Calculating distance of polygons from coastline using ArcGIS Desktop?

Calculating distance of polygons from coastline using ArcGIS Desktop?

I have a series of polygons representing parcels of land. I would like to calculate the distance of each polygon from the coastline. I also have a polygon of the entire region of interest, i.e. an island.

What is the ArcGIS Desktop 10.1 procedure to achieve what I am after?

On the assumption that you have an Advanced level license (you do not specify otherwise), I think you should try the Near (Analysis) tool:

Determines the distance from each feature in the input features to the nearest feature in the near features, within the search radius.

  • Both input features and near features can be point, multipoint, line, or polygon.

4 1.5 Geospatial Technology

Suppose that you have launched a new business that manufactures solar-powered lawnmowers. You are planning a mail campaign to bring this revolutionary new product to the attention of prospective buyers. However, since it is a small business, you cannot afford to sponsor coast-to-coast television commercials or to send brochures by mail to more than 100 million U.S. households. Instead, you plan to target the most likely customers – those who are environmentally conscious, have higher than average family incomes, and who live in areas where there are enough water and the sunshine to support lawns and solar power.

Fortunately, lots of data are available to help you define your mailing list. Household incomes are routinely reported to banks and other financial institutions when families apply for mortgages, loans, and credit cards. Personal tastes related to issues like the environment are reflected in behaviors such as magazine subscriptions and credit card purchases. Market research companies collect such data and transform it into information by creating “lifestyle segments” – categories of households that have similar incomes and tastes. Your solar lawnmower company can purchase lifestyle segment information by 5-digit ZIP code, or even by ZIP+4 codes, which designate individual households.

Getting away from it all

If you could pick one place in Dane County to get away, off the grid, far from the hum of any highway or city glow, where would you go? To the middle of some vast, forgotten field in Primrose or a nameless Mazomanie forest?

Turns out, you don’t even have to fill up your gas tank to get to the remotest spot in the county. Florida-based conservation biologists Rebecca and Ryan Means would not be surprised. They are concerned, like many, that natural spaces are being gobbled up by roadways. “The U.S. road network fills the national landscape so fully that it is no longer possible to be more than 5 miles from a road within the vast majority of the conterminous 48 United States,” according to their website, Remote Footprints.

So the couple set out to find and travel to the wildest patch in each state to document the “ecological and physical conditions” in order “to increase nationwide awareness about the importance of preserving our remaining roadless wildlands — forever.”

They call it Project Remote, and all across the country, folks like me are imitating their methods for defining and finding “remote” locales, curious about more close-to-home frontiers.

The project’s quantitative definition of remoteness is the point that is the farthest straight-line distance from a road or town, which they calculate using Geographical Information Systems (GIS).

In September the scientists, with daughter Skyla in tow, visited Wisconsin’s most outlying point — 17.2 miles from the nearest road, on the far northern tip of Outer Island, the northernmost and easternmost island of the Apostle Island Archipelago in Lake Superior.

U.S. Coast Guard Capt. Jeff Nourse, who runs Nourse Charters and has been navigating these waters for nearly four decades, gave them a ride there. He wasn’t all that surprised when they contacted him, given how secluded the area is.

“When the park was created,” he says, of the lakeshore becoming a state-protected area in 1970, “it kept people from privatizing the island.”

“I fired up the GPS and cruised the shoreline,” Nourse says. After about 90 minutes, traveling between 25 and 29 mph, “the northeast wind kicked up, but we found the spot,” a rock outcropping with white waves slapping and slipping over it.

“It’s a wondrous place,” Nourse says, home to bears, foxes, geese, sandhill cranes and mergansers, skirted by clear, cold water filled with lake trout and salmon. “The park kind of speaks for itself. It just happens to be my backyard and happens to have a lot of history.”

It also happens to have cell phone reception, the Means were disappointed to note. “Why can you go to the remotest spot and get cell service when I can’t even get it at my house?” Nourse quips. They stayed for about 45 minutes, taking photos and recording observations.

I began my journey to locate Dane County’s remotest location by creating a free trial account on, which offers a suite of mapping apps I had no idea how to use. I wanted to replicate the Means’ calculation by using the Euclidean Distance tool. But to do that, I learned, I would need access to ArcGIS Desktop and the Spatial Analyst Extension. And possibly a few university prerequisites.

To my great relief, when I reached out to the Wisconsin State Cartographers Office for support, program specialist Codie See viewed my query as an exciting opportunity to “engage some of our student lab in service learning.”

In two days he shared their analysis of the single most remote square meter within Dane County. The result was this spot: 43°12’40.5”N 89°07’00.4”W (aka 89.116771 43.211244 decimal degrees). “Interesting and surprising, I think!” Codie wrote in an email.

I clicked on the coordinates. I could see East Towne Mall on my screen. The pin was dropped in a green shape on the map labeled Deansville State Wildlife Area, a 14-mile jaunt from said shopping mall.

See pointed out that their analysis “uses roads and bike paths as a basis” for remoteness and that they removed “hydro” — otherwise my destination would have been the middle of Lake Mendota.

On a recent Saturday morning I drove there, my dog Marshall riding shotgun. The GPS routed us via I-94 to Highway N. I passed the Oaks Golf Course in Cottage Grove and followed County Highway TT until it turned into County Road TT. Rounding a corner at a dairy farm, the paved road abruptly becomes gravel.

Streams on either side of me, choked with duckweed, nearly reached up and touched my wheels. I felt like I was scouting locations for The Walking Dead. There was one truck ahead.

“Drive 1.5 miles,” Siri instructed, “then prepare to park and walk 350 feet.” I parked in a gravel lot used to access the wildlife area and I rolled down my car window.

“Pop-pop-pop!” I spied a speck of blaze orange moving toward my square meter. I looked at my outfit — blue jeans and black winter coat. I looked at my dog, the world’s most bang!-boom!-blast!-weary pup.

We sat and listened. A few cars whirred by. Another bullet cracked the air. I peered into the brush, gauging my mood. My tolerance for gunfire, which on a scale of 1-10 typically hovers around 0.5, was not at its peak.

A blue jay complained in the distance. Black-capped chickadees played musical chairs with tree branches, singing, ti-ti-ti-fee-bee. A helicopter thumped overhead.

Conflicted, I rolled up the window and pulled away, 350 feet shy of the most isolated speck of Dane County. On this day, Dane County’s remote wilderness was just a little too crowded for me.

Calculating distance of polygons from coastline using ArcGIS Desktop? - Geographic Information Systems

Mike Bradley, Research Associate

The University of Rhode Island

This layer represents hurricane surge for category 1 through 4 hurricanes striking the East Coast. Hurricane surge values were developed by the National Hurricane Center using the SLOSH (Sea Lake and overland Surge from Hurricanes) Model data.

"Poly_ID" = the orginal polygon number from SLOSH Publication date 2014

U.S. Environmental Protection Agency, Region 5 Jan Krysa Geospatial Data Steward mailing and physical address 77 West Jackson Boulevard Chicago IL

(312) 353-7210 [email protected]

Tests for integrity have not been performed. Complete

We obtained SLOSH (Sea Lakes and Overland Surge from Hurricanes) model output from the National Hurricane Center. This data was exported in ArcGIS shapefile format. The shapefile represented data from the closest SLOSH basin to the park. The attributes represented the water surface elevation in feet that would occur from a worst-case hurricane surge within each polygon for categories 1 through 4 (for Man Tide MOM's). elevations were collected in feet and converted to meters. This analysis uses the 'mean' tide field only. The shapefile was in a Geographic NAD 83 horizontal coordinate system and used the NAVD88 vertical datum. For all ASIS data, the point shapefile used the NAVD29 vertical datum and was then converted to NAVD88. For all four SLOSH Basin shapefiles, the 'Feature to Point' tool was used to create a point shapefile of the centroids of the SLOSH data polygons.

U.S. Environmental Protection Agency, Office of Environmental Information Lee Kyle Geospatial Data Owner mailing and physical address 1200 Pennsylvania Avenue, N.W. Washington DC

202-564-4622 [email protected]

Shapefile MHHW NAVD88 Difference ZIP Compressed Archive 42492 The National Park Service shall not be held liable for improper or incorrect use of the data described and/or contained herein. These data and related graphics (i.e., GIF or JPG format files) are not legal documents and are not intended to be used as such. The information contained in these data is dynamic and may change over time. The data are not better than the original sources from which they were derived. It is the responsibility of the data user to use the data appropriately and consistent within the limitations of geospatial data in general and these data in particular. The related graphics are intended to aid the data user in acquiring relevant data it is not appropriate to use the related graphics as data. The National Park Service gives no warranty, expressed or implied, as to the accuracy, reliability, or completeness of these data. It is strongly recommended that these data are directly acquired from an NPS server and not indirectly through other sources which may have changed the data in some way. Although these data have been processed successfully on computer systems at the National Park Service, no warranty expressed or implied is made regarding the utility of the data on other systems for general or scientific purposes, nor shall the act of distribution constitute any such warranty. This disclaimer applies both to individual use of the data and aggregate use with other data. 20140722 None FGDC Content Standard for Digital Geospatial Metadata FGDC-STD-001-1998 20140722 Shapefile File Geodatabase Feature Class SLOSH Category 1-4 Inundation Polygons

<DIV STYLE="text-align:Left"><DIV><DIV><P><SPAN><SPAN>The National Park Service shall not be held liable for improper or incorrect use of the data described and/or contained herein. These data and related graphics are not legal documents and are not intended to be used as such.</SPAN></SPAN></P><P><SPAN><SPAN>The information contained in these data is dynamic and may change over time. The data are not better than the original sources from which they were derived. It is the responsibility of the data user to use the data appropriately and consistent within the limitations of geospatial data in general and these data in particular. The related graphics are intended to aid the data user in acquiring relevant data it is not appropriate to use the related graphics as data.</SPAN></SPAN></P><P><SPAN><SPAN>The National Park Service gives no warranty, expressed or implied, as to the accuracy, reliability, or completeness of these data. It is strongly recommended that these data are directly acquired from an NPS server and not indirectly through other sources which may have changed the data in some way. Although these data have been processed successfully on a computer system at the National Park Service, no warranty expressed or implied is made regarding the utility of the data on another system or for general or scientific purposes, nor shall the act of distribution constitute any such warranty. This disclaimer applies both to individual use of the data and aggregate use with other data.</SPAN></SPAN></P><P><SPAN /></P></DIV></DIV></DIV> -138.21454852 -12.68151645 6.65223303 61.7110157 Publication date 2014-01-01 Microsoft Windows 7 Version 6.1 (Build 7601) Service Pack 1 Esri ArcGIS 1 -68.700085 -67.997327 44.455661 43.975489 Mike Bradley The University of Rhode Island Research Associate 1 Greenhouse Rd. Kingston RI US [email protected] 401 874 5054 Barbara Shaw NPS NER GIS Coordinator [email protected] Tests for integrity have not been performed. Complete

IRPS ASA (Applied Science Associates) was sub-contracted to create the inundation polygons. ASA created a Visual Basic code that uses the SLOSH MOM (in PRN format) and the elevation data (FLT format) as inputs for interpolation. Both inverse distance weighting (IDW) and inverse area weighting (IAW) were used to interpolate the SLOSH MOM output onto the LiDAR data. The result was a FLT file of water surface elevation. Raster calculator in ArcGIS was then used to subtract the LiDAR elevation from the outputted water surface elevation, resulting in a raster of positive and negative values. Negative values indicate regions where the LiDAR elevation was higher than that of the water surface elevation, meaning that inundation does not occur. Positive values indicate the surge heights associated with the inundation caused by the respective storm. The Map Algebra tool, within the Spatial Analyst extension, is used to identify values less than 0. This raster data converted to a single polygon, where values less than 0, along with disconnected regions which would not be flooded, were removed. The result is a single inundation polygon associated with each storm category. 2014-01-01T00:00:00 Mike Bradley The University of Rhode Island Research Associate 1 Greenhouse Rd. Kingston RI US 401 874 5054 Mike Bradley


We identify 13,209 land parcels along Georgia’s estuarine shoreline for analysis these parcels provide a census of armored and unarmored estuarine shoreline parcels. Chatham County comprises the largest number of these parcels (4856, 37%) and Bryan County comprises the fewest (955, 7%) (Fig. 2). In total, we estimate 2,997 parcels to be armored (23%). Chatham County also comprises the largest number of armored parcels (1473, 49%), whereas Liberty County comprises the fewest (242, 8%). When comparing armoring prevalence among counties, Bryan County has the highest percentage of armored parcels, with 37% of all estuarine shoreline parcels in Bryan County being armored (Fig. 2). In total, we estimate 4,004 parcels (30%) to be adjacent to a parcel with existing armoring.

Fraction of all shoreline parcels (blue) and all armored shoreline parcels (orange) by county. Fraction of armored shoreline parcels within a county (yellow).

Logistic regression analysis

The final selected logistic regression model for describing patterns of hard armoring among estuarine shoreline parcels in Georgia includes the predictor variables shown in Fig. 3. A likelihood ratio test supported the inclusion of neighborhood fixed effects (Chi-square = 323.476, with 172 degrees of freedom), and this model exhibits a correct classification rate of 88%. Chatham County is used as the reference for county fixed effects. Structure value appears to be best represented by replacement cost per building area, which we term “building value” ($/m 2 ). The explanatory power of shoreline length is improved by natural logarithm transformation. The urban classification descriptor variables (housing and population density) had inconsistent associations with armoring throughout model development, as well as small marginal effects and minimal influence on classification accuracy thus, we did not include a measure of urban classification in the final model. The final model (Fig. 3) indicates that eight of ten landscape and socioeconomic attributes selected a priori are strong predictors of the log-odds of shoreline armoring (p < 0.1). (See Appendix A, column two for numerical regression results).

Forest plot of the change in the log-odds of the probability of hard armoring resulting from a unit increase in the predictor variables included in the logistic regression model. Positive values indicate a positive association with hard armoring likelihood and negative values indicate a negative association with hard armoring likelihood. Bars are 95% confidence intervals. Parameter intervals that overlap zero do not significantly influence the probability of hard armoring (at 5% significance level).

We find that parcel slope (elevation/distance from the shore) has the largest effect on the log-odds probability of hard armoring, with a change in the log-odds value of 3.75 and a marginal effect of 0.33. Thus, a one-unit increase in the slope (from an average of 0.025) increases the likelihood of armoring by 33%. More telling, however, the elasticity of slope is 0.03, indicating that a one-percent increase in slope increases the probability of armoring by only 0.03%. Distance from the shoreline, on its own, has a small negative effect on armoring (marginal effect of −0.0006), while elevation does not have a statistically significant effect (independent of slope).

Also very impactful in the logistic regression model, the “neighbor armoring” coefficient indicates a change in the log-odds average value of 2.32 and a marginal effect of 0.18. Thus, being located next to an armored parcel increases the likelihood of armoring by 18% (holding all other predictor variables constant). This effect may reflect environmental forcings that are common to all parcels in a particular area, spatial spillovers in erosion risk due to installation of hard armoring on neighboring properties, or herding behavior (in which landowners adopt practices they see their neighbors using). To attempt to control for this, we include indicators for medium-energy or high-energy shoreline environments (relative to low-energy) and the historical erosion rate. Model results suggest medium-energy environments have no discernable impact on armoring, but high-energy shoreline environments increase the likelihood of hard armoring by 12%. A one-unit increase in the historical erosion rate increases the probability of armoring by 11%.

Other predictor variables exhibited modest effects in the logistic regression model. A one-meter increase in shoreline length reduces the log-odds by 0.09 (the likelihood of hard armoring by 0.0002 – marginal effect not statistically significant). An additional square-meter of parcel area increases the log-odds of hard armoring by 0.105, with a marginal effect of 0.009. A one-dollar increase in structure replacement cost (per m 2 ) increases the log-odds by 0.0093, with a marginal effect of 0.0008. Parcels located in Glynn, Liberty, and Bryan Counties are more likely to be armored, controlling for other predictors and neighborhood fixed effects, relative to Chatham County, while parcels in McIntosh County are less likely to be armored. Camden County was no different from Chatham County (all else being equal). The final model showed no evidence of spatial autocorrelation in errors (Moran’s I = 0.000246, p = 0.8125).

Calculating distance of polygons from coastline using ArcGIS Desktop? - Geographic Information Systems

Merkouri Christina, Kouli Maria. The Spatial Distribution and Location of Bronze Age Tumuli in Greece. In: Ancestral Landscape. Burial mounds in the Copper and Bronze Ages (Central and Eastern Europe – Balkans – Adriatic – Aegean, 4th-2nd millennium B.C.) Proceedings of the International Conference held in Udine, May 15th-18th 2008. Lyon : Maison de l'Orient et de la Méditerranée Jean Pouilloux, 2012. pp. 203-217. (Travaux de la Maison de l'Orient et de la Méditerranée. Série recherches archéologiques, 58)

The spatial distribution and location of Bronze Age tumuli in Greece

Christina Merkouri *, Maria Kouli **

Introductory facts

A funerary monument visible from a long distance, the tumulus was widespread in Greece from its first occurrence in the Early Bronze Age until Late Antiquity. Its make-up and the type of tombs that it contained (pithoi, shafts, cist graves, tholos tombs, chamber tombs) changed through the centuries, with regional variations in different time periods. The morphology of a tumulus depended on the environment and available building materials in each specific area, on the needs and living conditions of a particular group bonded by family ties and on their date. Because tumuli show much variation, several classifications with morphological, regional and chronological groups have been suggested in the past. The criteria taken into account include the tumuli’s structure and characteristics, both external (shape, size of enclosure walls, occurrence, either single or in groups within cemeteries) and internal (fill composition, number, type and placement of tombs, related burial customs). Ancestral Landscapes.

TMO 58, Maison de l’Orient et de la Méditerranée, Lyon, 2011

* H ellenic Ministry of Culture, Athens, Greece. ** L aboratory of Geoinformation, Department of Natural Resources Research & Environment, Technological Educational Institute of Crete, Chania, Greece.


Sánchez-Arcilla, A. et al. Climatic drivers of potential hazards in Mediterranean coasts. Reg Environ Change 11, 617–636, (2010).

UNEP/MAP. Mediterranean Strategy for Sustainable Development 2016-2025. Plan Bleu, Regional Activity Centre (Valbonne, 2016).

European Environment Agency. Horizon 2020 mediterranean report (2014).

Conte, D. & Lionello, P. Characteristics of large positive and negative surges in the Mediterranean Sea and their attenuation in future climate scenarios. Glob. Planet. Change 111, 159–173, (2013).

Casas-Prat, M. & Sierra, J. P. Trend analysis of wave storminess: wave direction and its impact on harbour agitation. Nat. Hazards Earth Syst. Sci 10, 2327–2340, (2010).

Jimenez, J. A., Valdemoro, H. I., Bosom, E., Sanchez-Arcilla, A. & Nicholls, R. J. Impacts of sea-level rise-induced erosion on the Catalan coast. Reg Environ Change 17, 593–603, (2017).

Vousdoukas, M. I., Mentaschi, L., Voukouvalas, E., Verlaan, M. & Feyen, L. Extreme sea levels on the rise along Europe's coasts. Earths Future 5, 304–323, (2017).

UNEP/MAP/PAP. Protocol on Integrated Coastal Zone Management in The Mediterranean (Priority Actions Programme Regional Activity Centre, Split, 2008).

Malvarez, G. C., Pintado, E. G., Navas, F. & Giordano, A Spatial data and its importance for the implementation of UNEP MAP ICZM Protocol for the Mediterranean. J Coast Conserv 19, 633–641, (2015).

Santoro, F., Lescrauwaet, A. K., Taylor, J. & Breton, F. Integrated Regional Assessments in support of ICZM in the Mediterranean and Black Sea Basins (Intergovernmental Oceanographic Commission of UNESCO: Paris, 2014).

United Nations Environment Programme/Mediterranean Action Plan (UNEP/MAP)-Plan Bleu. State of the environment and development in the Mediterranean (UNEP/MAP-Plan Bleu: Athens, 2009).

Hinkel, J. et al. A global analysis of erosion of sandy beaches and sea-level rise: An application of DIVA. Glob. Planet. Change 111, 150–158, (2013).

Hinkel, J. et al. Coastal flood damage and adaptation costs under 21st century sea-level rise. Proc. Natl. Acad. Sci. U. S. A 111, 3292–3297, (2014).

Spencer, T. et al. Global coastal wetland change under sea-level rise and related stresses: The DIVA Wetland Change Model. Glob. Planet. Change 139, 15–30, (2016).

Vafeidis, A. T. et al. A New Global Coastal Database for Impact and Vulnerability Analysis to Sea-Level Rise. J COASTAL RES 244, 917–924, (2008).

McFadden, L., Nicholls, R. J., Vafeidis, A. & Tol, R. S. J. A Methodology for Modeling Coastal Space for Global Assessment. J. Coastal Res 234, 911–920, (2007).

Brown, S. et al. Shifting perspectives on coastal impacts and adaptation. Nat. Clim. Change 4, 752–755, (2014).

Bartlett, D. J in Marine and Coastal Geographical Information Systems. (ed. Wright D. J. & Bartlett D. J. 11–24 (Taylor and Francis, 2000).

Brenner, J., Jimenez, J. A. & Sarda, R. Definition of homogeneous environmental management units for the Catalan coast. Environ. Manage 38, 993–1005, (2006).

Wolff, C., Vafeidis, A. T., Lincke, D., Marasmi, C. & Hinkel, J. Effects of scale and input data on assessing the future impacts of coastal flooding: An application of DIVA for the Emilia-Romagna coast. Front Mar Sci 3, 1–15, (2016).

Global Administrative Areas (GADM), (2015).

Schneider, A., Friedl, M. A. & Potere, D. A new map of global urban extent from MODIS satellite data. Environ. Res. Lett 4, 1–11, (2009).

Scheffers, A. M., Scheffers, S. R. & Kelletat, D. H. The Coastlines of the World with Google Earth (Springer Netherlands, 2012).

Wong, P. P. et al. in Climate Change 2014: Impacts,Adaptation, and Vulnerability (eds Field C. B. et al.) 361–409 (Cambridge University Press, 2014).

Nicholls, R. J. et al. Sea-level scenarios for evaluating coastal impacts. Wiley Interdiscip Rev Clim Change 5, 129–150, (2014).

Muis, S., Guneralp, B., Jongman, B., Aerts, J. C. & Ward, P. J. Flood risk and adaptation strategies under climate change and urban expansion: A probabilistic analysis using global data. Sci Total Environ 538, 445–457, (2015).

Lichter, M., Vafeidis, A. T., Nicholls, R. J. & Kaiser, G. Exploring Data-Related Uncertainties in Analyses of Land Area and Population in the “Low-Elevation Coastal Zone” (LECZ). J. Coastal Res 274, 757–768, (2011).

Neumann, B., Vafeidis, A. T., Zimmermann, J. & Nicholls, R. J. Future Coastal Population Growth and Exposure to Sea-Level Rise and Coastal Flooding - A Global Assessment (vol 10, e0118571, 2015). PLoS ONE 101, (2015).

McGranahan, G., Balk, D. & Anderson, B. The rising tide: assessing the risks of climate change and human settlements in low elevation coastal zones. Environ Urban 19, 17–37, (2007).

Muis, S., Verlaan, M., Winsemius, H. C., Aerts, J. C. J. H. & Ward, P. J. A global reanalysis of storm surges and extreme sea levels. Nat. Commun 7, 1–11, (2016).

Muis, S. et al. A comparison of two global datasets of extreme sea levels and resulting flood exposure. Earths Future 5, 379–392, (2017).

Hunter, J. R., Woodworth, P. L., Wahl, T. & Nicholls, R. J. Using global tide gauge data to validate and improve the representation of extreme sea levels in flood impact studies. Glob. Planet. Change 156, 34–45, (2017).

Hasselmann, S. et al. WAMDI group. The WAM model—a third generation ocean wave prediction model. J. Phys. Oceanogr 18, 1775–1810 (1988).

Lionello, P. & Sanna, A. Mediterranean wave climate variability and its links with NAO and Indian Monsoon. Clim. Dynam 25, 611–623, (2005).

Lionello, P., Cogo, S., Galati, M. B. & Sanna, A The Mediterranean surface wave climate inferred from future scenario simulations. Glob. Planet. Change 63, 152–162, (2008).

Burrough, P. A., McDonnell, R. Principles of geographical information systems. [ Rev. ed.] edn, (Oxford University Press, 1998).

Woodworth, P. L. et al. Towards a global higher-frequency sea level dataset. Geosci. Data J 3, 50–59, (2016).

Wahl, T. et al. Understanding extreme sea levels for broad-scale coastal impact and adaptation analysis. Nat. Commun 8, 1–12, (2017).

Wakelin, S. L. & Proctor, R. The impact of meteorology on modelling storm surges in the Adriatic Sea. Glob. Planet. Change 34, 97–119, (2002).

Vousdoukas, M. I., Voukouvalas, E., Annunziato, A., Giardino, A. & Feyen, L. Projections of extreme storm surge levels along Europe. Clim. Dynam 47, 3171–3190, (2016).

International Standard ISO 3166-1. Codes for the representation of names of countries and their subdivisions - Part 1: Country codes, ISO 3166-1 (2006).

World Bank. GDP per capita, PPP (current international $) (2016).

International Institute for Applied Systems Analysis. SSP Database (2015).

Crespo Cuaresma, J. Income projections for climate change research: A framework based on human capital dynamics. Glob Environ Change 42, 226–236, (2015).

KC, S. & Lutz, W. The human core of the shared socioeconomic pathways: Population scenarios by age, sex and level of education for all countries to 2100. Glob Environ Change 42, 181–192, (2014).

Vafeidis, A. T. et al. A New Global Coastal Database for Impact and Vulnerability Analysis to Sea-Level Rise. Journal of Coastal Research 24, 917–924, (2008).

Weatherall, P. et al. A new digital bathymetric model of the world's oceans. Earth Space Sci 2, 331–345, (2015).

Pickering, M. D. et al. The impact of future sea-level rise on the global tides. Cont Shelf Res 142, 50–68, (2017).

Pickering, M. D. The impact of future sea-level rise on the tides, Phd thesis (University of Southampton, 2014).

Peltier, W. R. Global glacial isostasy and the surface of the ice-age earth: The ice-5G (VM2) model and grace. Annu Rev Earth Planet Sci 32, 111–149, (2004).

Argus, D. F., Peltier, W. R., Drummond, R. & Moore, A. W. The Antarctica component of postglacial rebound model ICE-6G_C (VM5a) based on GPS positioning, exposure age dating of ice thicknesses, and relative sea level histories. Geophys J Int 198, 537–563, (2014).

Peltier, W. R., Argus, D. F. & Drummond, R. Space geodesy constrains ice age terminal deglaciation: The global ICE-6G_C (VM5a) model. J. Geophys. Res. Solid Earth 120, 450–487, (2015).

McOwen, C. et al. A global map of saltmarshes. Biodivers Data J 5, e11764, (2017).

World Tourism Organization. Yearbook of Tourism Statistics & Compendium of Tourism Statistics and data files. International tourism, number of arrivals (2014).

Rio, M. H., Mulet, S. & Picot, N. Beyond GOCE for the ocean circulation estimate: Synergetic use of altimetry, gravimetry, and in situ data provides new insight into geostrophic and Ekman currents. Geophys Res Lett 41, 8918–8925, (2014).

Jarvis, A., Reuter, H. I., Nelson, A. & Guevara, E. Hole-filled SRTM for the globe Version 4: Available from the CGIAR-CSI SRTM 90 m Database (2008).

Santini, M., Taramelli, A. & Sorichetta, A. ASPHAA: A GIS-Based Algorithm to Calculate Cell Area on a Latitude-Longitude (Geographic) Regular Grid. Trans GIS 14, 351–377, (2010).

Center for International Earth Science Information Network - CIESIN - Columbia University, International Food Policy Research Institute - IFPRI, The World Bank & Centro Internacional de Agricultura Tropical - CIAT (NASA Socioeconomic Data and Applications Center (SEDAC), Palisades, NY, 2011).

Center for International Earth Science Information Network - Columbia University. Gridded Population of the World, Version 4 (GPWv4): Population Count Adjusted to Match 2015 Revision of UN WPP Country Totals (NASA Socioeconomic Data and Applications Center, Palisades, NY, 2016).

European Space Agency & Université Catholique De Louvain. GlobCover (2009).

Reimann, L., Merkens, J.-L. & Vafeidis, A. T. Regionalized Shared Socioeconomic Pathways: narratives and spatial population projections for the Mediterranean coastal zone. Reg Environ Change 18, 235–245, (2017).

Merkens, J.-L., Reimann, L., Hinkel, J. & Vafeidis, A. T. Gridded population projections for the coastal zone under the Shared Socioeconomic Pathways. Glob. Planet. Change 145, 57–66, (2016).

Data Citations


We thank the following for their assistance and comments on earlier drafts: A. Townsend Peterson, J. Soberon, V. Sanchez-Cordero, S. Phillips, J. Losos, J. Wiens, J. Bastow, P. Wainwright, B. Shaffer, D. Schemske, T. Schoener, C. Moritz, R. Anderson, B. Monahan, A. Wright, R. Carlson, S. Veloz, S. Cameron, K. Moore, D. Grossenbacher, P. Ryan, S. Elmendorf, M. Wu, M. Brandley, T. Iglesias, and L. Moyle. We also thank the reviewers of this manuscript, in particular one anonymous reviewer whose extensive comments resulted in substantial improvements.


Species distribution modeling

Greater one-horned rhinoceros

Based on Area Under the Curve (AUC) of the Receiver Operator Curve (ROC) and on 100 bootstrap runs of omission/commission analysis with 20% test data showed that the rhinoceros model had a good fit [AUC = 0.96 (SE 0.0007)] and predictive ability (Fig. 1a,b). This model was also rated the best by True Skill Statistics (TSS) and had the lowest Akaike Information Criteria (AIC). Rhinoceros occurrence was best explained by (a) Distance from grassland, (b) Distance from forest, (c) Maximum temperature of warmest month (d) Annual Precipitation and, (e) Distance from water (Table 1). As expected, species response curve showed a decline in habitat suitability with increasing distance from grasslands (Fig. 1c). Rhinoceros were unlikely to occur at distances > 2 km from grasslands. Distance from grassland habitat explained the maximum variation (69%) in the rhinoceros distribution model. Rhinoceros occurrence declines rapidly from forest edge (Fig. 1d) which contributed 16% to the model. A temperature range between 32 to 40 °C during the warmest months of the year governed occurrence of rhinoceros (Fig. 1e). This covariate (Maximum temperature of the warmest month) explained 8.2% of the variation in the data in the best model. Rhinoceros occurred where annual rainfall exceeded 1700 mm and the probability of occurrence increased with increase in rainfall up to 4000 mm (Fig. 1f). Rhinoceros occurrence probability declined sharply with increasing distance to water up to 5 km, after this distance there was high variability in the response curve, This covariate contributed the least to the model at 1.9% (Fig. 1g).

(a) Receiver operator curve for assessing model fit, (b) Omission/Commission analysis for model accuracy of classifying test data. Species occurrence probability (response curves) obtained from 100 bootstrap runs of the best model explaining the distribution of greater one horned rhinoceros in Maxent to (c) distance to grassland, (d) distance to forest, (e) maximum temperature of warmest month, (f) annual precipitation, and (g) distance to water.

Swamp buffalo

The Buffalo model had a good fit with an AUC of 0.95 (SE 0.0027) but the predictive ability based on 100 bootstrap runs of omission/commission analysis with 20% test data (Fig. 2a,b) was not as good as that for rhinoceros. The TSS was the best for this model while the AIC was second best (Table 1). However, since this model had a good fit and made ecological sense supported by TSS we use it as the best model. The covariates of the best buffalo model were (a) Distance from grassland, (b) Distance from forest, (c) Annual Precipitation and, d) Distance from water (Table 1). Distance to grassland had the highest contribution to the model. Buffalo occurred in the proximity of grasslands and were not likely to be found beyond 1 km distance from grasslands (Fig. 2c). The model showed high buffalo habitat suitability at forest edges with low probability with increasing distances from forests (Fig. 2d). Areas having annual rainfall above 1500 mm were preferred (Fig. 2e). Habitat suitability for buffalo declined rapidly with increasing distance from water for up to 2 km after which habitat was unsuitable with high variability in model predictions (Fig. 2f).

(a) Receiver Operator Curve for assessing model fit, (b) Omission/Commission analysis for model accuracy of classifying test data. Response curves obtained from 100 bootstrap runs of modeling distribution of wild swamp buffalo in Maxent to (c) distance to grassland, (d) distance to forest, (e) Annual precipitation and (f) distance to water for the Terai-Brahmaputra floodplains landscape.

Population habitat viability analysis (PHVA)

Greater one-horned rhinoceros

PHVA for rhinoceros revealed small populations (K ≤ 10) could not persist (Table 2, Scenarios 1–4 and Table S1). Medium sized populations (K = 20–30) are shown to be viable when initial reintroductions are undertaken with > 8–10 rhinoceros and sites are occasionally supplemented, but these populations would not withstand poaching (Table 2, Scenarios 5–11 and Table S1,). Populations with K ≥ 50 have better chances of survival which increases with supplementation. These populations would also withstand low levels of poaching when supplemented at initial stages (Table 2, Scenarios 14 & 15). Populations

100 would survive long-term, even when subjected to a low-level poaching, and are able to retain high levels of heterozygosity even without immigrants (Table 2, Scenario 19–21 and Table S1). Populations with K ≥ 100 are ideal for long-term persistence (Table 2, Scenario 22). A meta-population comprising of Kaziranga, Orang and Laokhowa-Bura Chapori demonstrates a stochastic growth rate of 0.022 (SE 0.019). The extinction probability of that metapopulation is zero and heterozygosity is maintained at

Swamp buffalo

Small populations (K ≤ 20) exhibit low persistence probability despite supplementation (Table 3, Scenarios 1–4 and TableS2). Medium sized populations (K = 50) could persist with initial reintroduction of > 10 individuals supplemented for a decade, but would not sustain poaching offtake (Table 3, Scenarios 5-9 and Table S2). Large populations (K = 100–200 buffalos) remain viable in settings which experience natural catastrophes, and are also resilient to moderate poaching losses if initial founding population is > 30. Populations are sensitive to the size of founding population and depend upon continued supplementation (Table 3, Scenario 10–16 and Table S2). Areas which could sustain (K) > 250 were ideal for long-term persistence and could tolerate moderate poaching. However, all populations were sensitive to founding population size, and a founding population > 30 is optimal.

Identifying and prioritizing suitable habitats

The species distribution probability asc. layer obtained as the median from 100 Maxent bootstrap runs were exported to Arc GIS 10.5 to produce probability maps for rhinoceros (Fig. 3) and buffalo (Fig. 4), showed reasonable extents of suitable habitat outside of these species’ current range. Maps from conservative estimates of 95% lower limits also identified substantial patches of suitable habitat for reintroductions (Fig S1). Maximum training sensitivity plus specificity cumulative threshold values for 100 bootstrap runs for the rhinoceros model was less variable (13.95 ± SE 0.196 range 9.6–18.5) when compared to that of the buffalo model (19.77 ± SE 0.88 range 1.7–54.7). Maxent identified 11 and 7 habitat patches for rhinoceros and buffalo respectively that could sustain > 50 and > 100 individuals of each species outside of the current range (Table 4). After prioritizing these sites based on legal status, protection, management efficacies and minimal potential for human conflict we identified Corbett NP and Valmiki Tiger Reserve (TR) as top priority sites for rhinoceros reintroduction. For buffalo reintroductions the PA complex of Chitwan NP-Valmiki TR-Parsa Wildlife Sanctuary (WLS) and Bardia NP-Shuklaphanta NP-Dudhwa NP-Katerniagath WLS-Pilibhit TR were top priority (Table S3). While, Bardia NP and Shuklaphanta NP in Nepal and Dudhwa NP, Manas NP in India would benefit from Rhinoceros supplementation (Table 4 & Table S3).

Distribution probability of greater one-horned Rhinoceros across its global historic range in the Terai-Brahmaputra floodplain, where 1-Rajaji NP, 2-Hastinapur WLS 3-Corbett NP, 4- Shukalaphanta NP, 5-Pilibhit TR, 6-Bardia, 7-Keterniyaghat WLS, 8- Sohelwa WLS, 9- Chitwan NP-Valmik TR-Parsa WLS Complex, 10-Koshi Tappu RAMSAR Site, 11- Gorumara WLS, 12- Jaldapara WLS, 13- Manas-Royal NP-Manas NP Complex, 14-Sonai Rupai WLS, 15- Orang TR, 16- Kaziranga NP, 17-Dibru Saikhowa WLS, 18- D’Ering Memorial WLS, 19- Pobitora WLS. Created in ESRI ArcMap 10.5.1 (

Distribution probability of wild swamp buffalo across the Terai-Brahmaputra floodplain, where 1-Jhilmil Jheel, 2-Corbett NP, 3- Shukalaphanta NP, 4- Kishanpur WLS, 5-Keterniyaghat WLS, 6- Chitwan NP-Valmik TR-Parsa WLS Complex, 7-Koshi Tappu RAMSAR Site, 8- Gorumara WLS, 9- Jaldapara WLS, 10- Manas-Royal NP-Manas NP Complex, 11-Sonai Rupai WLS, 12- Laokhowa-Burachapori WLS, 13- Kaziranga NP, 14-Dibru Saikhowa WLS, 15- D’Ering Memorial WLS, 16- Pobitora WLS Created in ESRI ArcMap 10.5.1 (


Competing interests

The authors have declared there are no competing interests.

Authors’ contributions

CC performed statistical analyses of data, created all figures, drafted manuscript, presented findings at symposium. DWJ conceived the work, provided guidance and input on data analyses, drafted manuscript. ASF conceived the work, deployed tags, provided input to data analyses, and drafted the manuscript. PH, NG and HD contributed to logistical and field support and edited the manuscript. All authors read and approved the final manuscript.

Watch the video: Tutorial 1: ArcGIS Basic Tools for Beginners - Introduction