1. INTRODUCTION
Extensive research sponsored by the DOD has been conducted over the past three years to determine the feasibility of using a deterministic limited-area high-resolution numerical model to generate local climate statistics around the globe. This method is being developed to address the limitations of both the use of long-term observational datasets and the probabilistic methods based on statistical relationships. The limitations of using observational data and probabilistic methods include the nonavailability of long-term point observations for large areas of the world, the problem of representativeness of point observations in complex terrain, the relocation of observing sites, and the change in land use around a site over time. The quality of observations also changes in time as new instrumentation is used. Additionally, there is the problem of determining the statistical relationships between the climate observed at a point and the climate at another point at some distance away with differing elevation, slope or aspect.
The numerical model approach to generating local climate statistics involves determining local climate from a set of long period mesoscale simulations. The goal is to simulate the actual climate statistics for a particular period of time over a specified region. The objective is addressed by executing the model for a long period of time in a data assimilation mode and allowing the model to fuse the available data into a 3-D climatological dataset.
Initial research results showed promise that this method would provide useful climate information (Zack et al. 1996, Van Knowe et al. 1999, Doggett et al. 1999). The impact of model resolution, data assimilation, convective parameterization scheme, and planetary boundary layer formulation used in the method is discussed in previous papers.
This paper examines the modeling issues involving the sensitivity of the quality of simulated climate statistics to observed surface and upper air data availability and "warm" versus "cold" model starts. Methods developed to utilize the knowledge gained from available observations, to remove model bias and establish a confidence index for the generated climate statistics, will also be discussed.
2.
METHODOLOGY
The methodology involves conducting research for several different climate regimes around the world at a model resolution of 10 km and 40 km. The basic methodology involves initialization of the model with large-scale gridded data fields. The model then uses the large-scale gridded data fields and lateral boundary conditions and periodically ingests the available observed data through one of several possible data assimilation techniques (Newtonian relaxation periodic reanalysis, etc.). Next, the model dynamically fuses the available observations with its knowledge of the surface characteristics of the earth and the basic principles of physics to generate estimates of local climate statistics at locations for which no observational data is available. This technique has been given the name CLImate statistics by a dynamical MODel (CLIMOD) and is described more fully by Zack et al. (1996), Van Knowe et al (1999) Van Knowe et al (2000) and Doggett (2000).
3. MODEL CONFIGURATION
The simulation model used in CLIMOD is based on a deterministic numerical atmospheric mesoscale model called the Mesoscale Atmospheric Simulation System (MASS) (Kaplan et al. 1982; MESO 1995; Manobianco et al. 1996). For this research, MASS version 5.11 was used. Hydrostatic formulations of MASS were utilized in the studies. Unless otherwise noted, all the experiments used a baseline configuration of the Blackadar PBL scheme, a modified Kuo cumulus parameterization and 24-hour discontinuous data assimilation. All the surface model values used in the experiments were reduced to the 2-meter model level. The specific features of the MASS model are described in the MASS Reference Manual (MESO 1995).
4. EXPERIMENTAL DESIGN
It was recognized through the initial experiments of this effort that quality of the high resolution climatologies for a given month was impacted by the availability of observed data and whether the simulation was started without a spin-up time (cold start) or was a continuation of a longer simulation (warm start). The approach used to evaluate the impact of the availability of observed data to the CLIMOD method was done through a series of experiments of systematic withholding of data for both cold and warm starts. Five years of simulation experiments were performed (1973, 74, 75, 77 and 79) for January, April, August and October over the region of the Korean Peninsula. Table 1 lists each type of experiment performed and includes a brief description.
Table 1. List of Data-denial
Experiments
|
Experiment Type |
Description |
|
A. Control -Warm
start |
All available surface and
upper observed data used with a warm
start |
|
B. Control - Cold
start |
All available surface and
upper observed data used with a cold
start |
|
C. Total Denial-
Warm start |
All surface and upper observed data withheld with a warm start |
|
D. Total Denial-
Cold start |
All surface and upper observed
data withheld with a cold start |
|
E. Upper Air Only -
Warm start |
All available upper observed
data used with a warm start |
|
F. Upper Air Only -
Cold start |
All available upper observed
data used with a cold start |
|
G. Surface Only -
Warm start |
All available surface observed
data used with a warm start |
|
H. 50 % Denial
Uniform - Warm start |
50 % of available surface and
upper air observed data withheld in a uniform manner with a warm start |
|
I. 50 % Denial Hole
- Warm start |
50 % of available surface and
upper air observed data withheld in a non-uniform manner creating a data void
region (hole) with a warm start |
The modeling steps used for the experiments are shown in Fig. 1. The warm start runs used at least 30 days of climatological spin-up time before the output data was used to compile statistics, where as a cold start used no spin-up time. Of course, even in the case of a cold start, the model would build up simulated antecedent conditions throughout the month. However, for this set of experiments, the smallest climatological unit was considered to be one month so the accumulation of antecedent conditions within the month were not explicitly evaluated. When data was assimilated, incremental analysis update was used as described in Van Knowe et al (1999).
CLIMOD STEPS
1. Establish grid domains:
- 40 km and 10 km
2. Ingest atmospheric data
- Large scale gridded fields - all experiments
- Observational - surface and upper air if needed for experiment type
3. Conduct dynamic initialization:
- Assure mass and thermal field balance
4. Initiate the model run
5. Assimilate Lateral Boundary data every 6 hours
6. Assimilate Observed Data every 12 hours if needed for experiment type
Fig. 1. General modeling design of the experiments.
5. RESULTS
5.1 Data Denial Experiments
The results indicated that the type of start (warm or cold) and the amount of observed data assimilated did impact the model biases. Table 2 lists the mean experimental temperature results for three representative stations, Osan AB (47122), Hoengsong AB (47118) and Taegu AB (471420) in the Republic of Korea for January and July. These results are typical of the results for each variable studied. The results show that the biases, in general, were reduced by both the warm start and by assimilating observed data into the model.
Table 2. List of Experiment Results for Temperature
|
Data Availability Experiments by Month |
||||||
|
Temperature (T) °F |
||||||
|
Data Source (see Table 1) |
Avg. T °F |
Avg. Max T |
Avg. Min T |
|||
|
Jan |
Jul |
Jan |
Jul |
Jan |
Jul |
|
|
Observed |
25.6 |
74.2 |
33.1 |
79.2 |
18.9 |
70.1 |
|
Experiment A |
24.9 |
76.3 |
31.1 |
82.3 |
18.8 |
71.8 |
|
Experiment B |
22.8 |
77.0 |
30.1 |
85.9 |
17.2 |
72.5 |
|
Experiment C |
21.8 |
79.8 |
24.9 |
86.2 |
12.4 |
73.4 |
|
Experiment D |
18.2 |
80.3 |
23.7 |
91.2 |
10.0 |
73.9 |
|
Experiment E |
20.9 |
78.1 |
26.9 |
85.3 |
15.6 |
72.8 |
|
Experiment F |
20.5 |
79.1 |
26.1 |
87.5 |
14.1 |
73.1 |
|
Experiment G |
20.2 |
78.8 |
25.7 |
86.0 |
14.9 |
72.9 |
|
Experiment H |
23.9 |
76.9 |
30.8 |
82.4 |
17.7 |
72.0 |
|
Experiment I |
22.5 |
77.4 |
25.6 |
84.9 |
13.1 |
72.9 |
A comparison of CLIMOD simulated climate statistics with the observed climate statistics for January 1973, 74, 75, 77 and 79 for temperature, dewpoint, wind, precipitation and cloud cover for Osan AB Korea is depicted in Table 3.
Table 3. Simulated climate statistics compared with observed climate statistics for Osan
|
OSAN AB KOREA STATISTICS
FOR JANUARY 1973, 74,75,77, and 79 |
||||||
|
Data Src |
Tmax Avg. |
Tmin Avg. |
Tavg |
Tmax |
Tmin |
DP avg. |
|
OBS |
35.1 |
16.5 |
26.8 |
53.0 |
-6.0 |
16.1 |
|
OBS 30 Yr |
35.0 |
16.0 |
26.0 |
56.0 |
16.0 |
16.0 |
|
10K A |
35.1 |
17.5 |
27.2 |
48.9 |
4.4 |
15.5 |
|
40K A |
34.6 |
19.0 |
27.7 |
48.9 |
1.8 |
16.9 |
|
10K D |
36.7 |
17.6 |
28.0 |
52.7 |
2.9 |
14.2 |
|
40K D |
35.2 |
20.2 |
28.8 |
49.3 |
1.8 |
17.5 |
|
|
|
|
|
|
|
|
|
Data Src |
Wind Dir |
Spd |
Pcp |
Snow |
Snow Dept |
Cld Cov |
|
OBS |
060. |
3.0 |
9999 |
999 |
999 |
99 |
|
OBS 30 Yr |
060. |
3.0 |
1.00 |
7.0 |
999 |
38 |
|
10K A |
333. |
1.6 |
0.89 |
4.9 |
0.1 |
42 |
|
40K A |
321. |
2.4 |
1.60 |
1.3 |
0.0 |
44 |
|
10K D |
306. |
1.8 |
1.26 |
4.0 |
0.1 |
46 |
|
40K D |
279. |
2.9 |
0.94 |
1.1 |
0.0 |
47 |
The ability of the model to capture the diurnal cycle is highlighted in Fig. 2
showing Osan AB mean temperature for the month of July. A warm bias was
Fig. 2. Diurnal temperature cycles for the data availability experiments for July 1973, 74, 75, 77, and 79. (Note: B, A, and I overlap)
observed in all of the simulated experiments. The daytime high temperature bias was larger then the overnight low temperature bias. The assimilation of observed data was effective in reducing the model high temperature bias from 11.9 F° to 3.1 F°.
5.2 Bias Reduction Techniques
From the data-denial experiments results, it has been determined that a method to reduce the bias in the simulated surface climatological database was needed. In an attempt to achieve this aim, a post-simulation method has been developed to "blend" observations and model generated data. The objective is to develop a method that will yield a single seamless and consistent database in such a way that the simulated statistics surrounding sites where observed data is available closely reflect the observed data. The blending of the observed and simulated data may seem to be somewhat redundant since the intention is to assimilate all or most of the available observed data into the model simulations. Thus, the model-generated database will already be a type of blend between the observational data and the model physics. However, the model is capable of simulating only a portion of the variance of each parameter because it cannot simulate atmospheric features that are below the resolvable scale of the model grid.
The formulation of the blending method is based on the principle that the statistical database should be heavily influenced by the observational data near the actual observing sites and blend smoothly into the model statistics away from observational data sites. The final composite climatology should also contain the full distribution of data, not just means and variances, so that a large suite of statistical parameters can be obtained.
The blending procedure uses a weighted average of model and observational data. The weights depend upon the spatial statistics of each variable. In complex terrain, the weights will be smaller than in more uniform terrain regions. In the areas of more uniform terrain, the extent of the area of high spatial correlation will be larger and therefore allow a greater impact of observational data on the simulated data.
The spatial correlation depends on a variety of factors including the terrain, the atmospheric dynamics, and the time of year. An estimate of the spatial correlation is obtained by inter-comparing the large grid simulations, the nested grid simulations and observational data where available.
The method is now being tested for operational use. Early results for the combined average temperature at three representative stations, Osan AB (47122), Hoengsong AB (47118) and Taegu AB (471420) Osan AB, Korea are shown in Table 4.
Table 4. Comparison of observed, simulated, and blended temperature
statistics
|
Data Blending Experiments by Month Temperature (T) °F |
||||||
|
Data Source |
Avg T |
Avg Max T |
Avg Min T |
|||
|
Jan |
Jul |
Jan |
Jul |
Jan |
Jul |
|
|
Observed |
25.6 |
74.2 |
33.1 |
79.2 |
18.9 |
70.1 |
|
A |
24.9 |
76.3 |
31.1 |
82.3 |
18.8 |
71.8 |
|
Blended |
25.2 |
75.0 |
32.1 |
80.9 |
18.9 |
70.5 |
|
D |
18.2 |
80.3 |
23.7 |
91.2 |
10.0 |
|