KR102685689B1 - Apparatus and method for detailed estimation of relative humidity in consideration of land cover characteristics and influence of wind - Google Patents

Abstract

The present invention relates to a device and method for estimating relative humidity, and in particular, to a device and method for detailed relative humidity estimation taking into account ground cover characteristics and the influence of wind. The detailed relative humidity estimation device and method considering the ground cover characteristics and the influence of wind according to the present invention empirically quantifies the variation of water vapor pressure and the influence of wind speed according to the ground cover characteristics of the target area, resulting in relative humidity estimation error compared to the existing one. can be reduced.

Description

Apparatus and method for detailed estimation of relative humidity in consideration of land cover characteristics and influence of wind}

The present invention relates to a device and method for estimating relative humidity, and in particular, to a device and method for detailed relative humidity estimation taking into account ground cover characteristics and the influence of wind.

Relative humidity is one of several meteorological factors that affect crops. High humidity conditions, which are one of the direct environmental factors that have a negative impact on crop growth, are mainly dealt with in soil moisture, but atmospheric relative humidity is related to the occurrence of pests and diseases along with temperature, so appropriate response is necessary.

Relative humidity refers to the ratio of actual water vapor to the saturated water vapor and can be expressed as a ratio of vapor pressure, density, or mass. Water vapor pressure changes directly depending on the density or amount of water vapor. It decreases at a certain rate as the altitude in the atmosphere increases, and evaporation from the soil and water surface and transpiration through the stomata of plants supply water vapor into the atmosphere. In addition, evapotranspiration is influenced by various environmental and plant physiological factors, such as meteorological factors such as solar radiation energy, temperature, atmospheric pressure, wind speed, and humidity, characteristics of the evaporation surface, available moisture in the soil, and plant leaves and stomata.

To estimate relative humidity, methods have been used to obtain the vertical distribution of the troposphere from remote sensing information on water vapor or to statistically detail it from satellite data or point observation data. In Korea, relative humidity is basically included in the numerical model data provided by the Korea Meteorological Administration, but it has a grid size of 5km x 5km, so the resolution is not sufficient for complex terrain. In particular, due to Korea's topographical characteristics, there are many mountainous agricultural areas, so there are variations in weather distribution in complex terrain. The Rural Development Administration operates an agricultural meteorological disaster early warning system to generate weather forecasts and crop risk forecasts due to abnormal weather at the farm level and provide them to farmers one-on-one. To solve the problem of resolution of the Korea Meteorological Administration's forecast data, a digital elevation model (DEM) is used. , digital elevation model), and is undergoing a detailed process using high-resolution (30m×30m) geographical spatial information. The Korea Meteorological Administration's short- and medium-term forecast models, the Regional Date Assimilation and Prediction System (RDAPS) and Local Data Assimilation and Prediction System (LDAPS), and the Korea Local Analysis and Prediction Model (KLAPS), which are the short-term forecasts of the Korea Meteorological Administration. Numerical forecast model data such as System), or observation data from synoptic and disaster prevention weather observation networks are used as background data in the process of detailing weather information for the early warning system (early warning system for weather risk management in agricultural sector). In the process of producing farm-customized weather information, major weather simulation technologies such as detailed temperature, precipitation, solar radiation, and sunshine hours for agricultural and mountainous areas have been verified and improved over a long period of time. On the other hand, relative humidity does not take into account other factors other than basic altitude correction.

In the current early warning system, the weather information from the Korea Meteorological Administration for the target area is detailed at the 30m grid resolution level, converted into secondary and tertiary agricultural weather information, and weather disasters for each crop are monitored through real-time forecast information (updated every 3 hours). Because predictions must be made, calculations are performed on the grid itself or between spatially matching grids using relatively simple mathematical empirical quantifications. Therefore, it is necessary to devise a method to simulate farm-scale relative humidity in a manner consistent with the early warning system process.

Republic of Korea Open Utility Model No. 20-1998-0050664 (published on October 7, 1998)

The purpose of the present invention is to provide a detailed estimation device and method for relative humidity that takes into account ground cover characteristics and the influence of wind, which can reduce relative humidity estimation errors compared to existing ones.

The object of the present invention is not limited to the object mentioned above, and other objects and advantages of the present invention that are not mentioned can be understood by the following description and will be more clearly understood by the examples of the present invention. Additionally, it will be readily apparent that the objects and advantages of the present invention can be realized by the means and combinations thereof indicated in the patent claims.

The detailed relative humidity estimation device considering the ground cover characteristics and the influence of wind according to the present invention includes a water vapor pressure variation calculation unit that calculates the water vapor pressure variation, a water vapor pressure correction unit that corrects the water vapor pressure based on the water vapor pressure variation calculated in the water vapor pressure variation calculation unit, and It includes a relative humidity estimation unit that estimates relative humidity using the water vapor pressure corrected in the water vapor pressure correction unit.

In addition, the detailed relative humidity estimation method considering the ground cover characteristics and the influence of wind according to the present invention includes the steps of a water vapor pressure variation calculation unit calculating the water vapor pressure variation, and a water vapor pressure correction unit calculating the water vapor pressure based on the water vapor pressure variation calculated by the water vapor pressure variation calculation unit. It includes a step of correcting, and a step of the relative humidity estimating unit estimating the relative humidity using the water vapor pressure corrected by the water vapor pressure correction unit.

Water vapor pressure variation calculated in the water vapor pressure variation calculation unit ( )Is,

and

ego,

and

d is 365 days, i is the target point, V _i is the wind speed at target point i, and V _o is the background wind speed.

The detailed relative humidity estimation device and method considering the ground cover characteristics and the influence of wind according to the present invention empirically quantifies the variation of water vapor pressure and the influence of wind speed according to the ground cover characteristics of the target area, resulting in relative humidity estimation error compared to the existing one. can be reduced.

In addition to the above-described effects, specific effects of the present invention are described below while explaining specific details for carrying out the invention.

Figure 1 is a conceptual diagram of a detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention.
Figure 2 is a diagram showing the meteorological observation points for analysis and verification distributed in the target area, the KLAPS grid division, and the ground cover characteristics in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention.
Figure 3 is a graph of water vapor pressure variation compared to the reference point under wind-free conditions for each analysis point in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention.
Figure 4 shows the annual water vapor pressure deviation for each analysis point (A to I) compared to the reference point in the night time zone (1900 to 0600 LST) limited to no-wind conditions in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention. It is a distribution graph.
Figure 5 shows the annual water vapor pressure deviation for each analysis point (A to I) compared to the reference point during the daytime (0700 to 1800 LST) limited to no-wind conditions in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention. It is a distribution graph.
Figure 6 shows the maximum water vapor pressure variation for the dense forest area according to the wind speed (V0) at the reference point during the night time (1700-0600LST) in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention. This is a drawing showing the changes.
Figure 7 shows the wind speed deviation and water vapor pressure deviation at analysis points (A to I) compared to the reference point for night (left) and daytime (right) in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention. This is a comparison graph.
Figure 8 is a time-dependent comparison graph of the shear coefficient calculated from the wind speed for each layer in 2016 by the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention.
Figure 9 is a flowchart of a detailed estimation method for relative humidity considering the ground cover characteristics and wind influence according to the present invention.

The above-described objects, features, and advantages will be described in detail later with reference to the attached drawings, so that those skilled in the art will be able to easily implement the technical idea of the present invention. In describing the present invention, if it is determined that a detailed description of known technologies related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description will be omitted. Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings. In the drawings, identical reference numerals are used to indicate identical or similar components.

Figure 1 is a conceptual diagram of a detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention.

As shown in FIG. 1, the detailed relative humidity estimation device considering the ground cover characteristics and the influence of wind according to the present invention includes a water vapor pressure variation calculation unit 100 that calculates water vapor pressure variation, and a water vapor pressure variable that corrects the water vapor pressure based on the water vapor pressure variation. It includes a correction unit 200 and a relative humidity estimation unit 300 that estimates relative humidity using the corrected water vapor pressure.

The vapor pressure variation calculation unit 100 calculates the vapor pressure variation to correct the corrected atmospheric pressure.

Figure 2 is a diagram showing the meteorological observation points for analysis and verification distributed in the target area, the KLAPS grid division, and the ground cover characteristics in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention.

In the present invention, Agyang-myeon, Hadong-gun, Gyeongsangnam-do, forms a catchment, and is a typical agricultural and mountainous area, with rice paddies and fields in the lowlands, orchards at the foot and slopes of the mountains, and small villages consisting of farm houses near the roads. Here, referring to Figure 2, temperature, humidity, and wind speed observation data were collected from a number of simple weather observation equipment (STA Corporation Co., Ltd., South Korea) in 2012-2013 and 2015. Among them, the period from August 15, 2012, when the most spatially high-density data was collected to August 18, 2013, was used as a period to derive the variation characteristics of water vapor pressure by surface characteristics (asterisk in Figure 2).

Table 1 shows the weather observation points for analysis and verification of the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention and the ground cover information of the corresponding KLAPS grid.

Mdeling★ Verification● Site DEM Landcover Site DEM Landcover Num Landcover A 391 dense forest K 187 farmland One dense forest B 285 farmland L 181 farmland C 115 forest+town M 225 forest+grass 2 farmland D 54 farmland N 201 forest+grass 3 forest+grass E 48 town+farmland O 195 farmland 4 forest+grass F 90 town (parking) P 10 farmland 5 town+farmland G 200 forest+orchard Q 35 forest+grass H 326 forest+grass R 5 farmland (grass) I 8 paddy 　 J 17 town+farmland 　 　 　 　 　

Referring to Table 1, there are a total of 10 data utilization points for the relevant period, which are points where weather is observed 1.5 to 2 meters above the ground surface in the area where the equipment is installed and the missing period is less than 2 weeks. Ground surface characteristics are classified from equipment installed next to a deep forest (dense forest), inside an orchard next to a forest, on a lawn near a forest, around a village building next to a forest, in a rice field, on a farm (next to an orchard or on a field), around a farm and a village, and next to a building parking lot. did. Using data from these 10 locations, the variation of water vapor pressure by surface characteristics was quantified according to wind speed, and data from 7 meteorological observation points other than the 10 locations were used to verify the derived model. The verification period was from January 2 to October 3, 2015, a period of simultaneous observation of 7 locations. In addition, after collecting KLAPS humidity, temperature, and wind speed numerical data from the Korea Meteorological Administration for the relevant period, the values of the grid containing the verification points were extracted and used as background data for humidity estimation.

Table 2 shows the ratio of ground cover area in the KLAPS grid division, which includes the verification points of the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention, and the ratio of the entire ground cover area in South Korea (2010s). In Table 2, the water area is excluded.

Landcover Whole country
(%) KLAPS Number One 2 3 4 5 urban, town 5.6 3.1 5.5 1.5 2.1 10.9 farmland 25.7 6.0 32.8 15.2 19.5 28.1 forest 62.4 88.0 58.7 82.8 71.0 53.0 grassland 1.9 1.5 0.9 0.4 0.9 1.9 wetland 1.9 0.5 0.4 0.0 1.9 2.0 arid area 1.5 0.9 1.7 0.0 4.6 4.2 nodata 1.0 0.0 0.0 0.0 0.0 0.0 Total 100 100 100 100 100 100

KLAPS's surface cover information defined representative surface characteristics according to the area ratio of forest, farmland, grassland, wetland, and bare land within the KLAPS 5km × 5km grid division from the 30m grid surface cover map (National Geographic Information Institute) as of 2010 (Table 1 , Table 2). When the area ratio of forest in the grid was more than 80% and other ground cover areas such as agricultural land were less than 10%, it was defined as dense forest. However, even if the forest is more than 80%, if the agricultural land or grassland area is more than 10%, it was judged as forest + grass. Among the 10 analysis points, there were no cases where farmland (field) and grassland were separated, so farmland (field) and grassland were considered the same. Put it into a category. The proportion of farmland in the entire South Korean region was 25.7%, and if the proportion of farmland was greater than 25.7% within a 5km × 5km grid, it was classified as farmland. In addition, compared to the area ratio by land cover across the country, if the ratio of built-up area and agricultural land area was greater, it was considered town+farmland (Table 2).

In the present invention, among the analysis points in the target area, point J, which was installed in the agricultural land (field) in the village, was considered as the standard for water vapor pressure variation. Saturated water vapor pressure and water vapor pressure were calculated using the hourly temperature and humidity observed at the analysis points (A to J). Using point J as the standard, the water vapor pressure deviation (target point) was calculated for each of the remaining nine target points (A to I). value - reference point value) was calculated. This deviation value contains variation in water vapor pressure due to the surface cover characteristics of other points compared to the surface cover characteristics of point J, but will also reflect components caused by factors other than the surface cover (wind, altitude deviation, etc.). Since all 10 analysis points are included in a single catchment area (Agyang-myeon), the distribution of water vapor pressure according to altitude within one air mass is assumed to be explained according to the same lapse rate, and the altitude deviation above sea level is calculated for each of the 9 points based on point J. After obtaining it, it was corrected according to the method of Kim et al., (2020). In addition, water vapor pressure deviation was analyzed separately into daytime (07-18:00) and night time (19-6:00).

The maximum degree to which the water vapor pressure is higher than point J (town+farmland) is considered as the parameter M, and the case where the wind speed is 0m/s for both J and the target point (A~I) during the period from August 15, 2012 to August 18, 2013 After selection, the highest and lowest absolute values were found and the difference between the two was assigned as the M value. If the maximum value of the water vapor pressure deviation is a positive value and the minimum value is a negative value, which are opposite to each other, it is interpreted that factors other than ground cover and wind speed that cause the water vapor pressure variation play a large role, and the M value is calculated by changing the minimum value to a positive value. By doing so, the variation correction value was not excessively applied in the quantitative equation. However, in areas with clear seasonal influences, such as Korea, water vapor pressure is generally high in summer and low in winter, resulting in dryness. Therefore, in the analysis data, if precipitation occurring in late fall and winter resulted in a higher water vapor pressure deviation than in summer, which was an outlier and affected the derivation of M value, the case was deleted (2012.10.22, 2012.12.14). In addition, the day and night parameter M values were allowed to fluctuate seasonally, and the year-round change cosine curve based on the temperature suggested by Kim et al., 2016 was applied.

In Equation 1, d means 365 days in the year, with January 1st set as 1 and December 31st set as 365.

To express the wind offset of night or daytime M _d values, both the wind speed at the reference point and the wind speed at the point representing the water vapor pressure anomaly must be taken into account. Therefore, after extracting the case where one of the target point (one of A~I) and the reference point J was 0 m/s, a model equation was created to explain the water vapor pressure variation according to the wind speed intensity of the other point. First, when the wind speed V _i at the target point i is 0 m/s, the water vapor pressure deviation value at the analysis point is set as the dependent variable, and the wind speed V ₀ at the reference point J is set as the independent variable. As V ₀ increases, the water vapor pressure deviation changes. An empirical quantitative equation in the form of inverse proportionality was derived as shown in Equation 2 below. At this time, out of concern that if all analysis points were used, the composition ratio of analysis points (A to I) included in each wind speed section would vary, only the observed values from dense forest points expected to have the largest water vapor pressure deviation among surface characteristics were used. . Since the M _d value changes depending on the indicator characteristics, the k value was derived as a constant and then applied to all regions.

When applying the model in Equation 2, △VP _M is the maximum water vapor pressure variation that can occur under wind-free conditions at target point i, and the value of V ₀ corresponds to the background wind speed, which is the input data for the model. In addition, if the wind speed V _i at point i is known, the water vapor pressure variation estimation process does not end in Equation 2, but △VP _M is considered to be further attenuated according to V _i (Equation 3). V _i was recognized only up to the time average value of 2 m/s, and beyond that, the water vapor pressure variation △VP _i at i was set to 0.

The standard for the temporal wind speed of 2m/s is that the effect of cold air accumulation at night in the target area (Akyang-myeon) is almost canceled when the wind speed reaches 2m/s, and the warming effect of the temperature near the surface due to the influence of solar radiation during the day is 2m/s. It was set based on the fact that it drops to 10% of the no-wind level when it reaches , and the relationship between the water vapor pressure deviation across the entire analysis point and the wind speed at that point was confirmed.

The water vapor pressure correction unit 200 corrects the water vapor pressure based on the water vapor pressure variation calculated by the water vapor pressure variation calculation unit 100. This can be corrected by the water vapor pressure correction unit 200 applying the water vapor pressure variation calculated by the water vapor pressure variation calculation unit 100 to the water vapor pressure.

The relative humidity estimation unit 300 estimates the relative humidity based on the water vapor pressure corrected by the water vapor pressure correction unit 200. This allows the relative humidity estimation unit 300 to estimate the relative humidity by dividing the water vapor pressure corrected by the water vapor pressure correction unit 200 by the saturated water vapor pressure.

Next, the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention was verified.

During the verification period (January 2015 - October 2015), the KLAPS numerical data of 8 times a day at 3-hour intervals (0000, 0300, 0600, ?? 2100LST) were used as background data to compare the relative values of 7 verification points (K to Q) in the target area. Humidity was estimated. The M value was designated according to the indicator characteristics of each verification point and the indicator characteristics of the KLAPS grid (1 to 5) containing the corresponding point. If a random point i is a farm and most of the KLAPS grid division containing point i is a forest, the background humidity of KLAPS can be considered to indicate a forest area. Therefore, when estimating relative humidity, the water vapor pressure variation for forest areas should be removed and the water vapor pressure variation for farms should be added. Therefore, the maximum value of water vapor pressure variation M _d during the day or night on a specific day was obtained by subtracting the M ₀ value for the KLAPS section, which serves as background data, from the M _i value for the simulated point i. M _i and M ₀ are values to which the seasonal variation of Equation 1 was applied, and the calculated M _d was used in the calculation of Equation 2. Since the KLAPS wind speed corresponding to the V ₀ value is the wind speed value of 10 m above the ground, it was converted to the wind speed of 2 m above the ground. The main feature of the wind profile power law's shear modulus α value is that it varies depending on the surface roughness (Plate, 1971), but the only wind speed at each level observed near the target area is at point R installed in grassland and agricultural land. Since the α value has more pronounced daily fluctuations as the surface impact increases, an empirical α value was derived and used using the hourly average wind speeds at 1m, 3m, and 6m above the ground for one year in 2016, and the The difference from the actual wind speed was left in the area of error. On the other hand, the wind speed (V _i ) corresponding to verification point 7 did not have an estimated value, so observed values were inevitably used instead.

The water vapor pressure variation calculated through Equations 1 to 3 was added to the water vapor pressure calculated using KLAPS, and this was converted to relative humidity by dividing it by the KLAPS-based saturated water vapor pressure. The performance evaluation of the model compared KLAPS's humidity itself (old) with the technique for correcting water vapor pressure by ground cover (new model). The estimation errors of the two models were confirmed by ME (mean error), RMSE (root mean square error), and MAE (mean absolute error).

Figure 3 is a graph of water vapor pressure variation compared to the reference point under wind-free conditions for each analysis point in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention.

Table 3 is a table derived from the detailed relative humidity estimation device considering the ground cover characteristics and the influence of wind according to the present invention, the maximum moisture vapor pressure variation parameter M under windless conditions for each ground cover characteristic.

　 Town+paddy
(F) town+farmland
(E) farmland
(B&D) forest+town
(C) forest+grass
(H) dense forest
(A) Paddy
(I) night (19-06) 0.46 -0.13 0.62 1.26 0.82 1.82 -0.38 day (07-18) 0.66 -0.05 1.10 1.55 1.56 2.26 0.37

The variation in water vapor pressure compared to the reference point identified for wind-free conditions was high in the order of forest, agricultural land, and village. This shows that plants emit H2O during gas exchange through stomata, making the vegetation more humid. The water vapor pressure variation at points A to I compared in Figure 3 is a result of how much higher the water vapor pressure is at the analysis point compared to the reference point when the wind is 0 m/s at both the reference point and the target point during the analysis period. Even in windless conditions, the water vapor pressure deviation value occurred in both positive and negative values. Among them, the maximum positive value and the minimum negative value (maximum absolute value) were determined, and the opposite tendency value was calculated from the dominant side. The remaining value after canceling out was obtained. Overall, the variation in water vapor pressure during the day appears to be higher than that at night, suggesting that transpiration has increased due to the active photosynthetic process. However, since the number of cases where the wind-free condition is met during the day is less than that at night, the potential increase in water vapor pressure that may occur due to plant photosynthesis may actually be greater than this. In Figure 3, point E is a place where farmland and village are combined, and is relatively similar to the environment of the reference point (installed on grassland on one side of the playground of a closed elementary school), and the values of the remaining points are the parameter M values according to the characteristics of each indicator. (Table 3). Figure 4 shows the analysis point (A) compared to the reference point in the night time zone (1900-0600 LST) limited to no-wind conditions in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention. ~I) This is a graph of distribution of water vapor pressure deviation throughout the year. Figure 4 is the result of applying the M value for each point to Equation 1. Figure 5 shows the annual water vapor pressure deviation for each analysis point (A to I) compared to the reference point during the daytime (0700 to 1800 LST) limited to no-wind conditions in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention. It is a distribution graph. In Figure 5, the dotted line is the result of applying the M value for each point to Equation 1.

Figures 4 and 5 show all year-round water vapor pressure deviations in wind-free conditions for each analysis point at night and daytime, respectively, as a scatter plot, and the dotted line in the graph represents the water vapor pressure variation values in Figure 3 on July 24 (day of year = 205) and decreases as winter approaches according to Equation 1. The dotted lines in Figures 4 and 5 are considered to simulate close to the average value of all water vapor pressure deviations, but there was a large variation in water vapor pressure deviations even under windless conditions. The amount of transpiration of plants varies depending on the amount of moisture in the surface and the plant rhizosphere, and is also affected by physiological conditions such as photosynthesis and root activity, so even under the same wind-free conditions, differences occur depending on the season, time, and weather conditions (light, precipitation, etc.). , in this invention, the average value is applied empirically.

Referring to Figures 3 to 5, point A in the mountains where the surrounding area is comprised of forests showed the highest water vapor pressure relative to other points. Point B is near a pear orchard built on a mountain slope. The site is located on wild grassland and is not managed like a lawn. B had a lower water vapor pressure deviation than A, but it was higher than the reference point, showing a positive value. Point C is located at the top of the Hanok village built on a mountain ridge. It is surrounded by a forest, but the observation point is above the Hanok yard. Referring to Figures 4 and 5, the water vapor pressure deviation was smaller than that at point A and the variation was smaller than that at point B. In particular, the water vapor pressure deviation in the negative direction at C is relatively small compared to other points, and the size of the deviation is also insignificant, so the maximum quantified water vapor pressure variation (M value) was calculated to be higher than that at B. Point D is an agricultural land consisting of orchards and fields, but is exposed to the fields rather than the orchards, and the water vapor pressure deviation was slightly smaller than that of point C. Referring to Figure 3, since point E is a virtually similar environment to the reference point, there was little variation in water vapor pressure deviation. However, referring to Figures 4 and 5, in cases of actual deviation values, (+) and (-) trends were observed. All of this happened. Referring to Figures 3 to 5, point F, installed between the parking lot of the student training center building and the rice field next to it, also showed less bias in the (+) direction of the water vapor pressure deviation than other points. Referring to Figure 4, point G is inside an orchard built on a mountain slope, and a quantitative M value similar to point A was obtained at night, but only some of the actual water vapor pressure deviation values showed a large variation in the (+) direction, and many of them showed a large variation in the (+) direction. It stayed between -1 and 1 hPa. In addition, referring to Figure 5, the variation in water vapor pressure deviation during the day was not clear, so the M value was calculated as low as shown in Figure 3. H was installed on a lawn in the temple grounds located in a deep forest and the grass was always maintained, so the water vapor pressure deviation from the reference point was smaller than A due to the influence of the nearby forest but also the grassland cover at the observation site, and B~D It was at a similar level, but the variation was diverse. In the daytime, the M value was calculated at the same level as point C due to the large water vapor pressure deviation in the minor (+) direction. The M at night was calculated to be similar to, but slightly higher than, the M values at point C and agricultural areas B and D. While point I on a flat rice field is a completely open area, in the case of the reference point, since there are structures and vegetation around the reference point, the water vapor pressure deviation from the reference point appears to have a slight (-) tendency, especially in summer.

Finally, the remaining analysis points except G were used as quantitative M values representing the ground cover characteristics of the location. In the case of G, according to the rules for calculating the M value, day and night were calculated with opposite tendencies compared to other points. When looking at the overall water vapor pressure deviation distribution, it is similar to B to D, and in some cases, the water vapor pressure deviation in the high (+) direction is Because it does not exceed the range of A, which is a dense forest, G was excluded from the final M value designation. Points B and D were classified as the same farmland area, and the averaged M value was used. In addition, although point H was on grassland located within a deep forest, it was decided to also include cases where agricultural land was created within a deep forest. Since point C is a point where the forest is quite close, it was limited to cases where the forest dominates the artificial structures (buildings, yards, etc.). The dotted lines in Figures 4 and 5 are M _d , which is the result of setting the M value at each point as the highest value of the year on July 24 and applying the annual variation according to Equation 1.

Figure 6 shows the maximum water vapor pressure for the dense forest area according to the wind speed (V ₀ ) at the reference point during the night time (1700-0600LST) in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention. This is a diagram showing the change in mutation. In Figure 6, the maximum water vapor pressure variation is the highest value (MAX) and lowest value (MIN) found among the water vapor pressure deviation values observed when the observed wind speed at the forest point is 0 m/s.

The calculated M _d value was offset by the wind, as shown in the reduction curve according to the wind speed (V0) at the reference point in Figure 6. The two symbols (ⅹ, +) shown in FIG. 6 are the highest values of water vapor pressure deviation values for each group when the data is classified according to the wind speed at the reference point limited to the case where the wind speed at dense forest point A was 0 m/s during the night. (MAX) and minimum value (MIN). The label next to the symbol is the number of data in the group, and as V ₀ increases, the number of conditions in which A was windless greatly decreases. Meanwhile, regardless of the strength of V ₀ , water vapor pressure variation in the positive direction was always superior to water vapor pressure variation in the negative direction. The dotted line 'EST' is the result of deriving the water vapor pressure variation attenuation curve according to each wind speed in the form of Equation 2 with two symbol data, and MAX(=3.6073) and MIN(=) when V ₀ in Md is 0 m/s When the value (-1.7327) was entered, the k values were 5.8735 and 25.3843, respectively. This k value was commonly used as a parameter in Equation 2 for all regions and times, but the number of cases where the wind speed (V _i ) at the target point was 0 m/s was small for each ground cover, especially during the daytime, so it was used to meet the conditions. This is because it is difficult to obtain controlled data by intensity of V ₀ .

Figure 7 shows the wind speed deviation and water vapor pressure deviation at analysis points (A to I) compared to the reference point for night (left) and daytime (right) in the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention. This is a comparison graph. In Figure 7, if the deviation value is a positive value, it means that the value of the analysis point is greater than the reference point.

The water vapor pressure variation shown in FIG. 6 is based on Equation 2. When the deviation between the wind speed V _i and V ₀ at the target point is a (-) value, the wind at the target point is calm and the water vapor pressure variation can occur at its maximum size. It can be seen as expressing . In addition, Figure 7 shows the water vapor pressure deviation for all analysis points (A ~ I) when the wind speed (Vi) at the target point is stronger than V ₀ , that is, when the deviation between V _i and V ₀ is a (+) value. , As V _i became stronger, the variation of water vapor pressure rapidly decreased. No matter what the V ₀ wind speed is, when V _i exceeds 1~2 m/s, most of the variations in water vapor pressure stay within ±1 hPa, and when it exceeds about 2 m/s, the water vapor pressure deviation values converge to 0. When detailing relative humidity and water vapor pressure, the potential maximum water vapor pressure variation derived through Equation 2 by the wind speed value (=V ₀ ) from the background standard meteorological information (air temperature, humidity, wind speed, etc.) is estimated as Equation 3. It is corrected again based on the wind speed of the point (=V _i ), and it is judged that there is no problem in processing the water vapor pressure variation to 0 when the wind speed is 2 m/s or more.

Figure 8 is a time-dependent comparison graph of the shear coefficient calculated from the wind speed for each layer in 2016 by the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention. The upper graph of Figure 8 shows the average mesh shear modulus, and the lower graph shows the standard deviation. The shear modulus was calculated in three types: wind speed from 1m to 3m above the ground, from 1m to 6m, and from 3m to 6m.

The verification of the model according to the present invention, that is, the detailed relative humidity estimation device considering the ground cover characteristics and the influence of wind, can be done by using the relative humidity of KLAPS as is and evaluating the effect of ground cover and wind speed through the process of Equations 1 to 3. This was done by comparing cases that underwent a correction process. First, referring to the upper graph of FIG. 8, the shear modulus α value at point R used in the process of converting 10 m wind speed to 2 m wind speed using power law for application of KLAPS of the present invention is the α value calculated at wind speed close to the surface. The more days there were, the larger the daily fluctuations were. The α value calculated between wind speeds of 1m and 3m above the ground showed the most noticeable daily variation, while the α value between wind speeds of 3m and 6m above the ground showed little daily variation. Between 1m and 3m and 1m and 6m above the ground, the α value remained high during the night before sunrise, then decreased sharply after sunrise, remained low during the day, and showed a pattern of rapidly rising again from around sunset. The standard deviation of the α value was the lowest between 1m and 6m wind speeds, and the shear coefficient at this time was used to adjust the KLAPS wind speed.

Table 4 shows the results of applying the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention for each verification point (K to Q) compared with the relative humidity of KLAPS. In the detailed relative humidity estimation device considering the ground cover characteristics and wind influence according to the present invention, the dates/times when water vapor pressure correction was made by more than 0.1 hPa were counted.

　 　 KLAPS Model Site Count M.E. RMSE M.A.E. M.E. RMSE M.A.E. K 1,163 4.355 10.246 7.580 2.273 9.655 7.162 L 1,029 3.670 10.546 7.870 1.594 9.983 7.447 M 53 -1.222 6.962 5.365 -0.662 6.852 5.173 N - 0.001 9.110 6.660 - - - O 71 5.505 9.532 7.492 4.949 9.228 7.211 P 665 -9.851 11.931 10.218 -9.072 11.257 9.493 Q 1,019 -8.862 10.682 9.333 -7.931 9.883 8.486

Referring to Table 4, the relative humidity estimated from the temperature, humidity, and wind speed data of KLAPS and the wind speed data of the verification target point (K~Q) is RMSE 1.6~7.5%, MAE 3.6~ compared to the case where KLAPS data is used as is. It decreased by about 9.1%. In common, the bias of ME decreased at all verification points. At points K, L, and O, where the KLAPS relative humidity was excessive (ME > 0), the ME decreased after applying the model, while the relative humidity of KLAPS decreased Points M, P, and Q, which were smaller than the actual value (ME < 0), also improved convenience after applying the model. At point N, the ground cover characteristics of that point and the ground cover within the KLAPS area were classified as the same, so the corrected value in the model was 0. Points K and L are green tea fields on the mountainous slopes, and point O is a fern field on the mountain ridges. The entire KLAPS 5km×5km area covers most of the mountainous areas, so KLAPS humidity has a high value, but KLAPS humidity is high in partial agricultural areas within the area. It should be corrected even lower. On the other hand, point P, which is a plum orchard, has weather observation equipment installed in a general agricultural area consisting of fields and orchards. The KLAPS area that includes point P has a large agricultural land area and includes more villages and regions than other areas, so KLAPS The relative humidity value tended to be lower than the observed value at point P. This underestimation tendency was somewhat alleviated through model application. Point M is a fern field on a slope, but the immediate surrounding area is comprised of a forest. The ME of KLAPS humidity was slightly lower than 0, but the underestimation tendency was slightly improved by applying the model. Point Q is a grassland next to a forest near a village and a small vacant lot surrounded by trees. In KLAPS, the relative humidity was shown to be lower than actual, but the underestimation error was improved through the model. However, it was not possible to compare estimation errors between each point, because the number of cases (date/time) in which water vapor pressure corrections of 0.1 hPa or more occurred at each individual verification point were different (Table 4, Count).

As a result of applying a new model that applies ground cover and wind effects according to the present invention, the locally generated relative humidity estimation error remained at a level of less than 10%, but the convenience of the weather information estimation error makes it convenient when using the information for a long period of time. Considering this cumulative problem, even a slight reduction in error is considered a meaningful result.

Next, the detailed estimation method for relative humidity considering the ground cover characteristics and wind influence according to the present invention will be described with reference to the drawings. Among the content to be described later, content that overlaps with the description of the detailed relative humidity estimation device considering the ground cover characteristics and the influence of wind according to the present invention described above will be omitted or briefly described.

Figure 9 is a flowchart of a detailed estimation method for relative humidity considering the ground cover characteristics and wind influence according to the present invention.

As shown in FIG. 9, the detailed estimation method of relative humidity considering the ground cover characteristics and the influence of wind according to the present invention includes the step of calculating the water vapor pressure variation (S1), the step of correcting the water vapor pressure (S2), and the relative humidity It includes a step (S3) of estimating .

In the step (S1) of calculating the water vapor pressure variation, the water vapor pressure variation calculation unit calculates the water vapor pressure variation. This can be calculated as in Equation 3 described above.

In the step (S2) of correcting the water vapor pressure, the water vapor pressure correction unit corrects the water vapor pressure based on the water vapor pressure change calculated in the step (S1) of calculating the water vapor pressure change. As described above, this can be corrected by the water vapor pressure correction unit applying the water vapor pressure variation calculated by the water vapor pressure variation calculation unit 100 to the water vapor pressure.

In the step of estimating the relative humidity (S3), the relative humidity estimator estimates the relative humidity based on the water vapor pressure corrected in the step of correcting the water vapor pressure (S2). This means that the relative humidity estimation unit can estimate the relative humidity by dividing the water vapor pressure corrected by the water vapor pressure correction unit 200 by the saturated water vapor pressure.

As described above, the present invention has been described with reference to the illustrative drawings, but the present invention is not limited to the embodiments and drawings disclosed herein, and various modifications may be made by those skilled in the art within the scope of the technical idea of the present invention. It is obvious that transformation can occur. In addition, although the operational effects according to the configuration of the present invention were not explicitly described and explained while explaining the embodiments of the present invention above, it is natural that the predictable effects due to the configuration should also be recognized.

100: Water vapor pressure variation calculation unit 200: Water vapor pressure correction unit
300: Relative humidity estimation unit

Claims (4)

A vapor pressure variation calculation unit that calculates the water vapor pressure variation,
A vapor pressure correction unit that corrects the water vapor pressure based on the water vapor pressure variation calculated by the water vapor pressure variation calculation unit, and
It includes a relative humidity estimation unit that estimates relative humidity using the water vapor pressure corrected in the water vapor pressure correction unit,
The water vapor pressure variation calculated in the water vapor pressure variation calculation unit ( )Is,
and
remind ego,
remind and
where d is 365 days,
Where i is the target point,
The V _i is the wind speed at the target point i,
The V _o is a detailed relative humidity estimation device that takes into account the ground cover characteristics and wind influence, which is the background wind speed.
delete A step where the water vapor pressure variation calculation unit calculates the water vapor pressure variation,
A water vapor pressure correction unit correcting the water vapor pressure based on the water vapor pressure variation calculated by the water vapor pressure variation calculation unit, and
A relative humidity estimating unit includes a step of estimating the relative humidity using the water vapor pressure corrected by the water vapor pressure correction unit,
The water vapor pressure variation calculated in the water vapor pressure variation calculation unit ( )Is,
and
remind ego,
remind and
where d is 365 days,
Where i is the target point,
The V _i is the wind speed at the target point i,
The above V _o is a detailed estimation method of relative humidity that takes into account the ground cover characteristics, which is the background wind speed, and the influence of wind. delete

Images