28 29
ISSN 2477-9105
Número 32 Vol.1 (2024)
IBOMAS and Cressman Schemes for Temperaturatures Times in Chimborazo Province
¹ Escuela Superior Politécnica de Chimborazo, Facultad de Ciencias, Riobamba,Ecuador.
² Afiliación Independiente.
* cristina.ramos@espoch.edu.ec
Palabras claves: Datos meteorológicos, Datos climácos, Esmación de datos faltantes, Método de Cressman, Esquema de
Análisis de Mapas Objevo Interacvo de Barnes, Radio de inuencia.
Keywords: Weather data, Climate data, Missing data esmaon, Cressman method, Interacve Barnes Objecve Map
Analysis Scheme, Radius of inuence.
RESUMEN
ABSTRACT
ESQUEMAS IBOMAS Y CRESSMAN PARA SERIES TEMPORALES
DE TEMPERATURA EN LA PROVINCIA DE CHIMBORAZO
Fecha de recepción: Fecha de aceptación: Fecha de Publicación:
² Daniel Alejandro Reinoso Salas
² Jefferson Andrés Camacho Heredia
iD
iD
iD
iD
iD
iD
28 29
Esquemas IBOMAS y Cressman para series temporales de
temperatura en la provincia de Chimborazo
Ramos, Pozo, Balvoa, Miranda, Reinoso, Camacho.
errors, and outliers are problems encountered
structure can be reconstructed in two or three
I. INTRODUCTION
ensuring the integrity and reliability of the results.
database as part of the repository in ESPOCH
paradigm programming language.
II. MATERIALS AND METHODS
Figure. 1. Geographic Posion of Alao Staon
30 31
ISSN 2477-9105
Número 32 Vol.1 (2024)
Interacve Barnes Objecve Map Analysis
Scheme (IBOMAS)
Where rm is the distance between the grid point
i f(xi), and 0 < λ < 1 is the
K and K0 are
Where fi
case, Cressman's method was used, with one
that depends on its distance to the unsampled
Where di is the Euclidean distance between the
sampled and unsampled points, and R is the
Evaluaon of the performance of the missing
data esmaon methods
the performance and compare the models, thus
R²) and accuracy
between f(xmg0(xm) indicates the adjustment of
g0(x), a weighted
g0(xm) can be interpolated
And
And
w e=
m
r2
m
k(1)
w e=
m
r2
0
m
k(2)
λ
g
( (([ ]) ))
m=1 m=1
m=1
m m 0
m
m mm
M M
M
W W
W
x xx
ff f g
(i):= +
m=1 m
MW
(4)
Z(x,y) = fi (x,y)wi
wi
,
M
i=1
={,
>
wi
R
R
R - difor di
R + difor di
2
2
2
2
passes.
Cressman method
determined by the Euclidean distance.
= +
(xi xi-1) (yi yi-1)
di,
2 2
30 31
Esquemas IBOMAS y Cressman para series temporales de
temperatura en la provincia de Chimborazo
Ramos, Pozo, Balvoa, Miranda, Reinoso, Camacho.
Euclidean metric (Fig. 2).
Figure 2 illustrates the comparison between the
well with those of the original data, although
points. Despite these small discrepancies, the
III. RESULTS
(7)
(9)
(10)
(11)
tool for measuring the performance of a model
Where M represents the total amount of data,
zi
i,
temperature,
(Zi - Zi)
M
i=1
=MSE 1
n
2
(Zi - Zi)
M
i=1
=RSS
2
(Zi - Zi)
M
i=1
=MAE 1
n
(Zi - Zi)
(Zi - Zi)
M
M
i=1
i=1
=R2
2
2
,
Number of correct predicons
Total number of predicons
accuracy =
missing or unsampled data points within a region
original dataset.
data (depicted by the solid blue lines) with the
followed the trend of the original data, thereby
Figure. 2. Cressman method.
Figure. 3. First pass - IBOMAS
32 33
ISSN 2477-9105
Número 32 Vol.1 (2024)
against the original data. Each method followed
the general trend of the original data, which
passes) demonstrated a closer alignment with
accuracy compared to Cressman’s method.
Cressman´s method for accurate temperature
data. All the methods generally followed the
trend of the original temperature data, which
second pass, aligns more closely with the original
original data (solid blue lines) compared with the
from approximately fourth to eighteen, display
data successfully mirrored the trend of the
original data and accurately captured periodic
In the context of the Cressman method, the
with Cressman, discrepancies emerged with
Figure. 4. Second pass - IBOMAS.
Figure. 5. Esmaon Methods R=10
Figure. 6. Esmaon Methods R=30
32 33
Esquemas IBOMAS y Cressman para series temporales de
temperatura en la provincia de Chimborazo
Ramos, Pozo, Balvoa, Miranda, Reinoso, Camacho.
Figure. 7. Esmaon Methods R=60
Table 1. Error analyses
least accuracy.
Figure 7 presents a comparison of the
against the original temperature data. All
Although all methods capture these trends,
Cressman's method shows less precision than
method being the least precise.
Error analysis
importance to accurately represent atmospheric
how each method addresses data sparsity,
researchers can gain a deeper understanding of
its performance in capturing the atmospheric
enhancing the accuracy and reliability of
R=10 R=30 R=60
Error
metric IBOMAS 1 IBOMAS 2 Cressman IBOMAS 1 IBOMAS 2 Cressman IBOMAS 1 IBOMAS 2 Cressman
0.129 0.129 0.129
RSS
0.971
R^2 0.991 0.917 0.991 0.991
Accuracy 0.991 0.917 0.991 0.991
34 35
ISSN 2477-9105
Número 32 Vol.1 (2024)
R², and accuracy) tended to be close to zero or
in terms of accuracy and explanatory capacity
Cressman method. Accuracy was assessed using
99.9% in the second, whereas the Cressman
errors and yield more reliable results, thereby
contexts.
determined based on the distance between the
points represent the data points as well as possible
compute the weights for neighboring data points
IV. DISCUSSION
Cressman method (2,4,11).
employing a scale parameter to calculate the
weights for neighboring data points based on
adjusts the scale parameter using the disparity
Cressman method, there was an 11% disparity
phenomenon.
34 35
Esquemas IBOMAS y Cressman para series temporales de
temperatura en la provincia de Chimborazo
Ramos, Pozo, Balvoa, Miranda, Reinoso, Camacho.
second pass showed superior accuracy and
Cressman's method.
V. CONCLUSIONS
VII. REFERENCES
VI. ACKNOWLEDGEMENTS
Future studies should consider the
36 37
36 37
ISSN 2477-9105
Número 32 Vol.1 (2024)
