F and Williamson, Jebediah C and Herrick, Jeffrey E , copyright =. Desertification, land use, and the transformation of global drylands , volume =. Frontiers in Ecology and the Environment , keywords =

  29. [29]

    and Ravi, Sujith and Runyan, Christiane W

    D’Odorico, Paolo and Bhattachan, Abinash and Davis, Kyle F. and Ravi, Sujith and Runyan, Christiane W. , address =. Global desertification: Drivers and feedbacks , volume =. Advances in Water Resources , keywords =

  30. [30]

    and Uecker, Hannes and Feudel, Ulrike and Meron, Ehud , address =

    Zelnik, Yuval R. and Uecker, Hannes and Feudel, Ulrike and Meron, Ehud , address =. Desertification by front propagation? , volume =. Journal of Theoretical Biology , keywords =

  31. [31]

    Vogt, J. V. and Safriel, U. and Von Maltitz, G. and Sokona, Y. and Zougmore, R. and Bastin, G. and Hill, J. , address =. Monitoring and assessment of land degradation and desertification: Towards new conceptual and integrated approaches , volume =. Land Degradation & Development , keywords =

  32. [32]

    Desertification baseline: A bottleneck for addressing desertification , volume =

    Gui, Dongwei and Liu, Qi and Martínez-Valderrama, Jaime and Abd-Elmabod, Sameh Kotb and Ahmed, Zeeshan and Xu, Zhiwei and Lei, Jiaqiang , copyright =. Desertification baseline: A bottleneck for addressing desertification , volume =. Earth-science Reviews , keywords =

  33. [33]

    Mark Stafford and Lambin, Eric F and Turner, B.L

    Reynolds, James F and Smith, D. Mark Stafford and Lambin, Eric F and Turner, B.L. II and Mortimore, Michael and Batterbury, Simon P.J and Downing, Thomas E and Dowlatabadi, Hadi and Fernández, Roberto J and Herrick, Jeffrey E and Huber-Sannwald, Elisabeth and Jiang, Hong and Leemans, Rik and Lynam, Tim and Maestre, Fernando T and Ayarza, Miguel and Walke...

  34. [34]

    and Santos Felix, Evaldo and Ferreira, Karoline R

    Vendruscolo, Jhony and Perez Marin, Aldrin M. and Santos Felix, Evaldo and Ferreira, Karoline R. and Cavalheiro, Wanderson C. S. and Fernandes, Izaias M. , address =. Monitoring desertification in semiarid Brazil: Using the Desertification Degree Index (DDI) , volume =. Land Degradation & Development , keywords =

  35. [35]

    What Has Caused Desertification in China? , volume =

    Feng, Qi and Ma, Hua and Jiang, Xuemei and Wang, Xin and Cao, Shixiong , address =. What Has Caused Desertification in China? , volume =. Scientific Reports , keywords =

  36. [36]

    Regional desertification: A global synthesis , volume =

    Helldén, Ulf and Tottrup, Christian , copyright =. Regional desertification: A global synthesis , volume =. Global and Planetary Change , keywords =

  37. [37]

    Burrell, A. L. and Evans, J. P. and De Kauwe, M. G. , address =. Anthropogenic climate change has driven over 5 million km2 of drylands towards desertification , volume =. Nature Communications , keywords =

  38. [38]

    Vulnerability of Turkey to desetification with respect to precipitation and aridity conditions , volume =

    TÜRKEŞ, Murat , issn =. Vulnerability of Turkey to desetification with respect to precipitation and aridity conditions , volume =. Turkish Journal of Engineering & Environmental Sciences , keywords =

  39. [39]

    Drought and Desertification in Iran , volume =

    Emadodin, Iraj and Reinsch, Thorsten and Taube, Friedhelm , address =. Drought and Desertification in Iran , volume =. Hydrology , keywords =

  40. [40]

    Adam: A Method for Stochastic Optimization

    Adam: A method for stochastic optimization , author=. arXiv preprint arXiv:1412.6980 , year=

    work page internal anchor Pith review arXiv

  41. [41]

    2015 , howpublished=

    Keras , author=. 2015 , howpublished=

    2015

  42. [42]

    Climate classification revisited: from K

    Belda, Michal and Holtanov. Climate classification revisited: from K. Climate Research , volume=

  43. [43]

    Observed climate variability over Chad using multiple observational and reanalysis datasets , volume =

    Maharana, Pyarimohan and Abdel-Lathif, Ahmat Younous and Pattnayak, Kanhu Charan , copyright =. Observed climate variability over Chad using multiple observational and reanalysis datasets , volume =. Global and Planetary Change , keywords =

  44. [44]

    Earth's Future , volume=

    Climate risk and food security in Mali: A historical perspective on adaptation , author=. Earth's Future , volume=. 2017 , publisher=

    2017

  45. [45]

    CIBIO/InBIO Research Center in Biodiversity and Genetic Resources

    Geographical Atlas of Mauritania , author=. CIBIO/InBIO Research Center in Biodiversity and Genetic Resources. Biodeserts Report EN-02 , year=

  46. [46]

    2025 , url =

    Encyclopaedia Britannica , title =. 2025 , url =

    2025

  47. [47]

    Evaluating observed and projected future climate changes for the Arctic using the K

    Feng, Song and Ho, Chang-Hoi and Hu, Qi and Oglesby, Robert J and Jeong, Su-Jong and Kim, Baek-Min , journal=. Evaluating observed and projected future climate changes for the Arctic using the K. 2012 , publisher=

    2012

  48. [48]

    Evaluation of new CORDEX simulations using an updated K

    Remedio, Armelle Reca and Teichmann, Claas and Buntemeyer, Lars and Sieck, Kevin and Weber, Torsten and Rechid, Diana and Hoffmann, Peter and Nam, Christine and Kotova, Lola and Jacob, Daniela , journal=. Evaluation of new CORDEX simulations using an updated K. 2019 , publisher=

    2019

  49. [49]

    Atmospheric Science Letters , volume=

    Changes in Koppen--Trewartha climate classification over South America from RegCM4 projections , author=. Atmospheric Science Letters , volume=. 2017 , publisher=

    2017

  50. [50]

    Assessment of climate change in Europe from an ensemble of regional climate models by the use of K

    Gallardo, Clemente and Gil, Victoria and Hagel, Edit and Tejeda, C. Assessment of climate change in Europe from an ensemble of regional climate models by the use of K. International Journal of Climatology , volume=

  51. [51]

    Use of the Köppen-Trewartha climate classification to evaluate climatic refugia in statistically derived ecoregions for the People's Republic of China , volume =

    Baker, Barry and Diaz, Henry and Hargrove, William and Hoffman, Forrest , address =. Use of the Köppen-Trewartha climate classification to evaluate climatic refugia in statistically derived ecoregions for the People's Republic of China , volume =. Climatic Change , keywords =

  52. [52]

    Evaluation of

    Belda, Michal and Holtanová, Eva and Halenka, Tomáš and Kalvová, Jaroslava and Hlávka, Zdeněk , copyright =. Evaluation of. Climate Research , language =

  53. [53]

    Global warming-induced changes in climate zones based on CMIP5 projections , volume =

    Belda, Michal and Holtanová, Eva and Kalvová, Jaroslava and Halenka, Tomáš , copyright =. Global warming-induced changes in climate zones based on CMIP5 projections , volume =. Climate Research , language =

  54. [54]

    Ecological Economics , volume=

    Climate change and agriculture in the Sudan: Impact pathways beyond changes in mean rainfall and temperature , author=. Ecological Economics , volume=. 2020 , publisher=

    2020

  55. [55]

    Trends and impacts of climate change on crop production in Burkina Faso , volume =

    Sanou, Charles Lamoussa and Neya, Oblé and Agodzo, Sampson Kwaku and Antwi-Agyei, Philip and Bessah, Enoch and Belem, Mahamadou and Balima, Larba Hubert , issn =. Trends and impacts of climate change on crop production in Burkina Faso , volume =. Journal of Water and Climate Change , keywords =

  56. [56]

    The physiologic climate of Nigeria , volume =

    Eludoyin, Oyenike Mary and Adelekan, Ibidun Onikepo , address =. The physiologic climate of Nigeria , volume =. International Journal of Biometeorology , keywords =

  57. [57]

    Earth Interactions , volume=

    Analysis of mean climate conditions in Senegal (1971--98) , author=. Earth Interactions , volume=

    1971

  58. [58]

    Salih, Abubakr A. M. and Körnich, Heiner and Tjernström, Michael , address =. Climate impact of deforestation over South Sudan in a regional climate model , volume =. International Journal of Climatology , keywords =

  59. [59]

    Journal of Climate , volume=

    On using a clustering approach for global climate classification , author=. Journal of Climate , volume=

  60. [60]

    Data-Driven versus K

    Lasantha, Vajira and Oki, Taikan and Tokuda, Daisuke , journal=. Data-Driven versus K. 2022 , publisher=

    2022

  61. [61]

    Geographical Review , volume=

    An approach toward a rational classification of climate , author=. Geographical Review , volume=. 1948 , publisher=

    1948

  62. [62]

    Handbuch der klimatologie , year=

    Das geographische System de Klimate , author=. Handbuch der klimatologie , year=

  63. [63]

    1954 , publisher=

    An introduction to climate , author=. 1954 , publisher=

    1954

  64. [64]

    1943 , publisher=

    An introduction to weather and climate , author=. 1943 , publisher=

    1943

  65. [65]

    High-resolution (1 km) K

    Beck, Hylke E and McVicar, Tim R and Vergopolan, Noemi and Berg, Alexis and Lutsko, Nicholas J and Dufour, Ambroise and Zeng, Zhenzhong and Jiang, Xin and van Dijk, Albert IJM and Miralles, Diego G , journal=. High-resolution (1 km) K. 2023 , publisher=

    2023

  66. [66]

    Present and future K

    Beck, Hylke E and Zimmermann, Niklaus E and McVicar, Tim R and Vergopolan, Noemi and Berg, Alexis and Wood, Eric F , journal=. Present and future K. 2018 , publisher=

    2018

  67. [67]

    Climatic Change , volume=

    Assessment of CMIP5 global model simulations and climate change projections for the 21 st century using a modified Thornthwaite climate classification , author=. Climatic Change , volume=. 2014 , publisher=

    2014

  68. [68]

    Assessing the ability of the K

    Triantafyllou, GN and Tsonis, AA , journal=. Assessing the ability of the K. 1994 , publisher=

    1994

  69. [69]

    Geographical Review , volume=

    Problems in the classification of climates , author=. Geographical Review , volume=. 1943 , publisher=

    1943

  70. [70]

    Geographische Zeitschrift , volume=

    Versuch einer Klassifikation der Klimate, vorzugsweise nach ihren Beziehungen zur Pflanzenwelt , author=. Geographische Zeitschrift , volume=. 1900 , publisher=

    1900

  71. [71]

    K\"oppen-Geiger Climate Classification , booktitle =

    Barney Warf , year =. K\"oppen-Geiger Climate Classification , booktitle =. doi:10.4135/9781412939591 , url =

    work page doi:10.4135/9781412939591

  72. [72]

    and Rodell, M

    Beaudoing, H. and Rodell, M. , title =. 2020 , publisher =

    2020

  73. [73]

    Journal of Climate , volume=

    Development of a 50-year high-resolution global dataset of meteorological forcings for land surface modeling , author=. Journal of Climate , volume=

  74. [74]

    Procedia Computer Science , volume=

    Climate classifications: the value of unsupervised clustering , author=. Procedia Computer Science , volume=. 2012 , publisher=

    2012

  75. [75]

    Predicting climate types for the Continental

    Sathiaraj, D and Huang, Xiaoqi and Chen, J , journal=. Predicting climate types for the Continental. 2019 , publisher=

    2019

  76. [76]

    Cartography and Geographic Information Science , volume=

    Artificial neural networks as a method of spatial interpolation for digital elevation models , author=. Cartography and Geographic Information Science , volume=. 2002 , publisher=

    2002

  77. [77]

    Transactions of the ASAE , volume=

    Spatial interpolation of weather variables for single locations using artificial neural networks , author=. Transactions of the ASAE , volume=. 2004 , publisher=

    2004

  78. [78]

    Spatial Statistics , pages=

    Echo state network-enhanced symbolic regression for spatio-temporal binary stochastic cellular automata , author=. Spatial Statistics , pages=. 2024 , publisher=

    2024

  79. [79]

    Environmetrics , volume=

    Calibrated forecasts of quasi-periodic climate processes with deep echo state networks and penalized quantile regression , author=. Environmetrics , volume=. 2024 , publisher=

    2024

  80. [80]

    Proceedings of the AAAI conference on artificial intelligence , volume=

    Predicting the next location: A recurrent model with spatial and temporal contexts , author=. Proceedings of the AAAI conference on artificial intelligence , volume=

Showing first 80 references.