Background The Structure from Motion (SfM) technology based on the Unmanned Aerial Vehicle (UAV) platform is an important optical remote sensing technology to measure high-precision digital terrain. The SfM technology is often influenced by clouds and vegetation cover, whereas the UAV platform could be impacted by winds. This work aims to explore the optimal time window to measure terrain by the UAV-SfM survey in the Black Soil Region of Northeast China by considering dynamic changes of local climate and vegetation.

Methods The study area covers three climatic zones: The cold temperate zone, the middle temperate zone and the warm temperate zone. For different climatic zones in the Black Soil Region of Northeast China, the climate data of a typical site from 1988 to 2017 and the dynamics of vegetation coverage and snow cover in 2020 were analyzed. Typical DJI UAV operating parameters were used as the initial screening conditions for annual average climate data (temperature, wind speed and precipitation), and the medium vegetation coverage condition (FVC < 0.6) was then used as the screening basis for vegetation coverage to determine the appropriate period for the UAV terrain survey.

Results 1) The temperature and wind speed of the three meteorological stations were analyzed comprehensively. It was found that the temperature curve changed as a quadratic function, and the average wind speed curve and the maximum continuous wind speed curve had the same trend. Temperature and wind speed were satisfied in the cold temperate zone from mid-April to early October, in the middle temperate zone from early April to mid-October, and in the warm temperate zone from mid-March to mid-November. 2) The probability of daily precipitation curves in the three meteorological stations fluctuated greatly and were not significantly representative, thus it could not be used as the standard for estimating investigating optimal time window. 3) By analyzing the land cover conditions of the three sites, it was found that the ground was covered with snow from October to March of the following year, and the vegetation coverage exceeded 0.6 from May to October, which made it impossible to carry out UAV topographic survey. The cover conditions that met terrain acquisition range from March to the end of April and from the end of September to November. Comprehensive analysis of meteorological and cover conditions, we found that the most suitable periods for the cold, middle and warm temperate zones were from mid-April to early May, from mid-April to mid-May, and from mid-March to the end of April or from the end of September to the beginning of November.

Conclusions By applying the results of this paper, the UAV-SfM survey may effectively avoid adverse periods and ensure obtaining the best results. The findings of this work could guide terrain measuring works in the field of soil and water conservation.