Use prescribed burning: Grassland & shrubland

How is the evidence assessed?
  • Effectiveness
    not assessed
  • Certainty
    not assessed
  • Harms
    not assessed

Study locations

Key messages

  • Fourteen studies evaluated the effects of using prescribed burning in grassland and shrubland on reptile populations. Seven studies were in the USA, four were in Australia and one was in each of South Africa, Argentina and France.


  • Community composition (1 study): One replicated, before-and-after study in Australia found that reptile species composition was different before and immediately after burning in three grass types and remained different after vegetation grew back in one of three grass types.
  • Richness/diversity (5 studies): Two of three studies (including one replicated, controlled, before-and-after study) in South Africa, Argentina and the USA found that areas with annual burning had similar reptile species richness and diversity compared to unburned areas or that richness was similar across areas with a range of burn frequencies. The other study found that burned areas had higher reptile species richness than unburned areas. One replicated, site comparison study in Australia found that areas burned 1–4 years earlier had lower reptile species richness than areas burned 11–15 years earlier. One replicated, site comparison study in the USA found that areas with different burn frequencies had similar reptile species richness and diversity.


  • Abundance (11 studies): Three of six studies (including three replicated, randomized, controlled studies) in the USA and Argentina found that burned areas had a similar abundance of lizards, snakes and lizards and combined reptiles and amphibians compared to unburned areas. Two studies found that burning had mixed effects on the abundance of different reptile species and western yellow-bellied racer snakes.The other study found that burned areas had more eastern massasauga rattlesnakes than unburned areas. One replicated, site comparison study in Australia found that areas burned 1–4 years earlier had a lower abundance of reptiles than areas burned 11–15 years earlier. One controlled before-and-after study in the USA found that a burned area had a similar number of four snake species compared to when the area was managed by mowing. One site comparison study in France found that one reptile species was less abundant in areas managed by burning than areas grazed by sheep, whereas the abundance of five other species was similar in all areas. One replicated, before-and-after study in Australia found that immediately after burning, the abundance of reptiles was lower than before burning, but was similar after vegetation grew back. One replicated, randomized, site comparison study in Australia found that small-scale patch burning was associated with increased abundance of sand goanna burrows.


  • Use (1 study): One replicated, randomized, controlled study in Australia found that some rocky outcrops that were burned were recolonized by pink-tailed worm-lizards.

About key messages

Key messages provide a descriptive index to studies we have found that test this intervention.

Studies are not directly comparable or of equal value. When making decisions based on this evidence, you should consider factors such as study size, study design, reported metrics and relevance of the study to your situation, rather than simply counting the number of studies that support a particular interpretation.

Supporting evidence from individual studies

  1. A replicated, site comparison study in 1987–1990 in spinifex grasslands in the Northern Territory, Australia (Masters 1996) found that nine months to five years after prescribed burning, overall reptile species richness and abundance were lower than plots that had been burned 11–15 years earlier. Regenerating plots burned 9 months –4 years earlier had lower overall reptile abundance and species richness (abundance: 1–24 individuals/plot, richness: 1–12 species/plot) compared to mature plots burned 11–15 years earlier (4–47, 2–15). The relative abundances of species on regenerating plots changed after the first year of sampling, with terrestrial geckos becoming less common and Ctenotus species more common, whereas relative abundances of reptiles changed little in mature plots (see original paper for details of species individual and relative abundances). In 1987–1990 reptiles were surveyed 12 times (approximately every three months) in plots that had been burned in 1986 (‘regenerating’) or in 1976 (‘mature’; 3 plots/burn history). Surveys were carried out using drift fences with pitfall traps for three nights at a time (18 traps/plot).

    Study and other actions tested
  2. A replicated, site comparison study 2003–2004 in six watersheds in tallgrass prairie in Kansas, USA (Wilgers & Horne 2006) found that carrying out burning more frequently did not result in differences in combined reptile and amphibian abundance, species richness or diversity. Species richness and overall abundance were similar between areas with an annual burn (5–10 species/plot; 91 individuals), a four-year burn (6 species/plot; 115 individuals) or that remained unburned for 10–20 years (5–6 species/plot; 89 individuals), though the abundance of individual species in each treatment was mixed (see paper for details). Species evenness and diversity was similar across areas with different burn regimes, although reptile communities differed, with areas with more similar burning sharing more species (all results reported as indexes). Six watersheds were selected: two with annual burns; two burned every four years; and two that were unburned for 10–20 years. Two transects were established/watershed, and each transect (75 m long) incorporated cover boards, drift fencing (Y-trap array each end of transect) and funnel traps. In spring (1 month) and autumn (1 month) 2003–2004, traps were checked daily until temperatures reached 32°C and captured species were identified and marked.

    Study and other actions tested
  3. A replicated, site comparison study in 2004 in an area of grassland in Gauteng, South Africa (Masterson et al. 2008) found that reptile species richness was similar in areas that were last burned one, two or three years ago, and was also not affected by burn frequency over the previous 30 years. Neither time since last burn or frequency of burning over the past 30 years affected reptile species richness (data reported as statistical model result). In March–April 2004, reptiles were surveyed in nine sites that had last been burned one year ago (4 areas), two years ago (3 areas) or three years ago (2 areas). Burn frequency in the preceding 30 years of the sites varied from burning every 1–5 years. A total of 10 groups of traps (4 drift fences, 8 funnel and 8 pitfall traps) were established across the nine sites (1–2 groups/block). Traps were checked twice/day and all reptiles were identified to species level.

    Study and other actions tested
  4. A replicated, randomized, controlled, paired study in 2004 in juniper and mesquite shrublands in central Texas, USA (Radke et al. 2008) found that after prescribed burning, lizard captures were similar to unburned plots. Lizard captures were statistically similar in burned (8 lizards/plot) compared to unburned plots (4 lizards/plot). Low intensity prescribed burns were carried out in 0.2 ha plots, paired with unburned plots, in four locations on a former livestock ranch in February 2004. Lizards were surveyed using arboreal pitfall (1 m long, 8 cm diameter PVC tubes), glueboards (stapled to trees) and drift fences (one/plot) and terrestrial glueboards (a 4 x 4 grid 5 m around each arboreal trap) for four trapping sessions in March–August 2004 (152 traps and 5,908 total trap nights). 

    Study and other actions tested
  5. A replicated, randomized, controlled study in 1997–2001 in shrub and grassland in southern Texas, USA (Ruthven et al. 2008) found neither dormant (winter) season nor growing (summer) season prescribed burns affected the abundance of lizards or snakes in subsequent years. In the 2–3 years after dormant-season and growing-season prescribed burns, abundances of lizards and snakes in burned plots (dormant-season: 0.4–1.1 lizards/trap array/day, 0.1–0.3 snakes/trap array/day; growing-season: 18.2–19.8 lizards/trap array/day, 2.4–4.0 snakes/trap array/day) were similar to unburned plots (dormant-season: 0.8–1.4 lizards/trap array/day, 0.1–0.2 snakes/trap array/day; growing season: 13.6–14.8 lizards/trap array/day, 1.0–2.2 snakes/trap array/day). Dormant-season (December–February 1997–1998, 1999–2000) and growing-season (August 1999) prescribed burns were carried out in 2 ha plots (dormant: 3 plots, growing: 5) in a 15,200 acre study area. Reptiles were monitored using drift fences with pitfall traps (‘arrays’, dormant: 3 arrays/plot, growing: 1 array/plot). Dormant-season plots were monitored for 7–21 days each in May–August 1998–2000. Growing-season plots were monitored for 14 days each in May–September 2000–2001. Equivalent numbers of unburned plots were monitored at the same time. Prior to 1997, dormant-season plots had not been burned for ≥40 years. Growing-season plots had been prescribed burned in January–March 1997.

    Study and other actions tested
  6. A site comparison study in 2006 of cattle pasture in Corrientes, Argentina (Cano & Leynaud 2010) found that overall reptile diversity, species richness and abundance were similar in areas with annual burning and unburned areas. Overall reptile species richness, abundance and diversity were similar in sites with annual prescribed burning (richness: 4; abundance: 44, Shannon diversity index: 1.0) compared to sites that had not been burned for three or 12 years (richness: 3–4; abundance: 22–23, Shannon diversity index: 0.8–1.0). Some lizard species (e.g. Kentropyx viridistriga and Teius oculatus) were more abundant in annually burned sites, whereas others (e.g. Mabuya dorsivittata) were more abundant in unburned sites (see original paper for details). One site each (≥ 400 ha) was burned annually (August–September), left unburned for three years or 12 years. Monitoring was undertaken using drift-fencing with pitfall traps in January–April 2006 (80 survey days).

    Study and other actions tested
  7. A controlled study in 2005–2010 in a mixed coastal wetland, grass and scrubland and woodland habitat in California, USA (Thompson et al. 2013) found that four years after prescribed burning, western yellow-bellied racer snake Coluber constrictor mormon abundance was lower in burned than unburned sites, but that abundance was similar in burned and unburned sites from five years after burning took place. Four years after prescribed burning, western yellow-bellied racer snake abundance was lower (2008: 17 snakes/trap array) compared to unburned sites (49). In the fifth and sixth years after burning, snake abundance was similar in burned and unburned sites (2009 burned: 16 snakes/trap array vs. unburned: 25; 2010 burned: 19 vs. unburned: 30). Prescribed burns were carried out in a 213 ha area in autumn 2005 (64 ha) and 2006 (67 ha). Reptiles were surveyed in burned and adjacent unburned areas using traps, observation and coverboards. Traps were set in March–August 2007–2010 (277–1,140 trap days/year). Caught snakes (692 total individuals) were individually marked using PIT tags. Too few individuals were caught in the 2006 burn site to be included in analysis.

    Study and other actions tested
  8. A controlled, before-and-after study in 2006 and 2010 in two abandoned agricultural fields in New York State, USA (Dovčiak et al. 2014) found that prescribed burning increased the numbers of eastern massasauga rattlesnakes Sistrurus catenatus catenatus observed compared to before the fire and in an unburned area. After prescribed burning, eastern massasauga rattlesnakes were observed 27 times compared to no observations prior to burning and no observations in an unburned site over the same time period. The authors reported that rattlesnake occurrence was related to open habitats with low cover of leafy, non-woody plants (forbs) created by fire (see original paper for details). The study took place in two abandoned agricultural fields (disused for 15–20 years), one of which was burned in April 2010. Snakes were monitored using coverboards (in a 5 x 5 grid) per field before burning in 2006 and after burning in June–August 2010.

    Study and other actions tested
  9. A replicated, controlled, before-and-after study in 2011–2012 in four riparian grasslands in Missouri, USA (Larson 2014) found that areas with prescribed burning had higher reptile species richness compared to unburned areas. All results were reported as statistical model outputs. Reptile species richness was slightly higher in burned plots compared to unburned plots. Six turtles were found dead as a result of fire (two ornate box turtles Terrapene ornata, a western painted turtle Chrysemys picta bellii and three unidentified species). Snake presence was associated with 70–100% grass cover habitat that occurred the year following burning. Lizards were associated with burned or burned and heavily grazed plots, and turtles were associated with taller grass heights linked with light grazing. Patches of four watersheds (10–54 ha) were treated with prescribed burning (April 2011 or 2012) or were unmanaged during the past five years. Reptile monitoring took place 2–3 times/month in March–May 2011–2012 using coverboards and visual encounter surveys.

    Study and other actions tested
  10. A controlled, before-and-after study in 2006 and 2010 in disused crop fields in New York State, USA (Steen et al. 2015) reported that following prescribed burning, the abundance of four snake species did not increase. Results were not statistically tested. Two months after a prescribed burn in a field, counts were similar for eastern milksnakes Lampropeltis triangulum triangulum (0.002 snakes/coverboard), northern brownsnakes Storeria dekayi dekayi (0.040), eastern gartersnakes Thamnophis sirtalis sirtalis (0.181) and northern watersnakes Nerodia sipedon sipedon (0.004) compared to four years earlier (in 2006 milksnake: 0.001 snakes/coverboard; brownsnake: 0.020; gartersnake: 0.230; watersnake 0). The authors reported that counts of eastern milksnakes, northern brownsnakes and eastern gartersnakes may have declined in a neighbouring field that wasn’t burned over the same time period (see original paper for details). Snakes were monitored in two abandoned agricultural fields (1 km apart) that had been planted with crops until 15–20 years prior to the study, after which they had been managed by mowing biannually. Prescribed burning took place in one field in April 2010 instead of mowing in that year. Snakes were surveyed using coverboards in June-August 2006 and 2010 (25 coverboards/field, 20 total surveys).

    Study and other actions tested
  11. A replicated, before-and-after study in 2008–2010 in savanna and open woodland in north Queensland, Australia (Abom & Schwarzkopf 2016) found that while overall reptile abundance reduced immediately after prescribed burns compared to pre-burn in three different grass types, there was no difference in reptile abundance compared to pre- or post-burn once sites had revegetated. Reptile abundance was lower immediately post-burn compared to pre-burn (average abundance post-burn: 1.2–3.6; pre-burn: 3.2–5.2 reptiles/grass type), but there was no difference in abundance levels after plots had revegetated compared to pre-burn or immediately post-burn (average abundance up to 15 months post burn: 2.1–4.1 reptiles/grass type). Reptile species composition differed between pre-burn and post-burn in all three grass types (data reported as statistical model results). Reptile species composition also differed between pre-burn and revegetated plots for one native grass type (kangaroo grass Themeda triandra), but not for the other two. Eight plots for each of three different grass types (kangaroo grass, black spear grass Heteropogon contortus, and non-native grader grass Themeda quadrivalvis) were monitored in 2008–2010 (24 plots in total). Monitoring was undertaken pre-burn (>2 years before last burn), immediately post-burn and following revegetation (up to 15 months post-burn) using drift fences with pitfall and funnel traps.

    Study and other actions tested
  12. A replicated, randomized, controlled study in 2014–2015 in six rock and grassland areas in Australian Capital Territory, Australia (McDougall et al. 2016) found that following prescribed burns of rocky outcrops, Australian pink-tailed worm-lizards Aprasia parapulchella recolonised some rock outcrops within one year. No statistical analyses were carried out. Two worm-lizards were observed on plots that were burned compared to zero on unburned plots. A further four worm-lizards were observed in nearby high-quality habitat (4 worm-lizards and 3 shed skins observed). In April–May 2014, plots (4 x 4 m) in six replicate sites (150 m apart) were each randomly selected and burned (using a blow torch; one plot/site) or left unburned (one plot/site). A further plot at each site of high-quality habitat was also monitored. In February 2015, rocks were surveyed for lizards. All sightings of worm-lizards or shed skins were recorded.

    Study and other actions tested
  13. A site comparison study in 2016 in an area of heathland in Nouvelle-Aquitaine, France (Pernat et al. 2017) found that one of six reptile species was less abundant in sites that were burned 5–12 years previously than in a site grazed by sheep, whereas the other five species were similarly abundant across all sites. Fewer western green lizards Lacerta bilineata were found in any of the burned sites (0.1 lizards/site for all burned sites) than in the grazed site (1.5 lizards/site), whereas no difference was found between burned or grazed sites in the number of wall lizards Podarcis muralis (0–4 lizards/site) or the number of four snakes species (green whip snake Hierophis viridiflavus, viperine snake Natrix maura, grass snake Natrix natrix and European asp Vipera aspis; data not presented). An area of heathland (135 ha) was managed by prescribed burning or annual sheep grazing. Three burned sites (one each burned 5, 10 or 12 years ago) and one grazed site (all sites 8–10 ha) were selected. In 2016, a total of 96 cover boards (corrugated roofing tiles) were split between the four areas (24 boards/area), and 10 surveys were conducted in April–June. Reptiles found on or under cover boards were counted.

    Study and other actions tested
  14. A replicated, randomized, site comparison study in 2014–2016 in shrub and grass sandplain in Western Australia, Australia (Bird et al. 2018) found that small-scale patch burning was associated with increased sand goanna Varanus gouldii burrow abundance. Following several decades of annual small-scale burns, more sand goanna burrows were found in areas with a diverse burn history (results reported as statistical model outputs). Sand goanna burrows were particularly associated with no or early spinifex regrowth and mature, ready-to-be-burned spinifex habitats. The authors noted that sand goanna burrows found in plots with no or initial regrowth were likely to have been selected by over-wintering goannas prior to burning when the habitat was mature spinifex. Martu Aboriginal communities returned to the study area in 1984 and reinstated traditional winter patch burning since then. In July 2014–July 2016, seventy-six randomly-selected 1 ha plots (>1 km apart) in spinifex-dominated Troidia spp. desert were surveyed for sand goanna burrows. Plots were classified as: no regrowth present, early shoots present, mature plants with high plant diversity, mature spinifex able to carry a fire and deteriorating spinifex.

    Study and other actions tested
Please cite as:

Sainsbury K.A., Morgan W.H., Watson M., Rotem G., Bouskila A., Smith R.K. & Sutherland W.J. (2021) Reptile Conservation: Global Evidence for the Effects of Interventions for reptiles. Conservation Evidence Series Synopsis. University of Cambridge, Cambridge, UK.

Where has this evidence come from?

List of journals searched by synopsis

All the journals searched for all synopses

Reptile Conservation

This Action forms part of the Action Synopsis:

Reptile Conservation
Reptile Conservation

Reptile Conservation - Published 2021

Reptile synopsis

What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 21

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.

Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape ProgrammeRed List Champion - Arc Kent Wildlife Trust The Rufford Foundation Save the Frogs - Ghana Mauritian Wildlife Supporting Conservation Leaders
Sustainability Dashboard National Biodiversity Network Frog Life The international journey of Conservation - Oryx Cool Farm Alliance UNEP AWFA Bat Conservation InternationalPeople trust for endangered species Vincet Wildlife Trust