• Energy Research
• US close
• CN close
• CA close
• English close
• DRYAD close

Climate and vegetation change in a coastal marsh: two snapshots of groundwater dynamics and tidal flooding at Piermont Marsh, NY spanning 20 years We include water levels measured along a transect of groundwater wells in 1999 and 2019, statistical analyses of ground water data, tidal efficiency estimates, vegetation data from 1997, 2005, 2014, and 2018, measures of tide gauge data and sea level rise from the Battery, New York Harbor. We attach the following three groups of files: (1) Files related to data from Piermont Marsh, which includes water levels in wells, tide gauge data collected from the tidal channel, and vegetation data; (2) Files related to analysis of water levels at Piermont Marsh; (3) Files related to analysis of Battery tide gauge data, Battery tide predictions, and precipitation data ## Description of the data and file structure **(1) Files related to data from Piermont Marsh, which includes water levels in wells, tide gauge data collected from the tidal channel, and vegetation data** 1999PiermontWaterlevels.csv 2019PiermontWaterLevels.csv channel_1999.xls channel_2019.xls water_level_elevations.csv Vegetation.xls 1999PiermontWaterlevels.csv and 2019PiermontWaterLevels.csv - Water levels collected at Piermont marsh in groundwater wells, at 0-m, 6-m, 12-m, 18-m, 24-m, 36-m, and 48-m from a tidal channel. The files contain three fields: daytime, well, and elevation. The daytime is the date and time the water level was collected, hours in Eastern Daylight Time -4GMT. The well number refers to its location relative to the tidal channel, with #1 referring to 0-m, #2 referring to 6-m, #3 referring to 12-m, #4 referring to 18-m, #5 referring to 24-m, #6 referring to 36-m, and #7 referring to 48-m. The elevation field refers to the water level in meters relative to the NAVD88 datum. In 1999 water levels were collected 14 April 2019 - 26 May 2019. In 2019, water levels were collected 5 May 2019 - 30 June 2019. channel_1999.xls - This file shows the elevation of water level in the channel. There is a field for date and time, in GMT -4, and water level in meters relative to NGVD29. channel_2019.xls - This file shows the elevation of water level in the channel. There is a field for Date, Time, in GMT -4, absolute pressure in in mbar, temperature in degrees C, and water level in meters relative to NAVD88. water_level_elevations.csv - This csv file includes five fields. The first is "year" or the year collected (1999 or 2019). The second is "well" numbered 1-7. Well 1 is closest to the channel while 7 is the furthest from the channel. #1 referrs to 0-m from the channel, #2 referring to 6-m from the channel, #3 referring to 12-m from the channel, #4 referring to 18-m from the channel, #5 referring to 24-m from the channel, #6 referring to 36-m from the channel, and #7 referring to 48-m from the channel. The datetime field refers to the day and time the measure was made in day/month/year HH:MM AM/PM format. The next field is lunarcyle which refers to whether the measure was made during "spring" or "neap" tidal cycles. Spring was assigned to the tides the week of full or new moons, Neap was assigned to tides the week of the first and last quarter. The last is "elevation" and is the measure of water levels in meters relative to the NAVD88 datum. Vegetation.xls - This Excel file includes four sheets that each refer to a year of vegetation date - 1997, 2005, 2014, and 2017. The first field is "well" which has a number 1 through 7. The well number refers to its location relative to the tidal channel, with #1 referring to 0-m, #2 referring to 6-m, #3 referring to 12-m, #4 referring to 18-m, #5 referring to 24-m, #6 referring to 36-m, and #7 referring to 48-m. There is a field for latitude (lat) and longitude (long), which refers to the location of the shape in UTM, in meters, in the 18N. Cover refers to the plant cover type, area is the area of the polygon in square meters. **(2) Files related to analysis of water levels at Piermont Marsh** Distancefromsurface.R MinNeap_MarshSurface.csv MaxNeap_MarshSurface.csv MinSpring_MarshSurface.csv MaxSpring_MarshSurface.csv PiermontEfficiencyRggplot.csv Tidalefficiency.R The R file Distancefromsurface.R includes calculations of mean and variance of water levels, and as well as production of relevant figures. MinNeap_MarshSurface.csv file has low tide minimum water levels during neap tides (weeks centered on the moons first and third quarter). It includes the following fields: distance, year, water_elevation, marsh_elevation, and distance_surface. The field distance, is distance from the tidal channel, in meters. The field year, refers to is the year collected (1999 or 2019). The field water_elevation, is the elevation of the water level at low tide, in meters relative to the NGVD88 datum. The field marsh_elevation refers to the height of the marsh at that location, in meters relative to the NGVD88 datum. The field distance_surface is the difference between the marsh elevation and the water elevation. Positive values are values below the marsh surface, while negative values are values above the marsh surface. MaxNeap_MarshSurface.csv file has high tide maximum water levels during neap tides (weeks centered on the moons first and third quarter). It includes the following fields: distance, year, water_elevation, marsh_elevation, and distance_surface. The field distance, is distance from the tidal channel, in meters. The field year, refers to is the year collected (1999 or 2019). The field water_elevation, is the elevation of the water level at high tide, in meters relative to the NGVD88 datum. The field marsh_elevation refers to the height of the marsh at that location, in meters relative to the NGVD88 datum. The field distance_surface is the difference between the marsh elevation and the water elevation. Positive values are values below the marsh surface, while negative values are values above the marsh surface. MinSpring_MarshSurface.csv file has low tide minimum water levels during spring tides (weeks centered on the new and full moon). It includes the following fields: distance, year, water_elevation, marsh_elevation, and distance_surface. The field distance, is distance from the tidal channel, in meters. The field year, refers to is the year collected (1999 or 2019). The field water_elevation, is the elevation of the water level at low tide, in meters relative to the NGVD88 datum. The field marsh_elevation refers to the height of the marsh at that location, in meters relative to the NGVD88 datum. The field distance_surface is the difference between the marsh elevation and the water elevation. Positive values are values below the marsh surface, while negative values are values above the marsh surface. MaxSpring_MarshSurface.csv file has high tide maximum water levels during spring tides (weeks centered on the new and full moon). It includes the following fields: distance, year, water_elevation, marsh_elevation, and distance_surface. The field distance, is distance from the tidal channel, in meters. The field year, refers to is the year collected (1999 or 2019). The field water_elevation, is the elevation of the water level at high tide, in meters relative to the NGVD88 datum. The field marsh_elevation refers to the height of the marsh at that location, in meters relative to the NGVD88 datum. The field distance_surface is the difference between the marsh elevation and the water elevation. Positive values are values below the marsh surface, while negative values are values above the marsh surface. PiermontEfficiencyRggplot.csv - file lists the well number (1-7), distance (a number 1-14, which gives a unique identifier to each combination of well and year), year, which was the year the data was collected. The last field is efficiency. This field refers to the ratio between the change in water level over the course of a tidal cycle in the well to the change in the water level over the course of the tidal cycle at the Battery tide gauge, NYC. Tidalefficiency.R - file that plots and calculates tidal efficiency during 1999 and 2019 at each well. **(3) Files related to analysis of Battery tide gauge data, Battery tide predictions, and precipitation data** MSL_time.R 3348871.csv 3348873.csv Battery.csv Bat_wls.csv monthly.csv sin2.csv predictions.csv tide_l.csv wls.csv MSL_time.R - This R code uses several data files to conduct analysis of change over time in water levels and monthly anomalies in precipitation and water levels. All necessary packages are described. 3348871.csv and 3348873.csv - are weather data from Westchester County airport, station USW00094745 from 1997 to 2001 (3348873.csv) 2017 to 2022 (3348871.csv). The field station lists the station. The field Name is the name of the station, Westchester County Airport. The date is the day data was collected. AWND refers to Average daily wind speed in miles per hour. PGTM refers to peak gust time (hours and minutes, i.e., HHMM). PRCP refers to precipitation in inches, TMAX refers to the maximum daily temperature, in degrees Fahrenheit. TMIN refers to the minimum daily temperature, in degrees Fahrenheit. WDF2 is the direction of fastest 2-minute wind in degrees. WDF5 is the direction of fastest 5-second wind in degrees. WSF2 is the fastest 2-minute wind speed in miles per hour. WSF5 is the fastest 5-second wind speed in miles per hour. Missing data is replaced with -999. Battery.csv - all high tide levels for 1997 through 2022. The two fields are level, referring to high tide water levels in meters relative to the NAVD88 datum. The second field is year. Bat_wls.csv is monthly tide levels from the Battery tide gauge, NY. The year field refers to year including fraction. Mean high water (MHW) refers to monthly mean high water relative to the NAVD88 datum in meters. Mean sea level (MSL) refers to monthly mean sea level relative to the NAVD88 datum in meters. Mean tide level (MTL) refers to monthly mean tide level relative to the NAVD88 datum in meters.. Mean Low Water (MLW) refers to monthly mean low water relative to the NAVD88 datum in meters. monthly.csv - is mean high water and mean sea level from 1980-2022, by month. The field month refers to the month (January =1). MHW is monthly mean high water for all months, relative to the NAVD88 datum, and MSL is monthly mean sea level relative to the NAVD88 datum. sin2.csv is the monthly mean sea level at the Battery tide gauge (1980-2022), with a 1 year rolling window median smooth added. There are three fields, month, MSL, and year. Month is the number of months elapsed since January 1961. MSL is the monthly mean sea level in meters, relative to the NAVD88 datum, with a one year smoothing function applied. Year refers to the observation month, expressed in years and the fraction of years so January 1980 would be 1980, while February 1980 is depicted as 1980.083. predictions.csv - tide predictions for the Battery tide gauge, New York City. Fields are y, which stands for year, represented by year, including fractions representing months. High_p is the highest predicted tide of the month, in meters relative to the NAVD88 datum. MHW_p is the predicted mean high tide for the month relative to the NAVD88 datum. MLW_p is the predicted mean low tide for the month relative to the NAVD88 datum. MTL_p is the predicted mean tide level for the month relative to the NAVD88 datum. High_1 is the highest actual tide of the month, in meters relative to the NAVD88 datum. MHW_a is the actual mean high tide for the month relative to the NAVD88 datum. MLW_a is the actual mean low tide for the month relative to the NAVD88 datum. MTL_a is the actual mean tide level for the month relative to the NAVD88 datum. tide_l.csv is a file with the monthly mean high water (MHW_l), monthly mean tide level (MTL_l), and mean low water (MLW_l) for 1960 -2021. wls.csv is a file that has monthly water levels from 1999 to 2019, listing year (as a fraction, not just an integer for month), Highest, as the highest tide of the month in meters relative to the NAVD88 datum. MHW refers to the mean high water during the month in meters relative to the NAVD88 datum. MTL refers to the mean tidal level during the month in meters relative to the NAVD88 datum. MLW refers to the mean low water during the month in meters relative to the NAVD88 datum. ## Sharing/Access information Data was derived from the following external sources: * Vegetation shapefiles for the Hudson River NERR for 1997, 2005, and 2014, were obtained through personal request to Sarah Fernald, *Reserve Manager and Research Coordinator.* Files should be available through the Reserve website, although the link is not functional at this time: * The 2018 vegetation shapefiles were obtained from under the heading, [Hudson River Estuary tidal wetlands](https://data.gis.ny.gov/datasets/ee2723393f894e929dbd6dbdc84770de_0/explore?location=41.308770%2C-73.842410%2C9.10). * We acknowledge the NYS DEC Hudson River Estuary Program, NYS DEC Hudson River National Estuarine Research Reserve, and Cornell Institute for Resource Information Sciences for collection and curation of the Hudson River NERR vegetation data. * Tide gauge data and tide predictions for the Battery, NY were obtained from NOAA tides and currents website: * Precipitation data was obtained from the National Centers for Environmental Information, NOAA: . The station for which data was obtained was the Westchester County airport, station USW00094745. ## Code/Software We provide three R files, which we ran using R version 4.3.1 (2023-06-16), in R Studio 2022.02.1, Build 461. All required packages are described in the .R files. Distancefromsurface.R - This R code utilizes four data files that include low tides during spring tides, low tides during neap tides, high tides during spring tides, and high tides during neap files to compare average and variance in low and high tide water levels during 1999 and 2019 relative to the marsh surface and relative to the NAVD88 datum. Code is also included to produce plots. Tidalefficiency.R - file that plots and calculates tidal efficiency during 1999 and 2019 at each well. MSL_time.R - This R code uses several data files to conduct analysis of change over time in water levels and monthly anomalies in precipitation and water levels. Hydrological measurements were collected during the spring and summer of 1999 and 2019 in Piermont Marsh (coordinates 41.0361°, -73.9105°). These measurements covered a transect that was laid out perpendicular to a tidal channel. The objective of this study was to compare the current tidal flooding and groundwater table levels with the data from 1999. The goal was to assess the differences in tidal hydrology between these two distinct time periods, which also differed in terms of marsh and water level elevations. To determine groundwater levels and tidal flooding across the marsh, we installed seven water level loggers along a gradient, ranging from the tidal channel to the upland area. We constructed wells by suspending pressure transducers within 7.5 cm diameter perforated PVC pipes lined with screening to prevent sediment from entering the well. These wells were positioned one meter below the marsh surface, 0.6 meters above the soil surface, vented to the atmosphere, and only the section below the soil surface was perforated. Additionally, we installed concrete collars at the marsh surface around the wells to prevent preferential water flow down the well sides. These seven wells were placed along the original transect, perpendicular to the creek, with increasing distances (0 meters, 6 meters, 12 meters, 18 meters, 24 meters, 36 meters, and 48 meters). We installed and monitored the wells from May 5 to June 30, 2019, and from April 6 to May 26, 1999. In 2019, we measured the absolute elevation of the top of each well using RTK-enabled static GPS measurements from Leica GNSS GS14 rover units and static measurements with an AX1202 GG base station unit to reference water levels to the NAVD88 vertical datum. We measured reference water levels each time data was collected, which involved determining the distance from the top of the well to the water surface and converting it to elevation relative to the NAVD88 datum. To relate marsh elevation to water elevations, GPS surveys were conducted along the transect using a Leica GNSS GS14 rover unit. In 1999, elevation control for the wells and water levels was similarly measured using survey-grade GPS. We compared changes in the marsh water table with significant potential hydrological and vegetation changes that have occurred over the past 20 years. We calculated the rates of change in monthly water levels at Battery, NY for the period from 1999 to 2019 using two different methods. We modeled changes over time in monthly highest water levels, mean high water (MHW), mean tide level (MTL), and mean low water (MLW) using an ordinary least squares regression model with ARIMA errors to account for the autoregressive structure of tide data. We removed the annual cycle first using a curve with a 1-year periodicity. The ARIMA errors model was fitted using the "auto.arima" function from the "forecast" package. We calculated the squared correlation of fitted values to actual values to produce a pseudo-r2. For comparison, we calculated trends using ordinary least squares regression for the 1999-2019 period, although it's important to note that the temporal autocorrelation likely results in underestimated uncertainty. We obtained vegetation maps from the HRNERR for 1997, 2005, 2014, and 2018 to help assess changes in the coverage of plant species over time, as these changes could impact evapotranspiration and water table patterns. A 20-meter buffer zone was created around each well location, and the composition of vegetation within this buffer zone was quantified using QGIS version 3.30.2. While four time-points may not be sufficient for statistically identifying trends, we analyzed the changes observed. To put the measurement time periods in context and ensure that our selected seasons were not anomalous, we compared water levels in spring 1999 and 2019 relative to the astronomical cycles driving interannual sea level variability using data from the Battery tide gauge. We also compared spring high tide levels in 1999 and 2019 with surrounding years. The main astronomical cycles thought to influence tides include the 18.6-year lunar nodal cycle and the 4.4-year subharmonic of the 8.85-year lunar perigee cycle. As our 1999 and 2019 measurements were collected during slightly different time periods (April/May 1999 vs. May/June 2019), we also examined mean monthly water levels (1980-2022) from the NOAA Battery tidal gauge to identify potential artifacts. We obtained rainfall data from spring 1999 and 2019 from the nearest precipitation monitoring station (Westchester airport) to determine whether the measurements were made during an unusually wet or dry period. The sampling periods were 20 years apart, so they occurred at approximately the same point in the 18.6-year lunar nodal cycle. Pressure transducer data was processed using HOBOware Pro (Version 3.7.16, Onset Computer Corporation, Bourne, MA) with reference water levels collected in the field. The data were corrected for atmospheric pressure using the HOBOware barometric compensation assistant, using data from the Hudson River National Estuarine Research Reserve. Raw water elevation data from 1999 was analyzed in conjunction with the 2019 data. Water level data from 1999 were converted from the NVGD29 to NAVD 88 datum using NOAA VDatum v4.0.1 prior to analysis. Well seven's transducer experienced three brief malfunctions from May 30 to June 3, 2019, resulting in inaccurate elevation measurements for a total of 19.5 hours. These data were excluded from the analysis. In 1999, well seven also experienced malfunctions, which were corrected by Montalto into smoothed six-hour increments using average water elevation measurements and calculated error, calibrated using regression. No other well transducers appeared to have malfunctioned. Groundwater hydrology plays an important role in coastal marsh biogeochemical function, in part because groundwater dynamics drive the zonation of macrophyte community distribution. Changes that occur over time, such as sea level rise and shifts in habitat structure are likely altering groundwater dynamics and eco-hydrological zonation. We examined tidal flooding and marsh water table dynamics in 1999 and 2019 and mapped shifts in plant distributions over time, at Piermont Marsh, a brackish tidal marsh located along the Hudson River Estuary near New York City. We found evidence that the marsh surface was flooded more frequently in 2019 than in 1999, and that tides were propagating further into the marsh in 2019, although marsh surface elevation gains were largely matching that of sea level rise. The changes in groundwater hydrology that we observed are likely due to the high tide rising at a rate that is greater than that of mean sea level. In addition, we reported on changes in plant cover by P. australis, which has displaced native marsh vegetation at Piermont Marsh. Although P. australis has increased in cover, wrack deposition and plant die off associated Superstorm Sandy allowed for native vegetation to rebound in part of our focus area. These results suggest that climate change and plant community composition may interact to shape ecohydrologic zonation. Considering these results, we recommend that habitat models consider tidal range expansion and groundwater hydrology as metrics when predicting the impact of sea level rise on marsh resilience.

Although GPS coordinates for current populations are not included due to the potential threat of poaching, the climate variables for each species are provided. The records for extant gecko and skinks mainly came from the New Zealand's Department of Conervation Herpetofauna Database. After updating the taxonomy and cleaning the data to reflect the taxonomy as at 2019 of 43 geckos speceis recognised across seven genera and 61 species in genus, we then thinned the occurrence records at a 1 km resolution for all species then predicted distributions for those with > 15 records using species distribution models. The climate variables for each species were selected among annual mean temperature (bio1), maximum temperature of the warmest month (bio5), minimum temperature of the coldest month (bio6), mean temperature of driest quarter (bio9), mean temperature of wettest quarter (bio10), and precipitation of the driest quarter (bio17). To reduce multicollinearity in species distribution models for each species, we only retained climate variables with a variable inflation factor < 10. The climate variables were from the CHELSA database (https://chelsa-climate.org/), which can be freely downloaded for current and future scenarios. We also provide MCC tree files for the geckos and skinks. The phylogenetic trees have been constructed for NZ geckos by (Nielsen et al., 2011) and for NZ skinks by (Chapple et al., 2009). For geckos we used a subset of the sequences used by Nielsen et al. (2011) for four genes, two nuclear (RAG 1, PDC) and two mitochondrial (16S, ND2 along with flanking tRNA sequences). For skinks, we used sequences from Chapple et al. (2009) for one nuclear (RAG 1) and five mitochondrial (ND2, ND4, Cyt b, 12S and 16S) genes, and additional ND2 sequences for taxa not included in the original phylogeny (Chapple et al., 2011, p. 201). In total we used sequences for all recognised extant taxa (Hitchmough et al., 2016) as at 2019 except for three species of skink (O. aff. inconspicuum “Okuru”, O. robinsoni, and O. aff. inconspicuum “North Otago”) and two species of gecko (M. “Cupola” and W. “Kaikouras”) for which genetic data were not available. Aim: The primary drivers of species and population extirpations have been habitat loss, overexploitation, and invasive species, but human-mediated climate change is expected to be a major driver in future. To minimise biodiversity loss, conservation managers should identify species vulnerable to climate change and prioritise their protection. Here, we estimate climatic suitability for two speciose taxonomic groups, then use phylogenetic analyses to assess vulnerability to climate change. Location: Aotearoa New Zealand (NZ) Taxa: NZ lizards: diplodactylid geckos and eugongylinae skinks Methods: We built correlative species distribution models (SDMs) for NZ geckos and skinks to estimate climatic suitability under current climate and 2070 future-climate scenarios. We then used Bayesian phylogenetic mixed models (BPMMs) to assess vulnerability for both groups with predictor variables for life history traits (body size and activity phase) and current distribution (elevation and latitude). We explored two scenarios: an unlimited dispersal scenario, where projections track climate, and a no-dispersal scenario, where projections are restricted to areas currently identified as suitable. Results: SDMs projected vulnerability to climate change for most modelled lizards. For species’ ranges projected to decline in climatically suitable areas, average decreases were between 42–45% for geckos and 33–91% for skinks, although area did increase or remain stable for a minority of species. For the no-dispersal scenario, the average decrease for geckos was 37–52% and for skinks was 33–52%. Our BPMMs showed phylogenetic signal in climate change vulnerability for both groups, with elevation increasing vulnerability for geckos, and body size reducing vulnerability for skinks. Main conclusions: NZ lizards showed variable vulnerability to climate change, with most species’ ranges predicted to decrease. For species whose suitable climatic space is projected to disappear from within their current range, managed relocation could be considered to establish populations in regions that will be suitable under future climates.

From September of 2017, till August of 2019, water temperatures and A. aegypti larval presence was recorded in nine 19 liter buckets placed in the backyard of Jon Harrison’s home in Tempe, Arizona (33.339, -111.924), as it was known to experience high abundances of A. aegypti. Buckets were 5 – 10 m apart, and so should not be considered ecologically independent. Onset HOBO Pendant® UA-002-08 data loggers (Bourne, Massachusetts) were used to record temperature levels, and larval presence was observed every 1-10 days depending on season (frequently in the summer, less so in winter). If mosquito larvae were observed, they were collected from the bucket with a net and their species identity confirmed with a dissection scope. The data set labeled Figure 2 data provides the water temperatures in one representative bucket from 2017-2019 as shown in Fig. 2 of the manuscript. Larval rearing for mesocosm experiments The parents of larvae used in the mesocosm overwintering experiments were reared from Maricopa County, AZ, origin eggs collected by Maricopa County Vector Control from September to November of 2019. These eggs were placed in a 500 ml beaker, submerged, and hatched in a solution of 0.25 g/L baker’s yeast (Byttebier et al. 2014). As the 1st instar larvae emerged, they were fed TetraMin fish flakes every 1-2 days, making sure that an excess amount of food was visible in the container. The rearing density for the larvae was maintained at fewer than 500 animals per liter of water. As pupae began to appear, the beaker of larvae was placed in a 95-liter polymer-screened cage to contain the expected adults. Cotton balls saturated with 10% sucrose solution were made available for the adults as they began to emerge; these were taken away two days prior to blood feeding. One week after emerging, the adults were blood-fed using mice (IACUC protocol: 18-1662R). After a three-day gestation period, the females were supplied with moist seed-germinating paper to encourage oviposition. Once the females had finished ovipositing, the eggs were kept moist for an additional 48 hours before being dried, and placed in open zip lock sandwich bags which were stored at 100% humidity and 24°C. High humidity in the egg storage containers was achieved by storing damp paper towels along with the opened egg bags within a larger 3.8L bag. These eggs were kept for less than one month before the hatching procedure was repeated to produce the larvae for the experiment. In the lab, across all life stages, the mosquitoes were exposed to a 12:12 L/D photoperiod at 24°C. After hatching, the 2nd instar larvae were moved to their outdoor experimental mesocosms. The larvae were randomly distributed with 20 larvae supplied per each of three ambient mesocosms (Amb1, Amb2, Amb3) and six to warmed mesocosms (W1 – W6), which were warmed by varying amounts (W1 = least warmed, W6 = most warmed). The goal was to achieve a range of warming from very small warming (1-2°C in the least-warmed mesocosm (W1), to near-summer conditions in the most-warmed mesocosm (W6). Each mesocosm was a 150 ml clear plastic container, filled with 125 mL dechlorinated tap water. TetraMin fish flakes were supplied to each mesocosm, with more added every three days or when food was completely consumed. Although the mesocosms were open, we observed no mosquitoes flying in the field, and none were captured in local water buckets, and all A.a. in the mesocosms were of uniform stage, so we believe that this experiment was not affected by oviposition from wild mosquitoes. Manipulation of thermal conditions for larval outdoor rearing All mesocosms were placed on a table one meter above the ground and protected from rain, wind, and sunlight by a roof. The mesocosms were placed within individual lidless pine boxes (10x10x14 cm, 0.95 cm thick walls), and so were exposed to normal fluctuations in air temperature. Each warmed mesocosm was placed on 40mm2 thermoelectric plates with 40mm2 aluminum heatsinks attached using thermally conductive adhesive on each side. The warming orientation of the thermoelectric plate was positioned upwards, towards the mesocosms, to ensure adequate energy transfer from the heating units to the water. Each thermoelectric device was powered by two KORAD KD3005D 30V, 5A power supplies (Shenzhen, China). The thermoelectric plates were wired in parallel. Variable warming was produced by changing the supplied voltage. Temperatures were measured in the cups using HOBO Pendant® UA-002-08 data loggers submerged in the center of each cup. We did not measure temperature gradients within the mesocosms, but believe that they are likely to be small except possibly in the mesocosms that were maximally-warmed, as the mesocosms were small and mostly not strongly warmed above air temperature. Temperatures were logged each hour in each warmed mesocosm, and in one ambient treatment mesocosm. The data file labeled Figure 3 data provides the wate temperatures at hourly intervals during the experiment for one mesocosm at ambient temperature, mesocosm W1 (the least warmed mesocosm) and mesocosm W6 (the most warmed mesocosm) as shown in Fig. 3 of the manuscript. Global warming trends, human-assisted transport, and urbanization have allowed poleward expansion of many tropical vector species, but the specific mechanisms responsible for thermal mediation of range changes and ecological success of invaders remain poorly understood. Aedes aegypti (Diptera: Culicidae) is a tropical mosquito currently expanding into many higher-latitude regions including the urban desert region of Maricopa County, Arizona. Here, adult populations virtually disappear in winter and spring, and then increase exponentially through summer and fall, indicating that winter conditions remain a barrier to development of A. aegypti. To determine whether cold limits the winter development of A. aegypti larvae in Maricopa County, we surveyed for larval abundance, and tested their capacity to develop in ambient and warmed conditions. Aedes aegypti larvae were not observed in artificial aquatic habitats in winter and spring but were abundant in summer and fall, suggesting winter suppression of adults, larvae or both. Water temperatures in winter months fluctuated strongly; larvae were usually cold-paralyzed at night but active during the day. Despite daytime temperatures that allowed activity, larvae reared under ambient winter conditions were unable to develop to adulthood, perhaps due to repetitive cold damage. However, warming average temperature by 1.7°C allowed many larvae to successfully develop to adults. Because daytime highs in winter will often allow adult flight, it is possible that relatively minor additional winter warming may allow A. aegypti populations to develop and reproduce year-round in Maricopa County. # Data for Mesocosm studies suggest climate change may release Aedes aegypti (Diptera:Culicidae) larvae from cold-inhibition and enable year-round development in a desert city [https://doi.org/10.5061/dryad.nzs7h44z7](https://doi.org/10.5061/dryad.nzs7h44z7) Most of the data for this study are provided as supplementary files in the submitted manuscript. Here we provide representative thermal data. One file (Figure 2 data) contains the temperature data for the bucket kept under ambient conditions as shown in Figure 2, which also shows when Aedes aegypti larvae were found in the bucket. From to October 18 -November 29 2017, water temperatures were recorded every 6 minutes. Thereafter, water temperatures were recorded hourly until August 2, 2019. Another file (Figure 3 data) contains water temperatures for three of the mesocosms used in this study, as shown in the manuscript figure 3. This experiment ran from Jan 31, 2020 - March 1, 2020. One column sW1 was and ## Description of the data and file structure Figure 2 data has two columns, column A gives the date and column B the temperature of the ambient bucket in degrees Centigrade. Figure 3 data has four columns; column A gives the hours since the start of the experiment. Column B shows temperatures for an unheated mesocosm kept at ambient conditions. Column C shows temperatures for W6, the most warmed mesocosm (mean temperature 12C higher than the ambient mesocosm, to represent near-summer conditions). Column D shows temperatures for the least-warmed mesocosm (W1, mean temperature 1.8C higher than the ambient mesocosm). All temperatures are in degrees Centigrade.

The atmosphere concentration of CO2 is steadily increasing and causing climate change. To achieve the Paris 1.5 or 2 oC target, negative emissions technologies must be deployed in addition to reducing carbon emissions. The ocean is a large carbon sink but the potential of marine primary producers to contribute to carbon neutrality remains unclear. Here we review the alterations to carbon capture and sequestration of marine primary producers (including traditional ‘blue carbon’ plants, microalgae, and macroalgae) in the Anthropocene, and, for the first time, assess and compare the potential of various marine primary producers to carbon neutrality and climate change mitigation via biogeoengineering approaches. The contributions of marine primary producers to carbon sequestration have been decreasing in the Anthropocene due to the decrease in biomass driven by direct anthropogenic activities and climate change. The potential of blue carbon plants (mangroves, saltmarshes, and seagrasses) is limited by the available areas for their revegetation. Microalgae appear to have a large potential due to their ubiquity but how to enhance their carbon sequestration efficiency is very complex and uncertain. On the other hand, macroalgae can play an essential role in mitigating climate change through extensive offshore cultivation due to higher carbon sequestration capacity and substantial available areas. This approach seems both technically and economically feasible due to the development of offshore aquaculture and a well-established market for macroalgal products. Synthesis and applications: This paper provides new insights and suggests promising directions for utilizing marine primary producers to achieve the Paris temperature target. We propose that macroalgae cultivation can play an essential role in attaining carbon neutrality and climate change mitigation, although its ecological impacts need to be assessed further. To calculate the parameters presented in Table 1, the relevant keywords "mangroves, salt marshes, macroalgae, microalgae, global area, net primary productivity, CO2 sequestration" were searched through the ISI Web of Science and Google Scholar in July 2021. Recent data published after 2010 were collected and used since area and productivity of plants change with decade. For data with limited availability, such as net primary productivity (NPP) of seagrasses and global area and NPP of wild macroalgae, data collection was extended back to 1980. Total NPP and CO2 sequestration for mangroves, salt marshes, seagrasses and wild macroalgae were obtained by the multiplication of area and NPP/CO2 sequestration density and subjected to error propagation analysis. Data were expressed as means ± standard error.

# Data from: Eocene Shark Teeth from Peninsular Antarctica: Windows to Habitat Use and Paleoceanography. [https://doi.org/10.5061/dryad.qz612jmq2](https://doi.org/10.5061/dryad.qz612jmq2) The repository folder includes scripts and spreadsheets for phosphate oxygen stable isotope (δ18Op) analysis measured from shark tooth biogenic apatite collected from the Eocene deposits of the La Meseta and Submeseta formations (West Antarctica, Seymour Island). It also contains Fourier-Transform Infrared Spectroscopy (FTIR) analysis, a Bayesian model for temperature estimates, and model output extraction scripts from the iCESM simulation for the Early Eocene (Zhu et al., 2020). Scripts and data are stored in specific folders on the type of analysis. All scripts are in R or Python language. **Usage notes** **1 "iCESM modeling scripts" directory** The folder includes scripts in Jupiter Notebook format for extracting and plotting iCESM seawater outputs for the Eocene. The folder includes two files: 1) “d18Ow Analysis Script.ipynb” - This is a Python script primarily using the XArray library, to import iCESM output from Zhu et al. (2020), calculating δ18Ow, and reorganizing the output into monthly time intervals along 25 m and 115 m depth slices, while also averaging output down to these depths; 2) “NetCDF Plotting.ipynb” - this is a Python script primarily using the XArray, Matplotlib, and Cartopy libraries. The script writes a single callable function that creates Matplotlib contour plots from iCESM history output. Variables include temperature, salinity, ideal age, oxygen isotopes, and neodymium isotopes, and map projections include Plate Carree, Mollweide, and orthographic (centering on the Drake Passage). Options are built to enable scale normalization or to set maximum and minimum values for data and select colormaps from a predefined selection of Matplotlib’s “Spectral”, “Viridis”, “Coolwarm”, “GNUplot2”, “PiYG”, “RdYlBu”, and “RdYlGn”. For further questions on model output scripts, please email Adam Aleksinski at [aaleksin@purdue.edu](https://datadryad.org/stash/dataset/doi:10.5061/aaleksin@purdue.edu). **2 "d18O data and maps" directory** The folder includes δ18Op of shark tooth bioapatite and other datasets to interpret shark paleoecology. These datasets include: · δ18Op of shark tooth bioapatite (“shark FEST d18Op.csv”). Isotope measurements were run at the Stable Isotope Ecosystem Laboratory of (SIELO) University of California, Merced (California, USA). · Reference silver phosphate material δ18Op for analytical accuracy and precision (“TCEA reference materials.csv"). Isotope measurements were run at the Stable Isotope Ecosystem Laboratory of (SIELO) University of California, Merced (California, USA). · Bulk and serially sampled δ18Oc data of co-occurring bivalves (Ivany et al., 2008; Judd et al., 2019) (“Ivany et al. 2008_bulk.csv” and “Judd et al., 2019_serial sampling.csv"). · iCESM model temperature and δ18Ow outputs at 3x and 6x pre-industrial CO2 levels for the Early Eocene (Zhu et al., 2020) (“SpinupX3_25m_Mean_Monthly.nc”, “SpinupX6_25m_Mean_Monthly.nc.”, and “CA_x3CO2.csv”). Simulations are integrated from the surface to 25 m. · δ18O values of invertebrate species published in Longinelli (1965) and Longinelli & Nuti (1973), used to convert bulk δ18Oc (V-SMOW) data of bivalves into δ18Op (V-SMOW) values after δ18Oc (V-PDB) - δ18Oc (V-SMOW) conversion found in Kim et al. (2015) (“d18O carbonate and phosphate references.csv”). · R script for data analysis ("d18O data and maps.Rmd”). The script provides annotation through libraries, instrumental accuracy and precision tests, tables, statistical analysis, figures, and model output extractions. . ("TELM_diversity.csv") displays diversity trends of chondrichthyans across TELMs in one of the main figures of the manuscript. **2.1 Dataset description** **shark FEST d18Op.csv** · *Sample_ID*: Identification number of tooth specimens. · *Other_ID*: Temporary identification number of tooth specimens. · *Taxon*: Species assigned to shark tooth specimens. · *TELM*: Stratigraphic units of La Meseta (TELM 2-5; ~45 to ~37 Ma) and Submeseta formations (TELMs 6 and 7; ~37 to ~34 Ma) (Amenábar et al., 2020; Douglas et al., 2014; Montes et al., 2013). · *d18Op*: Mean δ18Op values of silver phosphate crystals precipitated from shark tooth bioapatite. Specimens were run in triplicates, corrected, and standardized on the V-SMOW scale. · *sd*: Standard deviation of silver phosphate triplicate samples per specimen. · *Protocol*: Silver phosphate protocols used to precipitate crystals from shark tooth bioapatite. We adopted the Rapid UC (“UC_Rapid”) and the SPORA (“SPORA”) protocols after Mine et al. and (2017) Larocca Conte et al. (2024) based on the tooth specimen size and sampling strategy. Descriptions of the methods are included in the main manuscript. · *Environment*: Inferred shark habitat based on taxonomy classified as benthic or pelagic environment. · *Collection*: Institutional abbreviations of museum collections from which shark tooth specimens are housed. NRM-PZ is the abbreviation for the Swedish Natural History Museum (Stockholm, Sweden), PRI is the abbreviation for the Paleontological Research Institute (Ithaca, New York, United States), and UCMP is the University of California Museum of Paleontology (Berkeley, California, United States). **TCEA reference materials.csv** · *Identifier_1*: unique identifier number per sample. · *sample*: reference silver phosphate materials (USGS 80 and USGS 81). · *amount*: weight of samples in mg. · *Area 28*: peak area of mass 28 (12C16O). · *Area 30*: peak area of mass 30 (12C18O). · *d18O_corrected*: corrected δ18Op value of reference materials following drift correction, linearity correction, and 2-point calibration to report values on the V-SMOW scale. **Ivany et al. 2008_bulk.csv** · *Telm*: Stratigraphic units of La Meseta (TELM 2-5; ~45 to ~37 Ma) and Submeseta formations (TELMs 6 and 7; ~37 to ~34 Ma) (Amenábar et al., 2020; Douglas et al., 2014; Montes et al., 2013). · *Locality*: Locality code from which bivalves were collected. · *Genus*: Genera of bivalves. Specimens are assigned to *Cucullaea* and *Eurhomalea* genera. · *Line*: Sampling areas of specimens. The sampling strategy is described in Ivany et al. (2008). · *d13C*: δ13C values of specimens from sampled lines. Values are reported in the V-PDB scale. · *d18Oc_PDB*: δ18Oc values of specimens from sampled lines. Values are reported in the V-PDB scale. **Judd et al., 2019_serial sampling.csv** · *Horizon:* horizons of the TELM 5 unit (La Meseta Formation) from which bivalves were collected. Horizon 1 is stratigraphically the lowest, while horizon 4 is the highest (Judd et al., 2019). · *ID*: Identification number of specimens. · *Latitude*: Geographic coordinate where bivalve specimens were collected. · *Longitude*: Geographic coordinate where bivalve specimens were collected. · *Surface sampled*: Specific sampling area, indicating whether sampling occurred in the interior or exterior portion of shells. · *distance*: The distance from the umbo in mm from which sampling occurred along a single shell. · *d18Oc_PDB*: δ18Oc values of specimens from sampled areas of shells. Values are reported on the V-PDB scale. **SpinupX3_25m_Mean_Monthly.nc** See section 1 ("iCESM modeling scripts" directory, “d18Ow Analysis Script.ipynb” script) for a full description of the iCESM model output extraction. **SpinupX6_25m_Mean_Monthly.nc** See section 1 ("iCESM modeling scripts" directory, “d18Ow Analysis Script.ipynb” script) for a full description of the iCESM model output extraction. **CA_x3CO2.csv** · *lat*: Geographic coordinate where temperature and δ18Ow model values are extracted from the iCESM simulation scaled at 3x preindustrial CO2 levels (values averaged within a seawater column depth of 25 m). · *long*: Geographic coordinate where temperature and δ18Ow model values are extracted from the iCESM simulation scaled at 3x preindustrial CO2 levels (values averaged within a seawater column depth of 25 m). · *T_mean*: Simulated seawater temperature values in °C. · *d18Ow*: Simulated seawater δ18Ow values (V-SMOW). · *d18Op*: Simulated seawater δ18Op values (V-SMOW). Values were calculated by using seawater temperature and δ18Ow arrays following the paleothermometer equation after Lécuyer et al. (2013). **d18O carbonate and phosphate references.csv** · *species*: Species of invertebrate taxa. · *type*: Specimen type, including barnacles, brachiopods, crabs, and mollusks. · *depth*: Depth of seawater column where specimens were collected, reported in meters below sea level when specified. · *d18Op*: δ18Op values of invertebrate specimens (V-SMOW). · *d18Oc_PDB*: δ18Oc values of invertebrate specimens (V-PDB). · *Reference*: Citations from which data were taken to build the dataset (Longinelli, 1965; Longinelli & Nuti, 1973). **TELM diversity.csv** · *genus:* genera of sharks and rays compiled from literature (Engelbrecht et al., 2016a, 2016b, 2017a, 2017b, 2019; Kriwet, 2005; Kriwet et al., 2016; Long, 1992; Marramá et al., 2018). · *species*: species of sharks and rays compiled from literature (Engelbrecht et al., 2016a, 2016b, 2017a, 2017b, 2019; Kriwet, 2005; Kriwet et al., 2016; Long, 1992; Marramá et al., 2018). · *Environment*: Inferred shark habitat based on taxonomy classified as benthic or pelagic environment. · *TELM*: Stratigraphic units of La Meseta (TELM 1-5; ~44 to ~37 Ma) and Submeseta formations (TELMs 6 and 7; ~37 to ~34 Ma) (Amenábar et al., 2020; Douglas et al., 2014; Montes et al., 2013). **3 “FTIR data” directory** The folder includes FTIR acquisitions and data analysis scripts on reference materials and shark tooth bioapatite for quality checks to test diagenesis effects on δ18Op of sharks. The folder includes: · The R project file “apatite_ftir.Rproj”. This project file navigates through scripts for raw data processing and data analysis. The background of the raw data was processed following custom R functions from Trayler et al. (2023; [https://github.com/robintrayler/collagen_demineralization](https://github.com/robintrayler/collagen_demineralization)). · The “.Rproj.user” folder includes project-specific temporary files (e.g. auto-saved source documents, window-state, etc.) stored by the R project file “apatite_ftir.Rproj”. The folder may be hidden depending on directory view options. · The “raw data” directory stores spectra acquisitions as .dpt files. Spectra files are stored in the folders “apatite” and “calcite” based on the material type. Spectra were obtained in the 400 – 4000 cm⁻¹ range using a Bruker Vertex 70 Far-Infrared in ATR located at the Nuclear Magnetic Resonance Facility at the University of California Merced (California, USA). · The “processed” directory includes processed spectra stored as .csv files (“apatite_data.csv” and “calcite_data.csv”) following the background correction (Trayler et al., 2023) and processed infrared data from Larocca Conte et al. (2024) (“Larocca Conte et al._SPORA_apatite_data.csv”) from which the NIST SRM 120c spectrum was filtered. Infrared spectra data in “Larocca Conte et al._SPORA_apatite_data.csv” were obtained and corrected following the same methodologies mentioned above. · The “R” directory includes R scripts of customized source functions for background correction (Trayler et al., 2023; inspect the "functions" directory and the R script "0_process_data.R") and data analysis (“data_analysis.R”). The scripts provide annotation through libraries and functions used for data processing and analysis. · Additional datasets. The “data_FTIR_d18O.csv” includes infrared data and δ18Op values of specimens, while the “Grunenwald et al., 2014_CO3.csv” is the dataset after Grunenwald et al. (2014) used to predict carbonate content from the materials featured in this work. **3.1 Dataset description** Spreadsheets included in the “processed” directory The datasets “apatite_data.csv”, “calcite_data.csv”, and “Larocca Conte et al._SPORA_apatite_data.csv” are structured with the following variables: · *wavenumber*: infrared wavenumber in cm-1. · *absorbance*: infrared absorbance value. · *file_name:* .dpt file name from which infrared wavenumber and absorbance values were obtained following the background correction. **data_FTIR_d18O.csv** · *file_name:* .dpt file name from which infrared wavenumber and absorbance values were obtained following the background correction. · *v4PO4_565_wavenumber*: Wavenumber of maximum infrared absorbance around the first νPO4 band, usually at 565 cm-1. · *v4PO4_565*: Peak absorbance value of the first ν4PO4 band (~565 cm-1). · *v4PO4_valley_wavenumber*: Wavenumber of valley between ν4PO4 bands. · *v4PO4_valley*: Absorbance value of the valley between ν4PO4 bands. · *v4PO4_603_wavenumber*: Wavenumber of maximum infrared absorbance around the second ν4PO4 band, usually at 603 cm-1. · *v4PO4_603*: Peak absorbance value of the second ν4PO4 band (~603 cm-1). · *CI*: Crystallinity index calculated after equation provided in (Shemesh, 1990) as (*v4PO4_565* + *v4PO4_603* / *v4PO4_valley*) (i.e., the sum of peak absorbance of νPO4 bands divided by the absorbance value of the valley between peaks). · *material*: Material type of samples (i.e., standard material, enameloid, dentin sampled from the crown or root area of shark teeth, and enameloid mixed with dentin). · *AUC_v3PO4*: Area under the curve of the ν3PO4 and ν1PO4 bands where maximum absorbance is at ~1025 cm-1 and ~960 cm-1, respectively. · *AUC_v3CO3*: Area under the curves of Type-A and Type-B carbonate bands having maximum infrared absorbance at ~1410 (Type-B), ~1456 (Type-B), and ~1545 cm-1 (Type-A). · *v3CO3_v3PO4_ratio*: Ratio between area under the curves of carbonate and phosphate bands (i.e., *AUC_v3CO3* / *AUC_v3PO4*). · *CO3_wt*: Estimated mean carbonate content following the equation in Grunenwald et al. (2014) (i.e. *CO3_wt* = 28.4793 (±1.4803) *v3CO3_v3PO4_ratio* + 0.1808(±0.2710); R2 = 0.985). · *CO3_wt_sd*: Standard deviation of estimated carbonate content calculated by propagating the error around coefficients provided in the Grunenwald et al. (2014) equation (see full equation in *CO3_wt*). · *Taxon*: Species assigned to shark tooth specimens. · *TELM*: Stratigraphic units of La Meseta (TELM 2-5; ~45 to ~37 Ma) and Submeseta formations (TELMs 6 and 7; ~37 to ~34 Ma) (Amenábar et al., 2020; Douglas et al., 2014; Montes et al., 2013). · *d18Op*: Mean δ18Op values of silver phosphate crystals precipitated from shark tooth bioapatite. Specimens were run in triplicates, corrected, and standardized on the V-SMOW scale. · *sd*: Standard deviation of silver phosphate triplicate samples per specimen. · *Collection*: Institutional abbreviations of museum collections where shark tooth specimens are housed. Infrared spectra were obtained from a selected subset of tooth specimens in the care of the Swedish Natural History Museum (NRM-PZ; Stockholm, Sweden). **Grunenwald et al., 2014_CO3.csv** · *sample*: Sample code. · *material*: Material type of samples (i.e., standard material, bone, and enamel). · *v3CO3*: Area under the curves of Type-A and Type-B carbonate bands having maximum infrared absorbance at ~1410 (Type-B), ~1456 (Type-B), and ~1545 cm-1 (Type-A). · *v3PO4*: *AUC_v3PO4*: Area under the curve of the ν3PO4 and ν1PO4 bands where maximum absorbance is at ~1025 cm-1 and ~960 cm-1, respectively. · *v3CO3_v3PO4_ratio*: *v3CO3_v3PO4_ratio*: Ratio between area under the curves of carbonate and phosphate bands (i.e., *v3CO3* /*v3PO4*). · *CO3_wt*: Carbonate content measured via CO2 coulometry. Further details about the analytical measurements are found in Grunenwald et al. (2014). **4 “Bayes_FEST_Temperautre Estimates” directory** The folder includes the Bayesian approach used to estimate posterior seawater temperature, δ18Ow values from δ18Op of sharks bioapatite using a Bayesian approach modified after Griffiths et al. (2023). The original scripts used in Griffiths et al. (2023) are reposited here: [https://github.com/robintrayler/bayesian_phosphate](https://github.com/robintrayler/bayesian_phosphate). The directory includes: · The R project file “Bayes_FEST.Rproj”. This project file navigates through scripts for raw data analysis. · The “.Rproj.user” folder includes project-specific temporary files (e.g. auto-saved source documents, window-state, etc.) stored by the R project file “Bayes_FEST.Rproj”. The folder may be hidden depending on directory view options. · The “data” folder includes the spreadsheets for modeled seawater temperature and δ18Ow values (“CA_x3CO2.csv”) and δ18Op values of shark tooth bioapatite (“shark FEST d18Op.csv”) used as prior information for the Bayesian model. We refer to section 2.1 for the full description of spreadsheets. · The “R” folder includes customized functions for the Bayesian model stored in the “functions” directory and the script for data analysis (“01_model_sharks.R”). The script includes a comparison of paleothermometer equations after Kolodny et al. (1983), Lécuyer et al. (2013), Longinelli & Nuti (1973), and (Pucéat et al. (2010) using the bulk δ18Op shark tooth bioapatite, simulated seawater temperature and δ18Ow values as prior inputs. While all paleothermometers estimate similar posterior bulk δ18Op close to empirical values, temperature estimates using the Pucéat et al. (2010) method are often the highest, generating estimates ~8°C higher than other equations. We therefore used the Lécuyer et al. (2013) paleothermomether for temperature estimates using δ18Op of shark bioapatite grouped by taxa because it: 1\) Provides consistent posterior temperature estimates relative to other equations (Longinelli & Nuti, 1973, Kolodny et al., 1983). 2\) provides temperature values from fish tooth specimens consistent with estimates of co-existing bivalves or brachiopod carbonate shells. The script provides annotation through libraries, statistical analysis, figures, and tables. **4 Software** **4.1 R** R and R Studio (R Development Core Team, 2024; RStudio Team, 2024) are required to run scripts included in the "d18O data and maps", “FTIR data”, and “Bayes_FEST_Temperautre Estimates” directories, which were created using versions 4.4.1 and 2024.04.02, respectively. Install the following libraries before running scripts: “cowplot” (Wilke, 2024), “colorspace” (Zeileis et al., 2020), “DescTools” (Signorell, 2024), “lattice” (Sarkar, 2008), “flextable” (Gohel & Skintzos, 2024), “ggh4x” (van den Brand, 2024), “ggnewscale” (Campitelli, 2024), “ggpubr” (Kassambara, 2023a), “ggspatial” (Dunnington, 2023), “ggstance” (Henry et al., 2024), “ggstar” (Xu, 2022), “greekLetters” (Kévin Allan Sales Rodrigues, 2023), “gridExtra” (Auguie, 2017), “mapdata” (code by Richard A. Becker & version by Ray Brownrigg., 2022); “mapproj” (for R by Ray Brownrigg et al., 2023), “maps” (code by Richard A. Becker et al., 2023), “ncdf4” (Pierce, 2023), “oce” (Kelley & Richards, 2023), “rasterVis” (Oscar Perpiñán & Robert Hijmans, 2023), “RColorBrewer” (Neuwirth, 2022), “rnaturalearth” (Massicotte & South, 2023), “rnaturalearthhires” (South et al., 2024),”rstatix” (Kassambara, 2023b), “scales” (Wickham et al., 2023), “tidyverse” (Wickham et al., 2019), “viridisLite” (Garnier et al., 2023). **4.2 Python** Python scripts, including “d18O Analysis Script.ipynb” and “NetCDF Plotting.ipynb”, utilize the Jupyter Notebook interactive ‘platform and are executed using Python version 3.9.16. Install the following libraries before running scripts: “xarray” (Hoyer & Joseph, 2017), “matplotlib” (Hunter, 2007), “cartopy” (Met Office, 2015). **5 References** Amenábar, C. R., Montes, M., Nozal, F., & Santillana, S. (2020). Dinoflagellate cysts of the la Meseta Formation (middle to late Eocene), Antarctic Peninsula: Implications for biostratigraphy, palaeoceanography and palaeoenvironment. *Geological Magazine*, *157*(3), 351–366. [https://doi.org/10.1017/S0016756819000591](https://doi.org/10.1017/S0016756819000591) Auguie, B. (2017). gridExtra: Miscellaneous Functions for “Grid” Graphics. Retrieved from [https://cran.r-project.org/package=gridExtra](https://cran.r-project.org/package=gridExtra) van den Brand, T. (2024). ggh4x: Hacks for “ggplot2.” Retrieved from [https://cran.r-project.org/package=ggh4x](https://cran.r-project.org/package=ggh4x) Campitelli, E. (2024). ggnewscale: Multiple Fill and Colour Scales in “ggplot2.” Retrieved from [https://cran.r-project.org/package=ggnewscale](https://cran.r-project.org/package=ggnewscale) code by Richard A. Becker, O. S., & version by Ray Brownrigg., A. R. W. R. (2022). mapdata: Extra Map Databases. Retrieved from [https://cran.r-project.org/package=mapdata](https://cran.r-project.org/package=mapdata) code by Richard A. Becker, O. S., version by Ray Brownrigg. Enhancements by Thomas P Minka, A. R. W. R., & Deckmyn., A. (2023). maps: Draw Geographical Maps. Retrieved from [https://cran.r-project.org/package=maps](https://cran.r-project.org/package=maps) Douglas, P. M. J., Affek, H. P., Ivany, L. C., Houben, A. J. P., Sijp, W. P., Sluijs, A., et al. (2014). Pronounced zonal heterogeneity in Eocene southern high-latitude sea surface temperatures. *Proceedings of the National Academy of Sciences of the United States of America*, *111*(18), 6582–6587. [https://doi.org/10.1073/pnas.1321441111](https://doi.org/10.1073/pnas.1321441111) Dunnington, D. (2023). ggspatial: Spatial Data Framework for ggplot2. Retrieved from [https://cran.r-project.org/package=ggspatial](https://cran.r-project.org/package=ggspatial) Engelbrecht, A., Mörs, T., Reguero, M. A., & Kriwet, J. (2016a). A new sawshark, Pristiophorus laevis, from the Eocene of Antarctica with comments on Pristiophorus lanceolatus. *Historical Biology*, *29*(6), 841–853. [https://doi.org/10.1080/08912963.2016.1252761](https://doi.org/10.1080/08912963.2016.1252761) Engelbrecht, A., Mörs, T., Reguero, M. A., & Kriwet, J. (2016b). Revision of Eocene Antarctic carpet sharks (Elasmobranchii, Orectolobiformes) from Seymour Island, Antarctic Peninsula. *Journal of Systematic Palaeontology*, *15*(12), 969–990. [https://doi.org/10.1080/14772019.2016.1266048](https://doi.org/10.1080/14772019.2016.1266048) Engelbrecht, A., Mörs, T., Reguero, M. A., & Kriwet, J. (2017a). Eocene squalomorph sharks (Chondrichthyes, Elasmobranchii) from Antarctica. *Journal of South American Earth Sciences*, *78*, 175–189. [https://doi.org/10.1016/j.jsames.2017.07.006](https://doi.org/10.1016/j.jsames.2017.07.006) Engelbrecht, A., Mörs, T., Reguero, M. A., & Kriwet, J. (2017b). New carcharhiniform sharks (Chondrichthyes, Elasmobranchii) from the early to middle Eocene of Seymour Island, Antarctic Peninsula. *Journal of Vertebrate Paleontology*, *37*(6). [https://doi.org/10.1080/02724634.2017.1371724](https://doi.org/10.1080/02724634.2017.1371724) Engelbrecht, A., Mörs, T., Reguero, M. A., & Kriwet, J. (2019). Skates and rays (Elasmobranchii, Batomorphii) from the Eocene La Meseta and Submeseta formations, Seymour Island, Antarctica. *Historical Biology*, *31*(8), 1028–1044. [https://doi.org/10.1080/08912963.2017.1417403](https://doi.org/10.1080/08912963.2017.1417403) for R by Ray Brownrigg, D. M. P., Minka, T. P., & transition to Plan 9 codebase by Roger Bivand. (2023). mapproj: Map Projections. Retrieved from [https://cran.r-project.org/package=mapproj](https://cran.r-project.org/package=mapproj) Garnier, Simon, Ross, Noam, Rudis, Robert, et al. (2023). {viridis(Lite)} - Colorblind-Friendly Color Maps for R. [https://doi.org/10.5281/zenodo.4678327](https://doi.org/10.5281/zenodo.4678327) Gohel, D., & Skintzos, P. (2024). flextable: Functions for Tabular Reporting. Retrieved from [https://cran.r-project.org/package=flextable](https://cran.r-project.org/package=flextable) Griffiths, M. L., Eagle, R. A., Kim, S. L., Flores, R. J., Becker, M. A., IV, H. M. M., et al. (2023). Endothermic physiology of extinct megatooth sharks. *Proceedings of the National Academy of Sciences*, *120*(27), e2218153120. [https://doi.org/10.1073/PNAS.2218153120](https://doi.org/10.1073/PNAS.2218153120) Grunenwald, A., Keyser, C., Sautereau, A. M., Crubézy, E., Ludes, B., & Drouet, C. (2014). Revisiting carbonate quantification in apatite (bio)minerals: A validated FTIR methodology. *Journal of Archaeological Science*, *49*(1), 134–141. [https://doi.org/10.1016/j.jas.2014.05.004](https://doi.org/10.1016/j.jas.2014.05.004) Henry, L., Wickham, H., & Chang, W. (2024). ggstance: Horizontal “ggplot2” Components. Retrieved from [https://cran.r-project.org/package=ggstance](https://cran.r-project.org/package=ggstance) Hoyer, S., & Joseph, H. (2017). xarray: N-D labeled Arrays and Datasets in Python. *Journal of Open Research Software*, *5*(1), 17. [https://doi.org/10.5334/jors.148](https://doi.org/10.5334/jors.148) Hunter, J. D. (2007). Matplotlib: A 2D graphics environment. *Computing in Science & Engineering*, *9*(3), 90–95. [https://doi.org/10.1109/MCSE.2007.55](https://doi.org/10.1109/MCSE.2007.55) Ivany, L. C., Lohmann, K. C., Hasiuk, F., Blake, D. B., Glass, A., Aronson, R. B., & Moody, R. M. (2008). Eocene climate record of a high southern latitude continental shelf: Seymour Island, Antarctica. *Bulletin of the Geological Society of America*, *120*(5–6), 659–678. [https://doi.org/10.1130/B26269.1](https://doi.org/10.1130/B26269.1) Judd, E. J., Ivany, L. C., DeConto, R. M., Halberstadt, A. R. W., Miklus, N. M., Junium, C. K., & Uveges, B. T. (2019). Seasonally Resolved Proxy Data From the Antarctic Peninsula Support a Heterogeneous Middle Eocene Southern Ocean. *Paleoceanography and Paleoclimatology*, *34*(5), 787–799. [https://doi.org/10.1029/2019PA003581](https://doi.org/10.1029/2019PA003581) Kassambara, A. (2023a). ggpubr: “ggplot2” Based Publication Ready Plots. Retrieved from [https://cran.r-project.org/package=ggpubr](https://cran.r-project.org/package=ggpubr) Kassambara, A. (2023b). rstatix: Pipe-Friendly Framework for Basic Statistical Tests. Retrieved from [https://cran.r-project.org/package=rstatix](https://cran.r-project.org/package=rstatix) Kelley, D., & Richards, C. (2023). oce: Analysis of Oceanographic Data. Retrieved from [https://cran.r-project.org/package=oce](https://cran.r-project.org/package=oce) Kévin Allan Sales Rodrigues. (2023). greekLetters: routines for writing Greek letters and mathematical symbols on the RStudio and RGui. Retrieved from [https://cran.r-project.org/package=greekLetters](https://cran.r-project.org/package=greekLetters) Kolodny, Y., Luz, B., & Navon, O. (1983). Oxygen isotope variations in phosphate of biogenic apatites, I. Fish bone apatite-rechecking the rules of the game. *Earth and Planetary Science Letters*, *64*(3), 398–404. [https://doi.org/10.1016/0012-821X(83)90100-0](https://doi.org/10.1016/0012-821X\(83\)90100-0) Kriwet, J. (2005). Additions to the Eocene selachian fauna of Antarctica with comments on Antarctic selachian diversity. *Journal of Vertebrate Paleontology*, *25*(1), 1–7. [https://doi.org/10.1671/0272-4634(2005)025\[0001:ATTESF\]2.0.CO;2](https://doi.org/10.1671/0272-4634\(2005\)025[0001:ATTESF]2.0.CO;2) Kriwet, J., Engelbrecht, A., Mörs, T., Reguero, M., & Pfaff, C. (2016). Ultimate Eocene (Priabonian) chondrichthyans (Holocephali, Elasmobranchii) of Antarctica. *Journal of Vertebrate Paleontology*, *36*(4). [https://doi.org/10.1080/02724634.2016.1160911](https://doi.org/10.1080/02724634.2016.1160911) Larocca Conte, G., Lopes, L. E., Mine, A. H., Trayler, R. B., & Kim, S. L. (2024). SPORA, a new silver phosphate precipitation protocol for oxygen isotope analysis of small, organic-rich bioapatite samples. *Chemical Geology*, *651*, 122000. [https://doi.org/10.1016/J.CHEMGEO.2024.122000](https://doi.org/10.1016/J.CHEMGEO.2024.122000) Lécuyer, C., Amiot, R., Touzeau, A., & Trotter, J. (2013). Calibration of the phosphate δ18O thermometer with carbonate-water oxygen isotope fractionation equations. *Chemical Geology*, *347*, 217–226. [https://doi.org/10.1016/j.chemgeo.2013.03.008](https://doi.org/10.1016/j.chemgeo.2013.03.008) Long, D. J. (1992). Sharks from the La Meseta Formation (Eocene), Seymour Island, Antarctic Peninsula. *Journal of Vertebrate Paleontology*, *12*(1), 11–32. [https://doi.org/10.1080/02724634.1992.10011428](https://doi.org/10.1080/02724634.1992.10011428) Longinelli, A. (1965). Oxygen isotopic composition of orthophosphate from shells of living marine organisms. *Nature*, *207*(4998), 716–719. [https://doi.org/10.1038/207716a0](https://doi.org/10.1038/207716a0) Longinelli, A., & Nuti, S. (1973). Revised phosphate-water isotopic temperature scale. *Earth and Planetary Science Letters*, *19*(3), 373–376. [https://doi.org/10.1016/0012-821X(73)90088-5](https://doi.org/10.1016/0012-821X\(73\)90088-5) Marramá, G., Engelbrecht, A., Mörs, T., Reguero, M. A., & Kriwet, J. (2018). The southernmost occurrence of Brachycarcharias (Lamniformes, Odontaspididae) from the Eocene of Antarctica provides new information about the paleobiogeography and paleobiology of Paleogene sand tiger sharks. *Rivista Italiana Di Paleontologia e Stratigrafia*, *124*(2), 283–297. Massicotte, P., & South, A. (2023). rnaturalearth: World Map Data from Natural Earth. Retrieved from [https://cran.r-project.org/package=rnaturalearth](https://cran.r-project.org/package=rnaturalearth) Met Office. (2015). Cartopy: a cartographic python library with a Matplotlib interface. Exeter, Devon. Retrieved from [https://scitools.org.uk/cartopy](https://scitools.org.uk/cartopy) Mine, A. H., Waldeck, A., Olack, G., Hoerner, M. E., Alex, S., & Colman, A. S. (2017). Microprecipitation and δ18O analysis of phosphate for paleoclimate and biogeochemistry research. *Chemical Geology*, *460*(March), 1–14. [https://doi.org/10.1016/j.chemgeo.2017.03.032](https://doi.org/10.1016/j.chemgeo.2017.03.032) Montes, M., Nozal, F., Santillana, S., Marenssi, S., & Olivero, E. (2013). Mapa Geológico de Isla Marambio (Seymour), Antártida, escala 1:20,000. *Serie Cartográfica*. Neuwirth, E. (2022). RColorBrewer: ColorBrewer Palettes. Retrieved from [https://cran.r-project.org/package=RColorBrewer](https://cran.r-project.org/package=RColorBrewer) Oscar Perpiñán, & Robert Hijmans. (2023). rasterVis. Retrieved from [https://oscarperpinan.github.io/rastervis/](https://oscarperpinan.github.io/rastervis/) Pierce, D. (2023). ncdf4: Interface to Unidata netCDF (Version 4 or Earlier) Format Data Files. Retrieved from [https://cran.r-project.org/package=ncdf4](https://cran.r-project.org/package=ncdf4) Pucéat, E., Joachimski, M. M., Bouilloux, A., Monna, F., Bonin, A., Motreuil, S., et al. (2010). Revised phosphate-water fractionation equation reassessing paleotemperatures derived from biogenic apatite. *Earth and Planetary Science Letters*, *298*(1–2), 135–142. [https://doi.org/10.1016/j.epsl.2010.07.034](https://doi.org/10.1016/j.epsl.2010.07.034) R Development Core Team. (2024). A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Vienna, Austria. RStudio Team. (2024). RStudio: Integrated Development for R. Boston, MA: RStudio, PBC. Retrieved from [http://www.rstudio.com/](http://www.rstudio.com/). Sarkar, D. (2008). *Lattice: Multivariate Data Visualization with R*. New York: Springer. Retrieved from [http://lmdvr.r-forge.r-project.org](http://lmdvr.r-forge.r-project.org) Shemesh, A. (1990). Crystallinity and diagenesis of sedimentary apatites. *Geochimica et Cosmochimica Acta*, *54*(9), 2433–2438. [https://doi.org/10.1016/0016-7037(90)90230-I](https://doi.org/10.1016/0016-7037\(90\)90230-I) Signorell, A. (2024). DescTools: Tools for Descriptive Statistics. Retrieved from [https://cran.r-project.org/package=DescTools](https://cran.r-project.org/package=DescTools) South, A., Michael, S., & Massicotte, P. (2024). rnaturalearthhires: High Resolution World Vector Map Data from Natural Earth used in rnaturalearth. Retrieved from [https://github.com/ropensci/rnaturalearthhires](https://github.com/ropensci/rnaturalearthhires) Trayler, R. B., Landa, P. V., & Kim, S. L. (2023). Evaluating the efficacy of collagen isolation using stable isotope analysis and infrared spectroscopy. *Journal of Archaeological Science*, *151*, 105727. [https://doi.org/10.1016/j.jas.2023.105727](https://doi.org/10.1016/j.jas.2023.105727) Wickham, H., Averick, M., Bryan, J., Chang, W., McGowan, L. D., François, R., et al. (2019). Welcome to the {tidyverse}. *Journal of Open Source Software*, *4*(43), 1686. [https://doi.org/10.21105/joss.01686](https://doi.org/10.21105/joss.01686) Wickham, H., Pedersen, T. L., & Seidel, D. (2023). scales: Scale Functions for Visualization. Retrieved from [https://cran.r-project.org/package=scales](https://cran.r-project.org/package=scales) Wilke, C. O. (2024). cowplot: Streamlined Plot Theme and Plot Annotations for “ggplot2.” Retrieved from [https://cran.r-project.org/package=cowplot](https://cran.r-project.org/package=cowplot) Xu, S. (2022). ggstar: Multiple Geometric Shape Point Layer for “ggplot2.” Retrieved from [https://cran.r-project.org/package=ggstar](https://cran.r-project.org/package=ggstar) Zeileis, A., Fisher, J. C., Hornik, K., Ihaka, R., McWhite, C. D., Murrell, P., et al. (2020). {colorspace}: A Toolbox for Manipulating and Assessing Colors and Palettes. *Journal of Statistical Software*, *96*(1), 1–49. [https://doi.org/10.18637/jss.v096.i01](https://doi.org/10.18637/jss.v096.i01) Zhu, J., Poulsen, C. J., Otto-Bliesner, B. L., Liu, Z., Brady, E. C., & Noone, D. C. (2020). Simulation of early Eocene water isotopes using an Earth system model and its implication for past climate reconstruction. *Earth and Planetary Science Letters*, *537*, 116164. [https://doi.org/10.1016/j.epsl.2020.116164](https://doi.org/10.1016/j.epsl.2020.116164) Eocene climate cooling, driven by the falling pCO2 and tectonic changes in the Southern Ocean, impacted marine ecosystems. Sharks in high-latitude oceans, sensitive to these changes, offer insights into both environmental shifts and biological responses, yet few paleoecological studies exist. The Middle-to-Late Eocene units on Seymour Island, Antarctica, provide a rich, diverse fossil record, including sharks. We analyzed the oxygen isotope composition of phosphate from shark tooth bioapatite (δ18Op) and compared our results to co-occurring bivalves and predictions from an isotope-enabled global climate model to investigate habitat use and environmental conditions. Bulk δ18Op values (mean 22.0 ± 1.3‰) show no significant changes through the Eocene. Furthermore, the variation in bulk δ18Op values often exceeds that in simulated seasonal and regional values. Pelagic and benthic sharks exhibit similar δ18Op values across units but are offset relative to bivalve and modeled values. Some taxa suggest movements into warmer or more brackish waters (e.g., Striatolamia, Carcharias) or deeper, colder waters (e.g., Pristiophorus). Taxa like Raja and Squalus display no shift, tracking local conditions in Seymour Island. The lack of difference in δ18Op values between pelagic and benthic sharks in the Late Eocene could suggest a poorly stratified water column, inconsistent with a fully opened Drake Passage. Our findings demonstrate that shark tooth bioapatite tracks the preferred habitat conditions for individual taxa rather than recording environmental conditions where they are found. A lack of secular variation in δ18Op values says more about species ecology than the absence of regional or global environmental changes. See methods in Larocca Conte, G., Aleksinski, A., Liao, A., Kriwet, J., Mörs, T., Trayler, R. B., Ivany, L. C., Huber, M., Kim, S. L. (2024). Eocene Shark Teeth From Peninsular Antarctica: Windows to Habitat Use and Paleoceanography. Paleoceanography and Paleoclimatology, 39, e2024PA004965.

Aim: We propose that forest trees create a vertical dimension for ecological niche variation that generates different regimes of climatic exposure, which in turn drives species elevation distributions. We test this hypothesis by statistically modelling the vertical and elevation distributions and microclimate exposure of rainforest ants. Location: Wet Tropics Bioregion, Australia Methods: We conducted 60 ground-to-canopy surveys to determine the vertical (tree) and elevation distributions, and microclimate exposure of ants (101 species) at 15 sites along four mountain ranges. We statistically modelled elevation range size as a function of ant species’ vertical niche breadth and exposure to temperature variance for 55 species found at two or more trees. Results: We found a positive association between vertical niche and elevation range of ant species: for every 3 m increase in vertical niche breadth our models predict a ~150% increase in mean elevation range size. Temperature variance increased with vertical height along the arboreal gradient and ant species exposure to temperature variance explained some of the variation in elevation range size. Main Conclusions: We demonstrate that arboreal ants have broader elevation ranges than ground-dwelling ants and are likely to have increased resilience to climatic variance. The capacity of species to expand their niche by climbing trees could influence their ability to persist over broader elevation ranges. We propose that wherever vertical layering exists - from oceans to forest ecosystems - vertical niche breadth is a potential mechanism driving macrogeographic distribution patterns and resilience to climate change. Data_collections.csv Main survey collections data in a site by species matrix showing all data for all sites surveyed. Tuna baited vials were placed every three metres from ground to canopy in trees at elevation sites at four subregion mountain ranges of the Australian Wet Tropics Bioregion. Note data file includes empty vials that lacked ants. Microclimate_AthertonTemp.csv This file contains Atherton Uplands temperature data from ibuttons deployed at one tree per elevation (200, 400, 600, 800, 1000) at every three metres in height in Dec-Jan 2017- 2018 set to record every half hour. See file Metadata for details of column names and data values.

Leaching dataset of dissolved organic carbon (DOC) and nitrogen (DON), nitrate (NO3+) and ammonium (NH4+) were collected from 6 cropping treatments (corn, switchgrass, miscanthus, native grass mix, restored prairie and poplar) established in the Bioenergy Cropping System Experiment (BCSE) which is a part of Great Lakes Bioenergy Research Center (www.glbrc.org) and Long Termn Ecological Research (LTER) program (www.lter.kbs.msu.edu). The site is located at the W.K. Kellogg Biological Station (42.3956° N, 85.3749° W and 288 m above sea level), 25 km from Kalamazoo in southwestern Michigan, USA. Prenart soil water samplers made of Teflon and silica (http://www.prenart.dk/soil-water-samplers/) were installed in blocks 1 and 2 of the BCSE (Fig. S1), and Eijkelkamp soil water samplers made of ceramic (http://www.eijkelkamp.com) were installed in blocks 3 and 4 (there were no soil water samplers in block 5). All samplers were installed at 1.2 m depth at a 45° angle from the soil surface, approximately 20 cm into the unconsolidated sand of the 2Bt2 and 2E/Bt horizons. Beginning in 2009, soil water was sampled at weekly to biweekly intervals during non-frozen periods (April to November) by applying 50 kPa of vacuum for 24 hours, during which water was collected in glass bottles. During the 2009 and 2010 sampling periods we obtained fewer soil water samples from blocks 1 and 2 where Prenart lysimeters were installed. We observed no consistent differences between the two sampler types in concentrations of the analytes reported here. Depending on the volume of leachate collected, water samples were filtered using either 0.45 µm pore size, 33-mm-dia. cellulose acetate membrane filters when volumes were <50 ml, or 0.45 µm, 47-mm-dia. Supor 450 membrane filters for larger volumes. Samples were analyzed for NO3-, NH4+, total dissolved nitrogen (TDN), and DOC. The NO3- concentration was determined using a Dionex ICS1000 ion chromatograph system with membrane suppression and conductivity detection; the detection limit of the system was 0.006 mg NO3--N L-1. The NH4+ concentration in the samples was determined using a Thermo Scientific (formerly Dionex) ICS1100 ion chromatograph system with membrane suppression and conductivity detection; the detection limit of the system was similar. The DOC and TDN concentrations were determined using a Shimadzu TOC-Vcph carbon analyzer with a total nitrogen module (TNM-1); the detection limit of the system was ~0.08 mg C L-1 and ~0.04 mg N L-1. DON concentrations were estimated as the difference between TDN and dissolved inorganic N (NO3- + NH4+) concentrations. The NH4+ concentrations were only measured in the 2013-2015 crop-years, but they were always small relative to NO3- and thus their inclusion or lack of it was inconsequential to the DON estimation. Leaching rates were estimated on a crop-year basis, defined as the period from planting or emergence of the crop in the year indicated through the ensuing year until the next year’s planting or emergence. For each sampling point, the concentration was linearly interpolated between sampling dates during non-freezing periods (April through November). The concentrations in the unsampled winter period (December through March) were also linearly interpolated based on the preceding November and subsequent April samples. Solute leaching (kg ha-1) was calculated by multiplying the daily solute concentration in pore-water (mg L -1) by the modeled daily drainage rates (m3 ha-1) from the overlying soil. The drainage rates were obtained using the SALUS (Systems Approach for Land Use Sustainability) model (Basso and Ritchie, 2015). SALUS simulates yield and environmental outcomes in response to weather, soil, management (planting dates, plant population, irrigation, nitrogen fertilizer application, tillage), and crop genetics. The SALUS water balance sub-model simulates surface run-off, saturated and unsaturated water flow, drainage, root water uptake, and evapotranspiration during growing and non-growing seasons (Basso and Ritchie, 2015). Drainage amounts and rates simulated by SALUS have been validated with measurements using large monolith lysimeters at a nearby site at KBS (Basso and Ritchie, 2005). On days when SALUS predicted no drainage, the leaching was assumed to be zero. The volume-weighted mean concentration for an entire crop-year was calculated as the sum of daily leaching (kg ha-1) divided by the sum of daily drainage rates (m3 ha-1). Weather data for the model were collected at the nearby KBS LTER meteorological station (lter.kbs.msu.edu). Leaching losses of dissolved organic carbon (DOC) and nitrogen (DON) from agricultural systems are important to water quality and carbon and nutrient balances but are rarely reported; the few available studies suggest linkages to litter production (DOC) and nitrogen fertilization (DON). In this study we examine the leaching of DOC, DON, NO3-, and NH4+ from no-till corn (maize) and perennial bioenergy crops (switchgrass, miscanthus, native grasses, restored prairie, and poplar) grown between 2009 and 2016 in a replicated field experiment in the upper Midwest U.S. Leaching was estimated from concentrations in soil water and modeled drainage (percolation) rates. DOC leaching rates (kg ha-1 yr-1) and volume-weighted mean concentrations (mg L-1) among cropping systems averaged 15.4 and 4.6, respectively; N fertilization had no effect and poplar lost the most DOC (21.8 and 6.9, respectively). DON leaching rates (kg ha-1 yr-1) and volume-weighted mean concentrations (mg L-1) under corn (the most heavily N-fertilized crop) averaged 4.5 and 1.0, respectively, which was higher than perennial grasses (mean: 1.5 and 0.5, respectively) and poplar (1.6 and 0.5, respectively). NO3- comprised the majority of total N leaching in all systems (59-92%). Average NO3- leaching (kg N ha-1 yr-1) under corn (35.3) was higher than perennial grasses (5.9) and poplar (7.2). NH4+ concentrations in soil water from all cropping systems were relatively low (<0.07 mg N L-1). Perennial crops leached more NO3- in the first few years after planting, and markedly less after. Among the fertilized crops, the leached N represented 14-38% of the added N over the study period; poplar lost the greatest proportion (38%) and corn was intermediate (23%). Requiring only one third or less of the N fertilization compared to corn, perennial bioenergy crops can substantially reduce N leaching and consequent movement into aquifers and surface waters. readme files are given that describe the data table

This work models a last-mile network design problem for an e-retailer with a capacitated two-echelon distribution structure - typical in e-retail last-mile distribution, catering to a market with a stochastic and dynamic daily customer demand requesting delivery within time-windows. Considering the distribution evnironment, this work formulates last-mile network design problem for this e-retailer as a dynamic-stochastic two capacitated location routing problem with time-windows. In doing so, this work splits the last-mile network design problem into its constituent strategic, tactical, and operational decisions. Here, the strategic decisions undertake long-term planning to develop a distribution structure with appropriate distribution facilities and a suitable delivery fleet to service the expected customer demand in the planning horizon. The tactical decisions pertain to medium-term day-to-day planning of last-mile delivery operations to establish efficient goods flow in this distribution structure to service the daily stochastic customer demand. And finally, operational decisions involve immediate short-term planning to fine-tune this last-mile delivery to service the requests arriving dynamically through the day. Note, the last-mile network design problem formulated as a location routing problem constitutes three subproblems encompassing facility location problem, customer allocation problem, and vehicle routing problem, each of which are NP-hard combinatorial optimization problems. To this end, this work develops an adaptive large neighborhood search meta-heuristic algorithm that searches through the neighborhood by destroying and consequently repairing the solution thereby reconfiguring large portions of the solution with specific operators that are chosen adaptively in each iteration of the algorithm, hence the name adaptive large neighborhood search. Further, considering the stochastic and dynamic nature of the delivery environment, this work develops a Monte-Carlo framework simulating each day in the planning horizon, with each day divided into 1-hr timeslots, and with each time-slot accepting customer requests for service by the end of the day. In particular, the framework assumes the e-retailer will delay route commitments until the last-feasible time-slot to accumulate customer requests and consequently assign them to an uncommitted delivery route. Note, a delivery route is committed once the e-retailer starts loading packages assigned to this delivery route onto the delivery vehicle assigned for this delivery route. At the end of every time-slot then, this framework assumes the e-retailer integrates the new customer requests by inserting these customer nodes into such uncommitted delivery routes in a manner that results in the least increase in distribution cost keeping the customer-distribution facility allocation fixed. Thus, the framework iterates through the time-slots with the e-retailer processing route commitments, accumulating customer requests, and subsequently integrating them into the delivery operations for the day. E-commerce has the potential to make urban goods flow economically viable, environmentally efficient, and socially equitable. However, as e-retailers compete with increasingly consumer-focused services, urban freight witnesses a significant increase in associated distribution costs and negative externalities particularly affecting those living close to logistics clusters. Hence, to remain competitive, e-retailers deploy alternate last-mile distribution strategies. These alternate strategies, such as those that include use of electric delivery trucks for last-mile operations, a fleet of crowdsourced drivers for last-mile delivery, consolidation facilities coupled with light-duty delivery vehicles for a multi-echelon distribution, or collection points for customer pickup, can restore sustainable urban goods flow. Thus, in this study, the authors investigate the opportunities and challenges associated with such alternate last-mile distribution strategies for an e-retailer offering expedited service with rush delivery within strict timeframes. To this end, the authors formulate a last-mile network design (LMND) problem as a dynamic-stochastic two-echelon capacitated location routing problem with time-windows (DS-2E-C-LRP-TW) addressed with an adaptive large neighborhood search (ALNS) metaheuristic.

For this manuscript, there were three types of methods performed to make our main conclusions: genetic diversity and structure analyses, downloading and mapping butternut fossil pollen during the last 20,000 years, and modeling and hindcasting butternut's (Juglans cinerea) distribution 20,000 years ago to present. Genetic analyses and species distribution modeling were performed in Emily Schumacher’s Github repository (https://github.com/ekschumacher/butternut) and pollen analyses and mapping were performed in Alissa Brown’s repository (https://github.com/alissab/juglans). Here is information detailing the Genetic data Data collection description: To perform genetic diversity and structure analyses on butternut, we used genetic data from the publication Hoban et al. (2010) and genetic data from newer sampling efforts on butternut from 2011 - 2015. Individuals were collected by Jeanne Romero-Severson, Sean Hoban, and Martin Williams over the course of ~ten years with a major sampling effort closer to 2009 followed up by another round of sampling 2012 - 2015. The initial 1,004 butternut individuals that were collected were genotyped by Sean Hoban and then the subsequent 757 individuals were genotyped in the Romero-Severson lab at Notre Dame non-consecutively. Genotyping was performed according to Hoban et al. (2008); DNA was extracted from fresh cut twigs using DNeasy Plant Mini kits (QIAGEN). PCR was performed by using 1.5 mM MgCl2, 1x PCR buffer [50 mm KCl, 10 mm Tris-HCl (pH 9.0), 0.1% Triton-X-100 (Fisher BioTech)], 0.2 mm dNTPs, 4 pm each forward and reverse primer, 4% Bovine Serum Albumin, 0.25 U TaKaRa Ex Taq Polymerase (Panvera), and 20 ng DNA template (10 μL total volume). The PCR temperature profile was as follows: 2 min at 94 °C; 30 cycles of 94 °C for 30 s, Ta for 30 s, and 72 °C for 30 s; 45 min at 60 °C; and 10 min at 72 °C on a PTC-225 Peltier Thermal Cycler (MJ Research). The process of assessing loci and rebinning for differences in years is detailed in the Schumacher et al. (2022) manuscript. Data files butternut_44pop.gen: Genepop file of original 1,761 butternut individuals, sampling described above, separated into original 44 sampling populations. butternut_24pop_nomd.gen: Genepop file of 1,635 butternut individuals, following rebinning based on researcher binning, reduced based on geographic isolation and missing data, organized into 24 populations. Used to generate all genetic diversity results. butternut_24pop_relate_red.gen: Genepop file of 993 butternut individuals, reduced for 25% relatedness, used to generate all clustering analyses. butternut_26pop_nomd.gen: Genepop file of 1,662 butternut individuals, reduced based on geographic isolation and missing data, including Quebec individuals, organized into 26 populations. Used to generate genetic diversity results with Quebec individuals. butternut_26pop_relate_red.gen: Genepop file of 1,015 butternut individuals, including Quebec individuals, reduced for 25% relatedness, used to generate clustering analyses with Quebec individuals. Fossil Pollen Data collection description: Pollen records for butternut were downloaded from Neotoma Paleoecology Database in 500-year time increments and visualized in 1,000 year-time increments 20,000 years ago to present. Data files butternut_pollen_data.csv: CSV of pollen records used for analyses and mapping. Includes original coordinates for each record (“og_long”, “og_lat”), the count of Juglans cinerea pollen at each site (“Juglans_cinerea_count”), and the age of the record (“Age”). To create the final maps, the coordinates were projected into Albers for each record (“Proj_Long,” “Proj_Lat”). Species Distribution Modeling and Hindcast Modeling Data collection description: We wanted to identify butternut's ecological preferences using boosted regression trees (BRT) and then hindcast distribution models into the past to identify migration pathways and locations of glacial refugia. Species distribution modeling was performed using boosted regression trees according to Elith et al. (2008). To run BRT, we needed to: 1. Reduce occurrence records to account for spatial autocorrelation, 2. Generate pseudo-absence points to identify the habitat where butternut is not found, 3. Obtain and extract the 19 bioclimatic variables at all points, 4. Select ecological variables least correlated with each other and most correlated with butternut presence. The BRT model that predicted butternut's ecological niche was then used to hypothesize butternut's suitable habitat and range shifts in the past. We downloaded occurrence records according to Beckman et al. (2019) as described here: https://github.com/MortonArb-ForestEcology/IMLS_CollectionsValue. The habitat suitability map generated from the BRT were projected into the past 20,000 years using Paleoclim variables (Brown et al., 2018). Data files butternut_BRT_var.csv: A CSV of the butternut presence and pseudoabsence points and extracted Bioclim variables (Fick & Hijman, 2017) used to run BRT in the final manuscript. Longitude and latitude coordinates are projected into Albers Equal Area Conic project, same with all of the ecological variables. Presence points are indicated with a 1 in the “PA” column and pseudo-absence points are indicated with a “0.” The variables most correlated with presence and least correlated with each other in this analysis were precipitation of the wettest month (“PwetM”), mean diurnal range (“MDR”), mean temperature of the driest quarter (“MTDQ”), mean temperature of the wettest quarter (“MTwetQ”), and seasonal precipitation (“precip_season”). References Brown, J. L., Hill, D. J., Dolan, A. M., Carnaval, A. C., & Haywood, A. M. (2018). PaleoClim, high spatial resolution paleoclimate surfaces for global land areas. Scientific Data, 5, 1-9 Elith, J., Leathwick, J. R., & Hastie, T. (2008). A working guide to boosted regression trees. Journal of Animal Ecology, 77, 802-813. Fick, S. E., & Hijmans, R. J. (2017). WorldClim 2: new 1‐km spatial resolution climate surfaces for global land areas. International Journal of Climatology, 37, 4302-4315. Hoban, S., Anderson, R., McCleary, T., Schlarbaum, S., and Romero-Severson, J. (2008). Thirteen nuclear microsatellite loci for butternut (Juglans cinerea L.). Molecular Ecology Resources, 8, 643-646. Hoban, S. M., Borkowski, D. S., Brosi, S. L., McCleary, T. S., Thompson, L. M., McLachlan, J. S., ... Romero-Severson, J. (2010). Range‐wide distribution of genetic diversity in the North American tree Juglans cinerea: A product of range shifts, not ecological marginality or recent population decline. Molecular Ecology, 19, 4876-4891. Aim: Range shifts are a key process that determine species distributions and genetic patterns. A previous investigation reported that Juglans cinerea (butternut) has lower genetic diversity at higher latitudes, hypothesized to be the result of range shifts following the last glacial period. However, genetic patterns can also be impacted by modern ecogeographic conditions. Therefore, we re-investigate genetic patterns of butternut with additional northern population sampling, hindcasted species distribution models, and fossil pollen records to clarify the impact of glaciation on butternut. Location: Eastern North America Taxon: Juglans cinerea (L., Juglandaceae) (butternut) Methods: Using 11 microsatellites, we examined range-wide spatial patterns of genetic diversity metrics (allelic richness, heterozygosity, FST) for previously studied butternut individuals and an additional 757 samples. We constructed hindcast species distribution models and mapped fossil pollen records to evaluate habitat suitability and evidence of species’ presence throughout space and time. Results: Contrary to previous work on butternut, we found that genetic diversity increased with distance to range edge, and previous latitudinal clines in diversity were likely due to a few outlier populations. Populations in New Brunswick, Canada were genetically distinct from other populations. At the Last Glacial Maximum, pollen records demonstrate butternut likely persisted near the glacial margin, and hindcast species distribution models identified suitable habitat in the southern United States and near Nova Scotia. Main conclusions: Genetic patterns in butternut may be shaped by both glaciation and modern environmental conditions. Pollen records and hindcast species distribution models combined with genetic distinctiveness in New Brunswick suggest that butternut may have persisted in cryptic northern refugia. We suggest that thorough sampling across a species range and evaluating multiple lines of evidence are essential to understanding past species movements. Data was cleaned and processed in R - genetic data cleaning and analyses and species distribution modeling methods were performed in Emily Schumacher's butternut repository and fossil pollen data cleaning and modeling was performed in Alissa Brown's juglans repository. Steps for performing data cleanining, analyses, and generating figures for the manuscript are described within each repo.