Gradsect Explained

A gradsect or gradient-directed transect is a low-input, high-return sampling method where the aim is to maximise information about the distribution of biota in any area of study. Most living things are rarely distributed at random, their placement being largely determined by a hierarchy of environmental factors. For this reason, standard statistical designs based on purely random sampling or systematic (e.g. grid-based) systems tend to be less efficient in recovering information about the distribution of taxa than sample designs that are purposively directed instead along deterministic environmental gradients.

Ecologists have long been aware of the significance of environmental gradient based approaches to better understand community dynamics and this is reflected especially in the work of Robert Whittaker (1967)[1] and others. Although in practice, life-scientists intuitively sample gradients, until the early 1980s there was little formal theoretical or empirical support for such an approach, sample design being driven largely by traditional statistical methods based on probability theory incorporating random sampling.

Origins

Intensively sampled landscape-based surveys in Australia provided a reference platform for developing and testing a less logistically demanding and yet statistically acceptable gradient-based survey design that avoided the need for random or purely grid-based sampling. These initial studies [2] and subsequently developed statistical support for purposive, gradient-based survey[3] provided a formalized, practical alternative to more logistically demanding traditional designs. It was here the term gradsect was coined that coupled purposive, transect sampling with a hierarchical framework of environmental gradients considered to be key determinants of species distribution.

Methodology

In constructing a gradsect, existing information is initially reviewed in which a hierarchy of environmental gradients is first identified either by visual means (maps, aerial photographs etc..) or through numerical analysis or spatial analysis of institutional or other data sources. A typical regional gradsect for example may be constructed according to a primary climate gradient (temperature, moisture, seasonality) then a secondary gradient (geomorphology, lithology, major and minor drainage systems), a tertiary gradient possibly represented by a local soil catena or local land use farming system or finer scale gradient levels representing local vegetational sequences. Through an inspection of spatial overlays of all gradients, a minimum number of sample locations is then purposively located to reflect, as far as possible, total environmental variation. For logistic and other purposes (such as improving the capacity to locate rare species) the steepest gradients are usually selected. In this way an ideal gradsect is constructed that may then be modified to accommodate logistic tradeoffs. The selection discipline requires that the fullest possible range of each hierarchical level is sampled. This commonly results in a set of progressively nested clusters of sample sites contained within the overarching primary gradient that may not reflect a linear distribution. At relatively local landscape scale, a primary gradients may be represented by salinity levels or water depth as in tidal wetlands or micro-topographic relief as in forest margins or a riparian zone. For most practical purposes, transects are commonly laid out along contours perpendicular to the main direction of the gradient. Iterative spatial analysis of environmental layers over a digital elevation model can then be used to identify areas requiring additional sampling thereby improving environmental representativeness.[4]

Advantages and limitations

Initial studies in gradsect development revealed considerable logistic and other advantages over more traditional non-gradient-based survey designs concerned primarily with random sampling. This finding is now widely supported especially in biodiversity and other areas of environmental surveying and conservation design (see Applications next). Apart from improved logistic efficiency, the gradsect method seeks to maximise environmental representativeness which has the dual advantage of potentially improving location of rarities and enhancing spatial modelling of species distribution. Because the underlying statistical model is not based on probability theory, gradsect sampling cannot be used to estimate numbers of species or other biological attributes per unit area. For that purpose some measure of random sampling needs to be built into the sample design.

Applications

Since the publication of gradsect theory in 1984, subsequent vegetational and landscape studies in regional Australia (Austin and Heyligers 1989);[5] Ludwig and Tongway (1995[6]) were followed by a successful evaluation of the method in faunal surveys in South Africa (Wessels et al.[7]). Since then applications involving gradsects have ranged from habitat suitability studies of fungi (Shearer and Crane 2011[8]), termites (Gillison et al. 2003[9]) other macro invertebrates (Lawes et al. 2005[10]); birds (Damalas 2005[11]) small and large mammals (Laurance 1994;[12] Ramono et al. 2009[13]). Vegetation studies using gradsects have been widely applied in many countries ranging from tidal wetlands (Parker et al. 2011[14]) and agricultural cropping systems and forested landscape mosaics (Gillison et al. 2004[15]) to infectious diseases (Boone et al. 2000[16]). At broader geographic and national scales (Grossman et al., 1998,[17] 2007;[18] USA/NPS 2012[19]) gradsects have been applied to guide field sampling and forest mapping in mountainous terrain (Sandman and Lertzmann 2003[20]) as well as wide-ranging remote sensing applications (Mallinis et al. 2008;[21] Rocchini et al. 2011[22]).

Notes and References

  1. Whittaker R. H. (1967). Gradient analysis of vegetation. Biological Reviews 42: 207–264.
  2. Gillison, A.N. (1984). Gradient oriented sampling for resource surveys – the gradsect method. In: K.R. Myers, C.R. Margules and I. Musto (eds.) Survey Methods for Nature Conservation pp. 349–74. Proc. Workshop held at Adelaide Univ. 31 Aug. to 31 Sept. 1983. (CSIRO (Aust.) Division of Water and Land Resources, Canberra)
  3. Gillison, A.N. and Brewer, K.R.W. (1985) The use of gradient directed transects or gradsects in natural resource surveys. Journal of Environmental Management 20; 103–127.
  4. Gillison, A.N. (2013). Plant Functional Types and Traits at the Community, Ecosystem and World Level, in Vegetation Ecology, Second Edition (eds E. van der Maarel and J. Franklin), John Wiley & Sons, Ltd, Oxford, UK. Ch 12, pp.347-386.
  5. Austin, M.P. and Heyligers, P.C. (1989). Vegetation survey design for conservation: gradsect sampling of forests in northeasternNew South Wales; Biological Conservation 50: 13–32.
  6. Ludwig, J.A. and Tongway, D.J. (1995). Spatial organization of landscapes and its function in semi-arid woodlands, Australia. Landscape Ecology. 10: 51–63.
  7. Wessels, K.J., Van Jaarsveld, A.S., Grimbeek, J.D. & Van der Linde, M.J. (1998). An evaluation of the gradsect biological survey method. Biodiversity and Conservation. 7: 1093–1121.
  8. Shearer, B.L. and Crane, C.E. (2011). Habitat suitability of soils from a topographic gradient across the Fitzgerald River National Park for invasion by Phytophthora cinnamomi. Australasian Plant Pathology 40: 168–179.
  9. Gillison, A.N., Jones, D.T., Susilo, F-X. and Bignell, D.E. (2003). Vegetation indicates diversity of soil macroinvertebrates: a case study with termites along a land-use intensification gradient in lowland Sumatra. Organisms Diversity & Evolution. 3: 111–126.
  10. Lawes, M.J., Kotze, D.J., Bourquin, S.L. and Morris, C. (2005). Epigaeic invertebrates as potential ecological indicators of afromontane forest condition in South Africa. Biotropica 37; 109–118.
  11. Damalas, A. (2005). Landscape ecology of birds on Mount Leconte, Great Smoky Mountains National Park Dissertation. Old Dominion University. 358 pages; AAT 3195595.
  12. Laurance, W.F. (1994). Rainforest Fragmentation and the Structure of Small Mammal Communities in Tropical Queensland. Biological Conservation. 69: 23–32.
  13. Ramono, W., Isnan, M.W., Sadjudin, H.R., Gunawan, H., Dahlan, E.N., Sectionov, Pairah, Hariyadi, A.R., Syamsudin, M., Talukdar, B.K. & Gillison, A.N. (2009). Report on a second habitat assessment for the Javan rhinoceros (Rhinoceros sondaicus sondaicus) within the island of Java. International Rhino Foundation.
  14. Parker, V.T., L. M. Schile, M.C. Vasey, and J.C. Callaway. (2011). Efficiency in assessment and monitoring methods: scaling down gradient-directed transects. Ecosphere. 2: 99
  15. Gillison, A.N., Liswanti, N. Budidarsono, S., van Noordwijk, M. and Tomich, T.P. (2004). Impact of cropping methods on biodiversity in coffee agroecosystems in Sumatra, Indonesia. Ecology and Society 9: 7. [online] URL: http://www.ecologyandsociety.org/vol9/iss2/art7
  16. Boone, J.D., McGuire, K.C., Otteson, E.W., DeBaca, R.S., Kuhn, E.A., Villard, P.F. & St Jeor, S.C. (2000). Remote Sensing and Geographic Information Systems: Charting Sin Nombre Virus Infections in Deer Mice. Emerging Infectious Diseases 6: 248–258.
  17. Grossman, D. H., Faber-Langendoen, D., Weakley, A.S. et al. (1998). International classification of ecological communities: terrestrial vegetation of the United States. Vol. I, The National Vegetation Classification System: development, status, and applications. The Nature Conservancy, Arlington, Virginia, USA.
  18. Grossman, D., Drake, J., Schindel, M., Hickson, D. et al. (2007). Classification of the Vegetation of Yosemite National Park and Surrounding Environs in Tuolumne, Mariposa, Madera and Mono Counties, California. NatureServe In Cooperation with the California Native Plant Society and California Natural Heritage Program Wildlife and Habitat Data Analysis Branch California Department of Fish and Game.
  19. National Park Service USA (2012). Gradsect and Field Sampling Plan for Big Bend National Park/Rio Grande National Wild and Scenic River.(Book)
  20. Sandmann, H. and Lertzman, K.P. (2003). Combining high-resolution aerial photography with gradient-directed transects to guide field sampling and forest mapping in mountainous terrain. Forest Science 49: 429–443.
  21. Mallinis, G., Koutsias, N., Tsakiri-Strati, M. and Karteris, M. (2008). Object-based classification using Quickbird imagery for delineating forest vegetation polygons in a Mediterranean test site. ISPRS Journal of Photogrammetry and Remote Sensing. 63: 237–250.
  22. Rocchini, D., McGlinn, D., Ricotta, C., Neteler, M. and Wohlgemuth, T. (2011). Landscape complexity and spatial scale influence the relationship between remotely sensed spectral diversity and survey-based plant species richness Journal of Vegetation Science. 22: 688–698.