Characteristic Flavor Substances of Guizhou Black Tea and the Environmental Factors Influencing Their Formation Using Stable Isotopes and Headspace Gas Chromatography-Ion Mobility Spectrometry

.


Introduction
Tea is among the most consumed beverages worldwide [1].Tea contains many healthy active ingredients, such as polyphenols, cafeine, theanine, and theafavins [2,3].According to the Food and Agriculture Organization of the United Nations statistics (FAOSTAT) database (https:// www.fao.org/faostat/),China is one of the world's largest tea producers and consumers, with a large production and market share of black tea.Guizhou Province is also among the major black tea-producing areas in China.In 2022, tea plantations in Guizhou covered an area of more than 4,666 square kilometers, with an output value of 57.09 billion Chinese yuan.
Tea's favor is the most important determinant of its quality.Flavor refers to tea taste and aroma, with volatile organic compounds (VOCs) being signifcant contributors to its characteristic favor.Te aroma of tea is predominantly made up of volatile favor compounds [4].Plant volatiles are mainly divided into terpenoids (with foral, fruity aroma), fatty acid metabolic compounds (alcohols, ketones, aldehydes, etc., with sweet aroma), and phenylalanine/aromatic cyclic metabolic compounds (including volatile phenylpropanoids and aromatic compounds with foral, nutty aroma) [1,5].Te terpenoids are synthesized primarily through the methylerythritol-4-phosphate (MEP) or mevalonate (MVA) pathways.Fatty acid metabolic compounds are primarily synthesized through the degradation of the unsaturated C18 fatty acids linolenic and linoleic acid in the lipoxygenase (LOX) pathway [6].Phenylalanine/aromatic cyclic metabolites are primarily synthesized through the shikimic acid pathway [7].At present, it has been found that the synthesis pathways of the above volatile components will be afected by external environmental factors.
Ion mobility spectroscopy (IMS) is a highly efective and sensitive analytical technique based on ion mobility differences in the gas phase under a constant electric feld.In addition, headspace-gas chromatography ion mobility spectrometry (HS-GC-IMS) combines the advantages of high resolution provided by gas chromatography and high sensitivity provided by ion transfer spectrometry [8][9][10][11] to rapidly detect trace VOCs in samples without requiring a sample pretreatment.Currently, HS-GC-IMS has been widely used in the detection of volatile components such as fruit juices [9], peppers [4], huajiao [12], green tea [13], and other foods.
In recent years, stable isotope technology has emerged as an excellent novel method for origin identifcation and has been widely utilized in various felds such as soil science, medicine, agriculture, biology, ecology, and environmental studies.Te isotopic fractionation efect is infuenced by various factors, including altitude, soil composition, water sources, topography, and atmospheric composition.Tese factors lead to variations in isotope ratios among organisms in diferent regions, contributing to distinct regional profles in isotope ratios.Terefore, the detection of the isotope composition of organisms can refect their growth environment and their characteristics as a result of adaptation to environmental changes.Moreover, since the stable isotope composition in living organisms is an intrinsic property that is not afected during processing or storage [14], stable isotope technology has also been used to identify teaproducing areas [15][16][17][18].
Numerous studies have been conducted on the identifcation and characterization of volatile compounds in tea to explore the mechanism of pleasurable aroma formation and reveal the relationship between tea metabolite composition and aroma production [10,[19][20][21].However, there are few studies on the relationship between the volatile components of tea and environmental factors.In this study, HS-GC-IMS was used to analyze the characteristic favor components of black tea and to investigate the key favor components of black tea from various regions.According to the characteristics that the synthesis of volatile components and isotopic composition of plants are afected by natural factors in the plant environment, the infuence of environmental factors on the unique favor components of Guizhou black tea and the relationship between the isotopic composition and environmental factors were innovatively determined by the ratio of C and N stable isotopes in black tea samples.It is hoped that a correlation model can be established between the formation of Guizhou black tea favor and environmental factors.Tis would provide a new research direction for the comprehensive study of Guizhou black tea and ofer theoretical guidance for the future development of the black tea industry.

Materials and Methods
2.1.Samples.All black tea samples used in this experiment were collected in the spring of 2022 from four main black tea-producing areas in Guizhou and two main black teaproducing areas outside of Guizhou in China.Te samples were collected by picking one bud, one leaf, one bud, and two leaves, and they were processed aseptically to make black tea using Chen's methods [19].Te information on sampling locations is shown in Table 1 and Figure 1.

Headspace Gas Chromatography-Ion Mobility
Spectrometry (HS-GC-IMS) 2.2.1.Instruments and Equipment.Te FA2204 Electronic analytical balance was purchased from China, and the Flavor Spec ® HS-GC-IMS was purchased from G.A.S. Germany.

HS-GC-IMS Determination Method.
Prior to the experiment, the tea samples were processed as described previously with minor modifcations [13,22].Accordingly, an HS-GC-IMS FlavourSpec ® (GAS mbH, Dortmund, Germany) instrument was utilized for the study.About 1 g of samples was injected into a 20 mL headspace vial and incubated for 20 min at 85 °C.Subsequently, 500 μL of headspace gas was automatically collected with a heated syringe (85 °C) and was injected into the HS-GC-IMS for analysis.Te samples were then carried into the capillary column (FS-SE-54-CB-1, 15 m × 0.53 mm, 60 °C) by nitrogen (purity ≥99.999%) using the following fow procedure: initial fow rate was maintained at 2 mL/min for 2 min, increased to 10 mL/min at 10 min, then 100 mL/min at 20 min, and fnally 150 mL/min at 30 min.After preseparation, analytes were ionized in positive ion mode using a 3H ionization source located in the ion mobility spectrometer ionization chamber and then transferred to the drift tube at 45 °C under the drift gas (nitrogen, purity ≥99.999%) at 150 mL/min.Te drift tube length was 98 mm, and its linear voltage was 500 V/cm.Volatile compounds were identifed by comparing retention indices (RIs) and were then compared with the data obtained from the NIST17 mass spectral library and the IMS database (GAS mbH, Dortmund, Germany).

HS-GC-IMS Result Analysis.
Te VOC analysis software was used to inspect the analytical spectrogram to determine the retention and migration time for qualitative analysis.Te internal standard method was used to calculate the substance content, and the content calculation formula is as follows: 2 Journal of Food Biochemistry where C s is the concentration of volatile substances, C i is the concentration of standard substances, A s is the peak area of volatile substances, and A i is the peak area of standard substances.
In addition, the HS-GC-IMS Library Search software was compared with the NIST and IMS databases for qualitative analysis of characteristic favor compounds.Te 3D diference map, 2D overhead diference map, and fngerprint map of VOCs were constructed using Reporter and Gallery Plot.Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) and Partial Least Squares Discriminant Analysis (PLS-DA) were used to analyze the diferences between the sample groups.
Te key favor substances were selected by calculating the volatile component VIP value and odor activity value (OAV) value.

Stable Isotope Analysis
2.3.1.Instruments and Equipment.Te XP6 electronic balance was purchased from Mettler-Toledo International Inc in Switzerland; the DI/CF-MAT253 gas isotope ratio mass spectrometer was purchased from Termo Fisher Scientifc; and the YXQM-0.4Lplanetary ball mill was purchased from MITR in China.

Measurements of C and N Element Content and Teir
Stable Isotope Ratios.Tea samples were crushed with a ball mill and sieved through a 100-mesh screen prior to the experiment.Approximately 1 mg from each tea sample was weighed and transferred to a 3.5 × 5 mm pressed tin capsule.Te samples were placed sequentially into the automatic solid sample tray of the element analyzer.After reduction resulting from combustion, the carbon and nitrogen elements in the samples were converted into pure CO 2 and N 2 gas.Ten, the isotope ratio mass spectrometer was used for the detection of isotope composition.
Te δ-value (‰) was determined by comparing the isotope value of the examined sample to internationally accepted standard materials.Te isotopic values were expressed according to the following formula: where R sample and R standard are the isotopic ratios of the sample and the standard materials, respectively.Isotope standards IAEA-CH-3 and IAEA-N1 were used for the δ 13 C and δ 15 N calibration.

Meteorological Information
Collection.GPS was used to locate and record the sampling locations' longitude, latitude, and altitude.Moreover, the China Meteorological Science Data Center was queried for the average annual precipitation, average high temperature, average low temperature, and number of sunny days.

Data Analysis.
Te correlation model between the carbon and nitrogen isotope ratios and environmental factors was established by the partial least squares (PLS) method.Te least squares method is a commonly implemented method in unitary linear regression analysis.It minimizes the sum of the squared errors between the observed and predicted values, generating a best-ftting line to obtain the functional relationship between the independent variable and the dependent variable.A quantitative prediction model was established to assess the relationship between carbon and nitrogen isotope ratios and environmental factors, using 8 environmental factors as independent variables and each stable isotope ratio as dependent variables.Te model included 18 samples, of which 16 were randomly selected and used as the training set and 2 as the prediction sets.Te model results were expressed in terms of coefcient of determination (R), root mean square error (RMSE), and relative prediction deviation (RPD) for all samples (where RPD � DS/RMSE).
Te box diagrams and bar diagrams were created using the Origin 2021 software.OPLS-DA was conducted on the SIMCA 14.1 program, and VIP value, partial least squares, and PLS analysis were performed on XLSTAT.

Analysis of Volatile Compounds of Black Tea Samples from Diferent Regions by HS-GC-IMS.
Te volatile favor compounds in 4 black tea samples from diferent regions in Guizhou and 2 black tea samples from regions outside of Guizhou were determined by GC-IMS.Using the Reporter plug-in of the included VOC analysis software, the volatile compounds of black tea from diferent regions were compared, and the results are shown in Figure 2.
Te GC-IMS 3D spectra of black tea samples from 6 regions are shown in Figure 2(a).Te variation in volatile compounds across diferent samples is clearly visible in the fgure.Te top view of the 2D GC-IMS diagram of black tea samples from 6 regions is shown in Figure 2(b).Te background of the image is blue, with the vertical axis representing the retention time (s) of gas chromatography, and the horizontal axis representing the ion migration time.4 Journal of Food Biochemistry Te red vertical line at 1.0 on the horizontal axis indicates the reaction ion peak (RIP).Both the migration time and RIP were normalized.Each peak on the left and right sides of the RIP corresponds to a volatile organic compound, and the color indicates the compound concentration: white indicates a lower concentration, red is a higher concentration, and a darker color corresponds to a greater concentration.As shown in Figure 2(b), most volatile compound signals of the black tea samples from 6 diferent Guizhou regions were in a retention time range of 0∼1750 s and drift time range of 1.0∼1.7 s, indicating that the volatile compounds of the black tea samples from diferent regions were highly similar.However, their content in the diferent regions was diferent.For example, the two GZZY samples were similar to WY and BS, while the GZAS, GZPA, and GZDY samples shared a higher similarity.Te WY sample was taken as a reference to accentuate the diferences, and the signal peak in the reference spectrum was deducted to obtain the diferential spectral map between diferent samples (Figure 2(c)).Te background was white after deduction, and red indicated that the corresponding compound content was higher than WY, while a darker color indicated greater diferences and a blue color indicated that the corresponding compound content was lower than WY.As shown in Figure 2(c), the volatile compound contents of GZAS, GZPA, and GZDY were signifcantly diferent from those of WY and BS (red box in the fgure).In contrast, the volatile compound contents of BS, GZZY, and WY were relatively low, with BS and WY still exhibiting some diferences (green box in the fgure).Finally, the volatile compounds of GZZY and WY samples were very similar.Terefore, overall, the diferences between Guizhou tea samples and non-Guizhou tea samples are small (purple box in fgure) and cannot be clearly distinguished.

Qualitative and Quantitative Analysis of Volatile
Flavor Compounds.To more intuitively assess the changes in the content of volatile compounds in diferent samples, the retention time and migration time of volatile favor compounds from samples from diferent regions were compared, and the HS-GC-IMS migration time data were compared with the NIST database to perform qualitative analysis of volatile compounds.Quantitative analysis was performed using the content calculation formula, and the results are shown in Table 2.A total of 184 volatile compounds were detected in the black tea samples, and 143 volatile compounds (including dimers and monomers) were identifcation through the database.As shown in Figure 3, there were 30 aldehydes (accounting for 21%∼27% of the total volatile compounds), 30 alcohols (16%∼24% of the total volatile compounds), 26 ketones (21%∼28% of the total volatile compounds), and 21 heterocyclic classes (7%∼14% of the total volatile compounds).
Tere were 18 esters (4%-7% of the total volatile compounds), 9 terpenes (1% of the total volatile compounds), 5 acids (11%-21% of the total volatile compounds), and 4 sulfdes (2%-3% of the total volatile compounds).Te total volatile compound content in the GZZY, BS, and WY samples was higher than that in the other three samples, and the volatile compounds of the WY and GZZY samples were highly similar.
Te Gallery Plot plug-in was used to plot the fngerprint spectrum of volatile compounds (Figure 4).Each row in the fgure represents all the signal peaks identifed in a sample, and each column represents the signal peaks of the same volatile organic compound in diferent samples, followed by M and D, which correspond to the monomers and dimers of the same compounds, and the number indicates the unidentifed peaks.

Analysis of Volatile Aroma Compounds in Black Tea
with OPLS-DA.OPLS-DA is a supervised discriminant analysis statistical method that can establish a relationship model between the samples and their volatile compound abundance to categorize them.Te model evaluation parameters were the independent variable ft index (R2X),         Te OPLS-DA model was established according to the volatile compounds identifed in black tea samples from 6 diferent regions.Te results are shown in Figure 5(a).Te six samples could be distinguished efectively, and the model indices R2X � 0.983, R2Y � 0.997, and Q2 � 0.993 indicate a very high and reliable prediction ability.Tus, the model can be used to distinguish black tea samples from diferent regions.In addition, to prevent the model from overftting the data, a replacement test (n � 200) was also carried out on the model.Te results are shown in Figure 5(b).All the green R2 values and blue Q2 values on the left are lower than the original point on the right, and the intercept between the regression line and the vertical axis (on the left) of the Q2 point is negative, indicating that the model does not overft and has good prediction ability.Tus, it can be used for discriminant analysis of aroma compounds from black tea samples of diferent origins.

Analysis of Volatile Aroma Compounds of Black Tea
with PLS-DA.Te 143 compounds identifed in Table 2 were assessed with PLS-DA.PLS-DA is a supervised discriminant analysis method, a multivariate statistical analysis method, which can determine the classifcation of research objects according to the observed or measured values of several variables.PLS-DA maximizes the diference between groups according to a predefned classifcation (Y variable), achieving better separation results than principal component analysis.
According to the PLS-DA score plots (Figure 6), WY and GZZY were in close proximity, indicating that the two samples have similar aroma types.In comparison, the other four samples were relatively dispersed and independent, indicating that the model can distinguish diferent black tea origins.

Analysis of Key Flavor Compounds of Black Tea from
Diferent Guizhou Regions.In order to further analyze the contribution rates of key favor compounds of black tea products from 4 Guizhou regions and 2 regions outside of Guizhou, a variable importance (VIP) >1 and a predicted P value <0.05 were used to identify the key volatile compounds.A total of 50 favor compounds were screened and are listed in Figure 7. Tey include 13 aldehydes, 10 alcohols, 7 ketones, 6 terpenes, 5 esters, 5 heterocycles, 3 sulfurcontaining compounds, and 1 acid.
Te OAV is the ratio of the concentration of each compound to its detection threshold.Te odor activity value can determine the contribution of each volatile compound in the sample to the overall aroma.When OAV ≥1, the aroma compound can theoretically be considered to contribute to the overall aroma.Moreover, the higher the OAV, the larger the contribution to aroma.Using the threshold values of related compounds provided by references [31,32], the OAV of all volatile compounds was calculated.Finally, 83 compounds with OAV ≥1 in at least one sample were obtained, including 1 acid, 13 alcohols, 22 aldehydes, 11 esters, 11 heterocyclic compounds, 16 ketones, 3 sulfdes, and 6 terpenes.When a volatile compound has an OAV ≥100 in the sample, it can be considered to have a greater contribution to the overall aroma.Terefore, all the volatile compounds with VIP greater than 1 or OAV ≥100 in the sample were considered as the main aroma compounds of black tea (68 in total), and cluster analysis was conducted accordingly (Figure 3).A stronger correlation between the compound and the overall aroma of the sample is indicated by a lighter color, while weaker correlations are indicated with darker colors.
As shown in Figure 8, the cluster analysis showed that GZZY and WY samples had a high similarity.Te GZDY and GZPA samples were also similar, while GZAS samples exhibited some overlap with the GZZY, WY, GZDY, and GZPA samples, while the BS samples were distinct.

Efects of Environmental and Meteorological Factors on the
Formation of the Characteristic Flavor of Black Tea.
Trough HS-GC-IMS analysis, the distinctive favor components of black tea samples from various origins were identifed.However, it was also observed that the similarity in volatile components between the GZZY sample, a black tea produced in Guizhou Province, and the WY sample, a black tea produced in Fujian Province, was higher than that of other black teas from Guizhou Province.To investigate this phenomenon further, the researchers conducted the following study.

Efects of Environmental Factors in Diferent Black Tea
Regions on Volatile Components of Black Tea.Trough the weather website, we checked the annual climate conditions of each producing area.It was found that the sample producing areas with similar volatile components had similar climate conditions.Terefore, it was speculated that environmental factors may be one of the reasons for the difference in volatile components of black tea.At present, studies have shown that the synthesis of volatile components in plants is afected by environmental factors.For example, high temperature can promote the synthesis and subsequent release of foral and fruity terpene compounds by inducing and increasing the activity of enzymes related to MEP and MVA pathways.It can also signifcantly express LOX in plants, promoting the accumulation of related metabolites (alcohols, ketones, aldehydes, etc.), most of which can bring a sweet odor [33].Te average temperature of the WY region and GZZY region is higher than that of the other four regions.Terefore, the content of related volatile components in the WY sample and GZZY sample is signifcantly higher than that of the other four samples, making their foral, fruity, and sweet fragrance stronger.Te shikimic acid pathway in plants is inhibited under long-term high-temperature but can be activated under intense light and low-temperature conditions, thus promoting the synthesis of phenylpropane and aromatic compounds with sweet, foral, and nutty aromas in plants [34].GZAS, GZPA, and GZDY, thanks to their high altitude, low average temperatures, and abundant sunlight, produce black teas that are fresh and light, with some nutty notes.Tis provides theoretical guidance for the study of the favor of black tea produced in Guizhou.However, as GC-IMS technology is an emerging technology, its database is not complete and refned, and there are still many compounds that cannot be annotated, so it is 14 Journal of Food Biochemistry impossible to know whether these compounds are the key favor compounds in black tea.We anticipate that GC-IMS results will become more precise as technology advances and GC-IMS data become more and more refned.

Relationship between C and N Stable Isotope Composition and Environmental Factors in Black Tea.
Environmental and meteorological factors not only afect the synthesis of volatile aroma compounds but also have a great infuence on the fractionation of stable isotopes in plants.
Diferent environmental meteorological factors can alter the ratio of stable isotopes in plants.For example, the carbon isotope fractionation of carbon assimilated in organisms is afected by multiple environmental factors such as temperature, precipitation, light, and atmospheric pressure [35].Specifcally, in arid areas with low air humidity and strong evaporation, plants reduce stomatal conductance, decreasing intercellular CO 2 concentration and increasing the δ 13 C value of carbon assimilated by photosynthates.During respiration, temperature also afects the CO 2 produced by the leaves and the δ 13 C values of the main organic compounds of the leaves [35].Nitrogen is an important biological element and one of the key components of amino acids, signifcantly impacting organisms' growth and development [36].Nitrogen isotope fractionation in plant tissues is mainly afected by soil properties, fertilization, altitude, temperature, and other factors [37,38].For example, when the ambient temperature increases, the biological activity of nitrifying bacteria and ammoniating bacteria in the soil is enhanced.Tis accelerates the rate of soil mineralization and nitrifcation, leading to an increase in the content of soil 15 N, consequently raising δ 15 N of plants [39][40][41].
Based on the aforementioned as a preliminary reference, it is hypothesized that there might exist a potential correlation between the volatile components of black tea and its internal isotopic compositions.By measuring the δ 13 C and δ 15 N values of black tea samples from six distinct regions, in conjunction with local environmental factors, we assessed the relationship between environmental factors and isotope fractionation across diferent regions, with the aim of establishing a relational model capable of refecting both the isotope ratios and volatile composition of black tea.
Te stable isotope ratios of black tea samples from 4 Guizhou regions and 2 non-Guizhou regions were obtained, as shown in Table 3.In addition, the δ 13 C values of the samples from Guizhou province ranged from −28.71‰ to −25.37‰, while the δ 15 N values ranged from 0.75‰ to 3.31‰.Te δ 13 C values of the WY samples ranged from −26.11‰ to −25.52‰, while the δ 15 N values ranged from 2.77‰ to 2.99‰.Furthermore, the δ 13 C and δ 15 N values of BS samples ranged between −27.53‰ and −26.44‰ and between 0.02‰ and 0.22‰, respectively.
In order to investigate the impact of environmental and meteorological factors on the ratio of stable isotopes of carbon and nitrogen in plants, a quantitative prediction model was developed using PLS analysis of the environmental meteorological factors and the δ 13 C values in each sample region (Figure 9).Te quantitative prediction model results are shown in Table 4.As R 2 and RPD increase, the RMSE decreases, and the model accuracy improves.Te R 2 values of the δ 13 C and δ 15 N models were 0.09 and 0.63, respectively.Te RMSE values were 0.94 and 0.64, respectively, and the RPD values were 0.94 and 1.06, respectively.Tis indicates that the δ 15 N prediction model based on eight environmental and meteorological factors is more accurate than the δ 13 C prediction model.
Further analysis of the importance of the model variables shows that if the VIP value of the environmental factor is all greater than 1, it indicates that the environmental factor has a signifcant impact on the ratio of carbon and nitrogen isotopes.Te results are depicted in Figure 9.
Variables with VIP values of carbon isotope ratios greater than 1 include annual rainfall, mean low temperature, mean high temperature, and altitude, indicating that temperature, rainfall, and altitude have signifcant efects on δ 13 C values.In addition, the positive correlation coefcients of the four factors showed that the δ 13 C value of black tea was positively correlated with the temperature and rainfall of the producing area and negatively correlated with the altitude of the producing area.Regarding the δ 15 N values, four environmental factors, total rainfall, number of sunny days, number of cloudy days, and mean low temperature, had VIP values greater than 1.Among the four factors, only the normalized correlation coefcients of number of sunny days were negative, while those of the other three factors were positive.Tis result shows that the δ 15 N value is afected by several factors, such as precipitation temperature and sunshine.
At present, a large number of studies show that when the environmental precipitation decreases or the soil moisture decreases, the water stress increases.In order to reduce water transpiration loss in the body, plants often close part of their stomata, reducing stomatal conductance and intercellular CO 2 concentration, resulting in an increase in plant δ 13 C [42].However, when the precipitation is too high or the soil moisture is too high, the soil microbial activity decreases, the respiration rate decreases sharply, the soil nitrifcation is inhibited, and the availability of soil inorganic nitrogen decreases, thus making the soil 15 N poor and resulting in the decrease of plant δ 15 N [43].According to Wang et al. [44], as the altitude increases, the temperature decreases, the atmospheric CO 2 partial pressure decreases, and the amount of CO 2 available to plants decreases.Plants, due to insufcient CO2 supply, cause carboxylase enzymes to be unable to fractionate 13C, leading to a direct synthesis of organic compounds, thereby resulting in an increase in plant δ 13 C.
In this study, the δ 13 C and δ 15 N values of black tea are positively correlated with the temperature of the producing area, which is consistent with the results of Julien et al. [35].However, the correlation results of δ 13 C and δ 15 N values of black tea with environmental elevation and rainfall are diferent from the current mainstream research results.Water et al. [45] believe that the positive ratio of δ 13 C value to environmental elevation may be due to the infuence of regional water use efciency and drought stress elimination.

Journal of Food Biochemistry
In addition, Schulze [46], Heaton [47], and Codron [48] also found that the δ 13 C and δ 15 N values in plants were positively correlated with regional rainfall under some special circumstances.To sum up, the efects of other factors on the stable isotope ratio of black tea may be superimposed, making the situation more complicated.Terefore, the

16
Journal of Food Biochemistry mechanism of how various environmental factors afect the stable isotope ratio of black tea still needs to be further studied.

Summary of the Efects of Environmental Factors on the Formation of Volatile Components in Black Tea.
According to the environmental and meteorological conditions of six black tea samples and the variation in volatile components in each sample, it was observed that the volatile components of black tea were signifcantly infuenced by temperature, sunlight, and other factors.Te levels of terpenoids, alcohols, ketones, and aldehydes in black tea samples were found to be higher in regions with elevated temperatures.Tese compounds contribute to the stronger foral, fruity, and sweet favors characteristic of teas produced in these regions.In regions with slightly lower average temperatures and abundant sunshine, the plants contain higher levels of substances with fragrant, foral, and nutty aromas.Tis results in tea produced in these areas having a lighter, sweeter smell, a fresher aroma, and a hint of nuttiness.
In addition to the volatile components, external environmental factors also infuence the internal isotopic composition of tea.Te δ 13 C and δ 15 N values of black tea samples from six regions were detected, and a PLS-VIP regression model was used for analysis.Te results showed that the δ 13 C value of black tea samples was mainly afected by the temperature of origin, altitude, and rainfall.It was positively correlated with temperature and rainfall but negatively correlated with altitude.Te δ 15 N value is mainly afected by precipitation, temperature, and sunshine.It is positively correlated with precipitation and temperature but negatively correlated with sunshine.
Combined with the analysis of the relationship between the two and the environmental factors in the producing area, it was found that the content of volatile components in black tea samples was positively correlated with δ 13 C and δ 15 N due to the environmental temperature in the producing area, indicating that the stable isotope ratio in black tea samples may refect the content of volatile components.However, due to the small sample size, this conclusion needs to be further studied after expanding the sample size.

Conclusion and Discussion
In this study, GC-IMS technology was implemented to determine volatile compounds from 4 black tea samples from Guizhou and 2 black tea samples from regions outside of Guizhou.A total of 184 volatile compounds were detected, and 143 compounds were annotated through database comparisons.Te data were analyzed by OPLS-DA, and the results showed that the HS-GC-IMS could efectively distinguish black tea samples from diferent regions.By calculating the VIP and OAV values of the volatile compounds in the samples, a total of 83 important volatile aroma compounds were selected.After analysis by PLS-DA and cluster analysis, the distinctive aroma compounds of black tea samples from diferent regions in Guizhou were identifed.Te distinctive aroma compounds in the GZZY sample were Limonene, 4-Methyl-3-penten-2-one, 2-Pentylfuran, 1-Propanethiol, Ethyl butanoate, gamma-Terpinene, and Ethyl pentanoate.Tese substances give GZZY samples strong foral and fruity aromas.Te distinctive aroma compounds in the GZAS sample were Limonene, 4-Methyl-3-penten-2-one, 2-Pentylfuran, 1-Propanethiol, Ethyl butanoate, gamma-Terpinene, and Ethyl pentanoate.Tese substances give the GZAS sample a fresh grassy, foral, and fruity aroma.Te characteristic aroma compounds in GZPA samples were 2-Methylpropanoic acid, (E)-2-Nonenal, o-Xylene, Methyl acetate, Hexanal, 1-Propanol, and 2-Acetyl-1-pyrroline. Tese substances have a strong sweet fragrance, which makes the GZPA sample smell with some honey odor; and in the GZDY sample 3-Methylbutanal, 3-Carene and Ethyl acetate.In addition, among the black tea samples from the other two regions, the characteristic favor substances in WY black tea are Dimethyl disulfde, 2-Propanol, 1-Hexanol, alpha-Terpinene, 2, 5-Dimethylpyrazine, and 5-Methyl-3heptanone.Tese ingredients give WY sample a special dry longan aroma.Te characteristic favor substances in BS black tea samples were cis-4-Heptenal, 1-Penten-3-one-D, Diacetyl, beta-Pinene, Phenylacetaldehyde, and (E, E)-2, 4-Hexadienal which gave BS black tea samples a strong fruity favor.
In addition, through the investigation of the climate environment of diferent producing areas, combined with the analysis of the volatile component content and stable C and N isotope ratios in each sample, it was found that the formation of volatile components in black tea was greatly afected by ambient temperature and light.In the regions with higher temperature, the volatile components with foral, fruity, and sweet aroma were higher, and the black tea had a stronger aroma.In the regions where the temperature is low but the light is strong, the fragrant and nutty components in black tea are higher, and the aroma of black tea is relatively elegant and fresh.At the same time, it was found that the stable C and N isotope ratios in black tea were also signifcantly afected by ambient temperature and were positively correlated.
Tis study examined how characteristic favor compounds and environmental factors afect the favor of Guizhou black tea.It ofers a theoretical framework for optimizing planting conditions and enhancing product quality.Additionally, it sets a new direction for in-depth research on Guizhou black tea and contributes positively to the industry's development.

Figure 3 :Figure 4 :
Figure 3: Volatile compound content and their proportion in samples from diferent tea-producing regions.

Figure 5 :
Figure 5: OPLS-DA analysis of volatile aroma compounds in black tea samples from diferent regions.

Figure 6 :
Figure 6: PLS-DA analysis of volatile aroma compounds in black tea samples from diferent regions.

Figure 7 :
Figure 7: VIP analysis of volatile compounds in black tea samples.

Figure 8 :
Figure 8: Cluster analysis of key favor compounds in black tea samples.

Figure 9 :
Figure 9: PLS analysis results of black tea samples from 6 regions.

Table 2 :
[4,12,23,24]no aroma description of the substance was found.a,b,c, and d indicate signifcant diference.Odor descriptions were from the literature[4,12,23,24]or from FEMA database. N R2X and R2Y represent the interpretation rate of the built model for X and Y matrices, respectively, and Q2 represents the prediction ability of the model.Te model is considered highly reliable when all three indices approach 1. Q2 > 0.5 indicates a reliable model, and Q2 > 0.9 indicates a highly reliable predictive model.

Table 3 :
Stable isotope data of black tea samples from 4 Guizhou regions and 2 non-Guizhou regions.

Table 4 :
Partial least squares regression models of δ 13 C and δ 15 N.