Climate velocities and species tracking in global mountain regions
Data files
Feb 02, 2024 version files 54.60 MB
-
organized_v7.rar
54.57 MB
-
README.md
25.39 KB
Abstract
Mountain ranges harbor high concentrations of endemic species and are indispensable refugia for lowland species under anthropogenic climate change1,2. Forecasting biodiversity redistribution hinges on assessing whether species can track shifting isotherms as climate warms3,4. However, a global analysis of isotherm shift velocities along elevation gradients is hindered by the scarcity of weather stations in mountainous regions5. We address this by mapping the lapse rate of temperature (LRT) across mountain regions globally using satellite data (SLRT) and laws of thermodynamics to account for water vapour6 (i.e., the moist adiabatic lapse rate: MALRT). Dividing the rate of surface warming from 1971 to 2020 by either the SLRT or MALRT, we provide the first maps of vertical isotherm shift velocities. We identify 17 mountain regions with exceptionally high vertical isotherm shift velocities (> 11.67 m/yr for the SLRT, > 8.25 m/yr for the MALRT), predominantly in dry areas but also in wet regions with shallow lapse rates like Northern Sumatra, the Brazilian Highlands, and Southern Africa. By linking these velocities to species range shift velocities, we report instances of close tracking in mountains with lower climate velocities. However, many species lag behind, suggesting persisting range shift dynamics even if we manage to curb climate change trajectories. Our findings are vital for devising global conservation strategies, particularly in the 17 high-velocity mountain regions we identified.
References
- Rahbek, C. et al. Building mountain biodiversity: Geological and evolutionary processes. Science 365, 1114-1119 (2019).
- Rahbek, C. et al. Humboldt’s enigma: What causes global patterns of mountain biodiversity? Science 365, 1108-1113 (2019).
- Chen, I. C., Hill, J. K., Ohlemuller, R., Roy, D. B. & Thomas, C. D. Rapid Range Shifts of Species Associated with High Levels of Climate Warming. Science 333, 1024-1026 (2011).
- Lenoir, J. et al. Species better track climate warming in the oceans than on land. Nature Ecology & Evolution, 1-16 (2020).
- Pepin, N. et al. Elevation-dependent warming in mountain regions of the world. Nat Clim Change 5, 424-430 (2015).
- Holton, J. R. & Hakim, G. J. An introduction to dynamic meteorology. Vol. 88 (Academic press, 2012).
README: Chan et al., Climate Velocities and Species Tracking in Global Mountain Regions (2024)
[Access this dataset on Dryad] https://doi.org/10.5061/dryad.1rn8pk0wm
This data collection accompanies the Chan et al., 2024 paper which evaluates climate velocities in global mountain regions and compares these with species range shift velocities. It includes climatic, geographic, and biological data used in the study.
Description of the data and file structure
The dataset is organized into two main folders: 'data' and 'scripts'.
'data' folder
The 'data' folder contains datasets generated for the study's figures and intermediate products. Maps in this collection are primarily stored in two formats: TIFF and CSV matrix. Both formats generally adhere to a global resolution of 360 x 720 (0.5 degrees).
#Datasets that are not directly readable are likely intended for use with scripts for further figure production and not for direct human interpretation. These datasets are provided to facilitate reproducibility.
Key Abbreviations:
- MALRT: moist adiabatic lapse rate of temperature
- SLRT: satellite-derived version of lapse rate of temperature
- tmp: temperature
- vap: water vapour
- cld: cloud coverage
- climate velocity: can be based on MALRT, SLRT, or constant LRT
Sub-folders
'MALRT_SLRT_Climate_Velocity': Contains key final products, including MALRT, SLRT, climate velocity (based on MALRT), and the map of high climate velocity global mountain regions (Google Earth file; also provided as Supplementary Data 3 in Chan et al., 2024).
For MALRT and SLRT: scaling factor: 100; unit: c/km; NA: -9999 (Map resolution: 0.5 degrees)
For climate velocity (based on MALRT): unit: m/yr; NA: -9999 (Map resolution: 0.5 degrees)
'result_figs': Contains figures derived from the study's results.
Additional folders are named according to their respective analyses in Chan et al., 2024:
'data_for_SLRT': Datasets used for calculating SLRT.
- [cru_ts4.05.2011-2020.malr.csv]: MALRT globally from 2011-2020 derived from CRU dataset; scaling factor: 100; unit: c/km; NA: -9999
- [CRU2_dem_05d.csv]: elevation data map used in CRU data generation; unit: m; NA: -9999
- [ETOPO1_bed_005d.tif]: elevation data map based on ETOPO1; unit: m
'MALRT_SLRT_ML_results': Random forest analysis results for environmental variables explaining MALRT and SLRT.
- [salrt_median_rfr_keep_rows_data20230418_rm004_latlng+malr.csv]: results derived from the random forest script (Env_Random-Forest.py); see Supplementary Information in Chan et al., 2024 for details
- [salrt_median_rfr_keep_rows_data20230418_rm004_latlng+tmp+pre.csv]: results derived from the random forest script (Env_Random-Forest.py); see Supplementary Information in Chan et al., 2024 for details
- [salrt_median_rfr_keep_rows_data20230418_rm004_malr+tmp+pre.csv]: results derived from the random forest script (Env_Random-Forest.py); see Supplementary Information in Chan et al., 2024 for details
- Columns (applicable to the three files above):
- abs.lat: absolute latitude (unit: relative importance)
- longitude: longitude (unit: relative importance)
- malr: MALRT (unit: relative importance)
- pre: mean annual precipitation (unit: relative importance)
- tmp: temperature (unit: relative importance)
- r2_oob: out-of-box R-squared
- r2_x: overall R-squared
- Columns (applicable to the three files above):
'range_shift_ML_results': Random forest analysis results for environmental variables explaining species range shifts.
- [*malr.csv]: results derived from the random forest script (Bio_Random-Forest.py); see Extended Data Fig. 5 in Chan et al., 2024 for details
- Columns:
- centroid.lon: longitude (unit: relative importance)
- abs_lat: absolute latitude (unit: relative importance)
- tmp_rate_over_malr: climate velocity based on MALRT (unit: relative importance)
- r2_oob: out-of-box R-squared
- r2_x: overall R-squared
- Columns:
- [*mod_lr_median.csv]: results derived from the random forest script (Bio_Random-Forest.py); see Extended Data Fig. 5 in Chan et al., 2024 for details
- Columns:
- centroid.lon: longitude (unit: relative importance)
- abs_lat: absolute latitude (unit: relative importance)
- tmp_rate_over_modlr: climate velocity based on SLRT (unit: relative importance)
- r2_oob: out-of-box R-squared
- r2_x: overall R-squared
- Columns:
- [*rev.csv]: results derived from the random forest script (Bio_Random-Forest.py); see Extended Data Fig. 5 in Chan et al., 2024 for details
- Columns:
- variables from 'range_shift_dataset_Dec2021.csv' (unit: relative importance)
- r2_oob: out-of-box R-squared
- r2_x: overall R-squared
- Columns:
'study_area_tables': Intermediate products used for quantifying study regions in the BioShift dataset.
- [*.xls]: raw tables indicate the polygon size of each studied area in BioShifts dataset, further integrated by the script 'integrate_area_info.nb'
- Columns:
- Name: the name of the studied regions
- Shape_Length: the perimeter of the polygon
- Shape_Area: the area of the studied polygon
- Columns:
'variable_maps' and 'variable_maps-2': Climatic and environmental variables used in the analyses. Note that there is some overlap between these two folders. Some maps became redundant during the review process.
- [cru_ts4.05.1971-1980.tmp.tif]: tiff format of 'cru_ts4.05.1971-1980.tmp.csv'
- [cru_ts4.05.2011-2020.cld.tif]: averaged cloud coverage from 2011-2020 derived from CRU dataset; scaling factor: 10; unit: C; NA: -999
- [cru_ts4.05.2011-2020.malr.tif]: MALRT globally from 2011-2020 derived from CRU dataset; scaling factor: 100; unit: c/km; NA: NA
- [cru_ts4.05.2011-2020.pre.tif]: averaged annual precipitation from 2011-2020 derived from CRU dataset; unit: mm; NA: -999
- [cru_ts4.05.2011-2020.tmp.tif]: tiff format of 'cru_ts4.05.2011-2020.tmp.csv'
- [cru_ts4.05.2011-2020.vap.tif]: averaged water vapour globally from 2011-2020 derived from CRU dataset; unit: Pa; NA: -999
- [cru_ts4.05.Climate_velocity.1975_VS_2015_mt.tif]: tiff format of 'cru_ts4.05.Climate_velocity.1975_VS_2015_mt.csv'
- [cru_ts4.05.MOD11C3.Climate_velocity.1975_VS_2015.tif]: climate velocity globally (1970-1980 vs 2011-2020) based on SLRT; unit: m/yr; NA: NA
- [cru_ts4.05.MOD11C3.Climate_velocity.1975_VS_2015_mt.High-velocity_regions_matrix-labeled.tif]: high climatic velocity regions in global mountains (not included in the final manuscript)
- [cru_ts4.05.MOD11C3.Isotherm.1975_VS_2015.tif]: temperature difference (1970-1980 vs 2011-2020) divided by SLRT; unit: m; NA: NA
- [cru_ts4.05.Tmp_rate.1975_VS_2015.tif]: temperature difference (1970-1980 vs 2011-2020) calculated over a 40-year time span; unit: C/yr; NA: NA
- [bioRea_ras05.tif]: raster data of biogeographic realm (not included in the final manuscript)
- [EcoFecet_de_Majority_005d.txt]: ecological facet based on 'EF_value_translator.txt' (not included in the final manuscript)
- [EF_value_translator.txt]: a dictionary for interpreting numbers in 'EcoFecet_de_Majority_005d.txt' (not included in the final manuscript)
- [Latitude.tif]: a map indicating latitude values globally; unit: degree
- [CRU2_dem_05d-georef.tif]: elevation data map used in CRU data generation; unit: m; NA: -9999
- [MOD11C3_A2011-2020_all-MALR_regional-elevation_SRTM_avg.tif]: averaged elevation for each grid of the SLRT map; unit: m
- [MOD11C3_A2011-2020_all-MALR_regional-elevation_SRTM_rng.tif]: elevational range for each grid of the SLRT map; unit: m
- [MOD11C3_A2011-2020_all-MALR_regional-lp_mean.tif]: tiff format of 'MOD11C3_A2011-2020_all-MALR_regional-lp_mean_mt.csv', but globally
- [MOD11C3_A2011-2020_all-MALR_regional-lp_median.tif]: tiff format of 'MOD11C3_A2011-2020_all-MALR_regional-lp_median_mt.csv', but globally
- [MCD12C1.A2011-2019.006.Land_type-LAI_base_05d.tif]: land use land type data based no leaf area index; details available in the MODIS MCD12C1 product documentation (not included in the final manuscript)
- [MOD13C2.A2011-2020_EVI_05d.tif]: enhanced vegetation index; unit: EVI (not included in the final manuscript)
- [MODCF_CloudForestPrediction_05d.tif]: global distribution of cloud forest (not included in the final manuscript)
- [SRTM_island_maks_ref_x_300_y_30.tif]: distribution of islands, with details available in Chan et al., 2024
- [aspectcosine_05d_SRTM.tif], [aspectsine_05d_SRTM.tif], [evenness_01_05_05d.tif], [Homogeneity_01_05_05d.tif], [pcurv_05d_GMTEDmd.tif], [roughness_05d_SRTM.tif], [slope_05d_SRTM.tif], [tcurv_05d_GMTEDmd.tif], [vrm_05d_SRTM.tif]): These datasets offer a range of topographical and textural features of the Earth's surface, with details accessible via EarthEnv (https://www.earthenv.org/).
'weather_station_data': Climatic data from GHCN used for comparison with MALRT and SLRT.
- [*.csv]: raw GHCN monthly data datasets downloaded from Google Cloud, with details available on the Google Cloud Platform Website (https://console.cloud.google.com/marketplace/product/noaa-public/ghcn-m)
- Columns:
- id: station id
- year: year
- element: the statistical method used to summarize the data
- value*: monthly data (unit: C)
- qcflag*: note of quality check for the corresponding monthly data
- Columns:
Unnested files
- [all_study_area_info.csv]: Summary of the size of studied areas from the BioShifts dataset.
- [all_study_area_info_size_selection.csv]: Terrestrial studied areas smaller than 1 were selected for biological analysis.
- Columns:
- n_polygons: the total number of polygons within each studied region
- Shape_Length: the perimeter of the polygon
- Shape_Area: the area of the studied polygon
- Less_than_1: polygon size less than one: 1; polygon size greater than one: 0
- Marine: marine: 1; non-marine: 0
- Terrestrial: terrestrial: 1; non-terrestrial: 0
- problematic: flags datasets within small terrestrial studied regions that could not be processed (problematic: 1, others: 0)
- Columns:
- [BioShifts.csv]: The raw bioShifts dataset downloaded from https://doi.org/10.6084/m9.figshare.7413365.v1
- [CHELSA_2.1.2011-2020.malr.csv]: MALRT from 2011-2020 derived from CHELSA dataset; scaling factor: 100; unit: c/km; NA: -9999
- [CHELSA_2.1.2011-2020.tmp.csv]: averaged temperature from 2011-2020 derived from CHELSA dataset; scaling factor: 10; unit: K
- [CHELSA_2.1.2011-2020.vap.csv]: averaged water vapour from 2011-2020 derived from CHELSA dataset; unit: Pa; NA: -9999
- [cru_ts4.05.1971-1980.tmp.csv]: averaged temperature from 1971-1980 derived from CRU dataset; scaling factor: 10; unit: C; NA: -999
- [cru_ts4.05.2011-2020.malr_mt.csv]: MALRT in global mountain regions from 2011-2020 derived from CRU dataset; scaling factor: 100; unit: c/km; NA: -9999
- [cru_ts4.05.2011-2020.tmp.csv]: averaged temperature from 2011-2020 derived from CRU dataset; scaling factor: 10; unit: C; NA: -999
- [cru_ts4.05.2011-2020.tmp_mt.csv]: averaged temperature in global mountain regions from 2011-2020 derived from CRU dataset; scaling factor: 10; unit: C; NA: -9999
- [cru_ts4.05.2011-2020.vap_mt.csv]: averaged water vapour in global mountain regions from 2011-2020 derived from CRU dataset; scaling factor: 10; unit: hPa; NA: -9999
- [cru_ts4.05.Climate_velocity.1975_VS_2015_mt-norm.csv]: normalized climate velocity in global mountain regions (1970-1980 vs 2011-2020) based on MALRT; unit: Z-score; NA: -9999
- [cru_ts4.05.Climate_velocity.1975_VS_2015_mt.csv]: climate velocity in global mountain regions (1970-1980 vs 2011-2020) based on MALRT; unit: m/yr; NA: -9999
- [cru_ts4.05.MOD11C3.Climate_velocity.1975_VS_2015_mt-norm.csv]: normalized climate velocity in global mountain regions (1970-1980 vs 2011-2020) based on SLRT; unit: Z-score; NA: -9999
- [cru_ts4.05.MOD11C3.Climate_velocity.1975_VS_2015_mt.csv]: climate velocity in global mountain regions (1970-1980 vs 2011-2020) based on SLRT; unit: m/yr; NA: -9999
- [cru_ts4_1975_VS_2015_climvelocity_to_const_lp_5.5_mt-norm.csv]: normalized climate velocity in global mountain regions (1970-1980 vs 2011-2020) based on constant lapse rate of 5.5C/km; unit: Z-score; NA: -9999
- [cru_ts4_1975_VS_2015_climvelocity_to_const_lp_5.5_mt.csv]: normalized climate velocity in global mountain regions (1970-1980 vs 2011-2020) based on constant lapse rate of 5.5C/km; unit: m/yr; NA: -9999
- [GHCN-M_Google_cloud-yr_2011-2019.csv]: GHCN dataset derived from files in 'weather_station_data'; scaling factor: 100; unit: C; NA: -9999 or NA
- Columns:
- 'value*' represents the data for a certain month
- Columns:
- [GHCN-M_Google_cloud_station_info.csv]: Geographic data for each station in GHCN
- [grid.lr.dat.rm.4.percent.outliers.csv]: Dataset with 2% SLRT outliers in both tails removed from the complete dataset of climatic and geographic factors ('variable_table_mt.csv')
- [high_climV_region_MALR_n_MOD11C3-cut_off-0.01_endanger-high_res.kmz]: The ultimate high climate velocity regions based on both MALRT and SLRT (with removal of 1% SLRT outliers from both tails); Google Earth file; also provided as Supplementary Data 3 in Chan et al., 2024
[high_climV_region_MALR_n_MOD11C3-env_var_summary.csv]: Summarized climatic information for the 17 identified high climate velocity global mountain regions; also provided as Supplementary Data 1 in Chan et al., 2024;
- Columns:
- cld: cloud coverage (scaling factor, 10; unit: %)
- vap: water vapour (unit: Pa)
- Tmp.rate: temperature rate (unit: C/yr)
- malr: MALRT (scaling factor, 100; unit: C/km)
- climV: climate velocity (unit: m/yr)
- MOD11C3.A2011.2020.lr: SLRT (scaling factor, 100; unit: C/km)
- MOD11C3.climV: climate velocity (SLRT; unit: m/yr)
- CRU2.elev: elevation (unit: m)
- sd: standard deviation (unit: following each variable)
- n.grids: sample size (unit: grid)
- Columns:
[high_climV_region_MALR_n_MOD11C3-simple-labeled.csv]: Regions included in the high climate velocity global mountain regions (considering both MALRT and SLRT), labeled according to 'high_climV_region_MALR_n_MOD11C3-simple-labeled_dictionary.csv'; NA: 0
[high_climV_region_MALR_n_MOD11C3-simple-labeled_dictionary.csv]: the corresponding dictionary indicating values labeled in 'high_climV_region_MALR_n_MOD11C3-simple-labeled.csv'
[high_climV_region_MALR_n_MOD11C3-simple.csv]: Regions included in the high climate velocity global mountain regions, labeled by 1; NA: 0
[MOD11C3.A2011-2020_all-MALR_regional_sampling(15-19_09-05-2021).csv]: SLRT from 2011-2020 (integrating day and night satellite data); NA: -9999 or -99.99
[MOD11C3.A2011-2020_day-MALR_regional_sampling(22-49_09-04-2021).csv]: day-time SLRT from 2011-2020 (byproduct; not included in the final manuscript); NA: -9999 or -99.99
[MOD11C3.A2011-2020_night-MALR_regional_sampling(06-57_09-05-2021).csv]: night-time SLRT from 2011-2020 (byproduct; not included in the final manuscript); NA: -9999 or -99.99
- Columns (applicable to the three files above):
- y_id: row number
- x_id: column number
- lp_median: median of SLRT (unit: C/km)
- lp_iqr: interquartile range of SLRT (unit: C/km)
- lp_mean: mean of SLRT (unit: C/km)
- lp_sd: standard deviation of SLRT (unit: C/km)
- lp_r2_avg: averaged R-squared for multiple transects in a grid (unit: C/km)
- lp_r2_sd: standard deviation of R-squared for multiple transects in a grid (unit: C/km)
- wilcoxon_p-log: log(p-value) by comparing SLRT with the corresponding MALRT (Wilcoxon test)
- t-test_p-log: log(p-value) by comparing SLRT with the corresponding MALRT (t-test)
- available_transects: number of available transects in a grid
- MALR_CRU: MALRT derived from CRU data (unit: C/km)
- elevation_CRU: elevation used in CRU datasets (unit: m)
- elevation_SRTM_avg: average elevation derived from SRTM (unit: m)
- elevation_SRTM_sd: standard deviation of elevation derived from SRTM (unit: m)
- elevation_SRTM_rng: elevational range derived from SRTM (unit: m)
- Columns (applicable to the three files above):
[MOD11C3_A2011-2020_all-MALR_regional-lp_iqr_mt.csv]: interquartile range of SLRT derived in each grid of global mountain regions from 2011-2020; scaling factor: 100; unit: C/km; NA: -9999 (byproduct)
[MOD11C3_A2011-2020_all-MALR_regional-lp_mean_mt.csv]: mean of SLRT derived in each grid of global mountain regions from 2011-2020; scaling factor: 100; unit: C/km; NA: -9999 (byproduct)
[MOD11C3_A2011-2020_all-MALR_regional-lp_median_mt.csv]: median of SLRT derived in each grid of global mountain regions from 2011-2020; scaling factor: 100; unit: C/km; NA: -9999
[MOD11C3_A2011-2020_all-MALR_regional-lp_sd_mt.csv]: standard deviation of SLRT derived in each grid of global mountain regions from 2011-2020; scaling factor: 100; unit: C/km; NA: -9999 (byproduct)
[mt_weather_station_list_9_grids.csv]: selected GHCN weather stations in global mountain regions (Find Methods in Chan et al., 2024 for details)
Columns:
- id: GHCN ID
- latitude: (unit: degree)
- longitude: (unit: degree)
- stnelev: elevation (unit: m)
- name: station name
- grelev: the elevation of the corresponding 0.5 x 0.5-degree grid cell (unit: m)
- stveg: vegetation coverage; refer to the original files in the 'weather_station_data' folder
- row.id: row number of the grid location on the MALRT map
- col.id: column number of the grid location on the MALRT map
- lat.l: the southern boundary of the study window (unit: degree)
- lat.h: the northern boundary of the study window (unit: degree)
- lon.l: the western boundary of the study window (unit: degree)
- lon.h: the eastern boundary of the study window (unit: degree)
- ws.grid.id: a unique id assigned to the selected stations
[mts_GMBA_ras_map.csv]: raster format of GMBA dataset; regions included in the GMBA are labeled by 1; NA: -9999
[range_shift_dataset_Dec2021.csv]: integration of filtered BioShifts dataset with climatic and geographic information.
- Columns:
- climatic variables are specified in the Key Abbreviation section; the year span was indicated in each column name
- geographic variables are provided in the section describing the 'variable_maps' folder
- biological variables can be found in 'BioShifts.csv'
- Columns:
[References.csv]: corresponding references for the BioShifts dataset; note that the reference numbers are independent of Chan et al., 2024
[Study_Areas_terra_small_merge_centroid.csv]: centroids of each filtered studied region in the BioShifts dataset
- Columns:
- X: longitude (unit: degree)
- Y: latitude (unit: degree)
- Columns:
[variable_table_mt.csv]: complete dataset of climatic and geographic factors
- Columns:
- climatic variables are specified in the Key Abbreviation section; the year span was indicated in each column name
- geographic variables are provided in the section describing the 'variable_maps' folder
- Columns:
[weather_station_LRT_df.csv]: detailed statistical results for the LRT based on weather station data
- Columns:
- Estimate: lapse rate of temperature (unit: C/km)
- Std. Error: standard error
- t value: t-value
- Pr(>|t|): p-value
- adj.r.2: adjusted R-squared value
- d.f: degrees of freedom
- row.id: row number of the grid location on the MALRT map
- col.id: column number of the grid location on the MALRT map
- elev.min: the elevation of the lowest station included in the analysis (unit: m)
- elev.max: the elevation of the highest station included in the analysis (unit: m)
- elev.mean: the average elevation of all stations included in the analysis (unit: m)
- elev.median: the median elevation of all stations included in the analysis (unit: m)
- malrt: corresponding MALRT (unit: C/km)
- salrt: corresponding SLRT (unit: C/km)
- sig.level: p-value smaller than 0.05: 1; others: 0
- elev.rng: ele.max - elev.min (unit: m)
- lat.idx: id preserved for developmental purpose
- lon.idx: id preserved for developmental purpose
- Columns:
[wilcoxon_test (different sample size)-p_xxxxxx_2022Jan.csv]: series of dataset derived under different levels of p-value (please find Methods and Extended Data Fig. 6 in Chan et al., 2024 for details). These are intended for use in figures production and may not be directly interpretable by humans. Some are byproducts.
'scripts' folder
Contains twenty three executable scripts, each with annotations for understanding and replication purposes.
Sharing/Access information
Detailed descriptions of data sources are provided in Extended Data Table 1 and the Data Availability section in Chan et al., 2024.
Code/Software
archived in the 'scripts' folder
Software:
Python 3.6.10 & 3.7.9 (with jupyter notebook environment)
Wolfram Mathematica 12
R 4.04
(The software applications are compatible with MacOS, Windows, and Linux operating systems, making them accessible across various environments. To download and install the software, users can simply follow the instructions provided by the software providers. The installation time may vary based on the hardware setup of each individual system.)
(Suggested hardware specification: RAM > 16gb; suggested storage > 16gb)
The necessary Python and R packages are directly listed within each script. Users has to install these packages before running the script.
- Execute [climate_calculate_mean.ipynb] to derive annual, 5-year, and decadal means (not provided in the datasets) [2-4 hr]
- Execute [CRU_MALRT.ipynb] to dervie the moist adiabatic lapse rate (MALRT) for a given period of time [< 1 hr]
- Execute [MODIS_img_yr_avg.ipynb] followed by [MODIS_img_period_avg.ipynb] to generate averaged land surface temperature (LST) datasets for a given period [1-2 hr]
- Execute [MOD11C3_ALRT.ipynb] based on all required data in the folder
and a folder of MOD11C3_006 LST data [4-8 hr] Note: Since MOD11C3_006 has three sub-dimensions (all, day, and night), three result datasets will be generated. The analyses in this study is based on all LST. - Execute [table_to_map_mt_regions.nb] using the file
, which derived from GMBA mountain inventory v1.2, to generate maps (0.5-degree resolution) in csv format. [< 10 min] - Execute [To_rate_and_climate_velocity.nb] to calculate climate velocity from temperature difference and ALRTs. [< 10 min]
Note: All environmental datasets are provided in geoTIFF format with a 0.5-degree spatial resolution in folder
- Execute [all_maps_variables.nb] to generate a table of environmental variables where each pixel corresponds to a row [< 10 min]
- Execute [Env_analysis-1.R] to generate figures and perform analysis on the geographical and climatic aspects [< 1 hr]
- Execute [Env_Random-Forest.py] to evaluate the relationship between MALRT and SALRT [< 5 min for 100 iterations]
Range shift analysis
- Execute [integrate_area_info.nb] to generate a list of studied sites with area information [< 30 min]
- Execute [Bio_analysis-1.R] to create a range-shift dataset with corresponding environmental variables [< 1 hr]
- Execute [Bio_Random-Forest.py] for random forest analysis [< 1 hr]
- Execute [biological_data_random_forest_plot.R] to obtain statistical results from the random forest analysis [< 5 min for 100 iterations]
- Execute [Bio_analysis-2.R] to generate figures and perform analysis on the biological aspects [< 1 hr]
Additional analyses
- Execute [Env_analysis-1_add.R] to remove SALRT outliers based on different % of interest [< 10 min]
- Execute [SALRT_variation-n_transect.R] to see the interaction between number of transects and SALRT reliability [< 10 min]
- Execute [tmp_warming_rate_vs_lapse_rate.R] to see the elevational trend of temperature rate and MALRT [< 10 min]
Comparison MALRT with weather station data (GHCN)
- Execute [weather_station_data_comparison.R] to generate a dataset for comparison between MALRT and station-based LRT [< 10 min]
- Execute [weather_station_data_comparison.R] to plot the results [< 10 min]
Comparison CRU and CHELSA based MALRT
- Execute [down_scaling_geotiff.py] to systematically downscale the CHELSA data to 0.05d [2-4 hr]
- Execute [CHELSA_MALRT.ipynb] to dervie the moist adiabatic lapse rate (MALRT) for a given period of time [< 1 hr]
- Execute [comparison_between_CRU_n_CHELSA.R] to compare between CHELSA MALRT and CRU MALRT [< 10 min]
Methods
Please read the Methods in Chan et. al., 2024 for details.
Usage notes
Key datasets are in /data/MALRT_SLRT_Climate_Velocity
Software:
Python 3.6.10 & 3.7.9 (with jupyter notebook environment)
Wolfram Mathematica 12
R 4.04
(The software applications are compatible with MacOS, Windows, and Linux operating systems, making them accessible across various environments. To download and install the software, users can simply follow the instructions provided by the software providers. The installation time may vary based on the hardware setup of each individual system.)
(Suggested hardware specification: RAM > 16gb; suggested storage > 16gb)
The necessary Python and R packages are directly listed within each script. Users has to install these packages before running the script.