Regresar

Resumen del producto

Acevedo-Acosta, J.D., A., Martinez Lopez & L., Carro-Calvo (2025). Diatom-based climate reconstruction and modeling: Modern calibration set and paleodata. Paleoceanography and Paleoclimatology. 40(4): e2024PA005006. DOI: 10.1029/2024PA005006.

Diatom-based climate reconstruction and modeling: Modern calibration set and paleodata

Juan David Acevedo-Acosta 1, Aida Martinez Lopez 2 y Leopoldo Carro-Calvo 3

1 Instituto Politécnico Nacional, CICIMAR
2 Instituto Politécnico Nacional, Centro Interdisciplinario de Ciencias Marinas, Instituto Politécnico Nacional, CICIMAR
3 Universidad Rey Juan Carlos, Madrid, España

The past 2000 years have been crucial to evaluating climate models and their projections. Weexplored an artificial neural network (ANN) to reconstruct (the past 600 years) and model the climate for thecoming decades. For reconstruction of air temperature (AT) and sea surface temperature (SST), time series ofinstrumental data and diatoms fluxes, contemporaneous and preserved in a laminated sediment core from theAlfonso Basin, were used to train a feedforward ANN (6 neurons for AT and 14, 14 for SST). We demonstratedthat ANNs have an advantage in simplifying and determining the nonlinear relations of the climate system withthe assemblages of planktonic diatoms identified as AT and SST proxies. During validation, significant distancecorrelations were achieved for reconstructions (r2dist = 0.65 for AT and 0.63 for SST; p < 0.005) and prediction(r2dist = 0.76 for AT and 0.70 for SST; p < 0.005). In this way, a better performance was obtained [AT (SST):r = 0.98 (0.99), Root mean square error (RMSE) = 1.69 (1.88)] than the reconstructions by other authors for theregion using conventional methods. Finally, present-time and paleodata integration into the ANNs forprediction allowed improving the number of steps into the future with greater certainty, precision, performance[AT (SST): r = 0.94 (0.94), RMSE = 2.83 (5.98), Mean Absolute Percent Error = 2.98 (3.97)], and predictionaccuracy (>90%), with values within the range of variation of the instrumental data. However, we recognize thatincreasing data in the ANN training base and higher resolution paleodata from laminated sediments are stillnecessary, so it is crucial to maintain the regional observatories. 

Palabras clave: artificial neural network; climate reconstructions and predictions; paleodata

Para obtener una copia del documento contacta la personal de la biblioteca a través del correo bibliocicimar{a}ipn.mx

Regresar